MU: Super Mini Modular synth project construction journal.

Moog, Synthesizers.com, MOTM, Modcan, Moon and others..... Go big!

Moderators: lisa, luketeaford, Kent, Joe.

Post Reply
User avatar
hsosdrum
Veteran Wiggler
Posts: 737
Joined: Thu May 25, 2017 11:02 pm
Location: Southern California, USA

Post by hsosdrum » Mon Apr 30, 2018 5:03 pm

Rex: I can't wait to hear a demo comparing the sound with the Drive all the way up and the Drive all the way down. If it's as great as you say (and I have no reason to think that it isn't), I'll have to dig my way back through this thread to find out what you did with that Drive circuit and possibly make one myself!
:nana:

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Mon Apr 30, 2018 9:01 pm

hsosdrum wrote:Rex: I can't wait to hear a demo comparing the sound with the Drive all the way up and the Drive all the way down. If it's as great as you say (and I have no reason to think that it isn't), I'll have to dig my way back through this thread to find out what you did with that Drive circuit and possibly make one myself!
:nana:
All too easy ... (if you have a Q113) connect one of the bus outputs back to one of the unused inputs, then use that input's level knob to adjust the feedback depth. So, connect one of the jacks that is labelled "Output" to one of the input jacks on the same side of the panel as the Output jack with a small patch cable. Done. Then use the input channel's knob to control the feedback depth (aka "drive level").

Here's a pic I posted in an old thread I posted regarding the technique.

Image

Image

I liked it so much that I ended up hardwiring the setup as a switched option on the Q113 ... flip either switch and the feedback loop for that particular bus was instantly engaged, just on that bus ...

Image

If you don't have a Q113, here's how to do the wiring in a Q112. Solder a single wire from the "Output" jack's tip to the Ch3 input jack's tip. Then use Ch3's level knob to control the amount of drive (feedback). If you wish you can place a toggle switch between the two jacks so you can engage the feedback loop just by flipping the toggle up. But I found the switch to be redundant, because reducing the feedback level to ~zero~ (full counter clockwise) is just as useful.

Image

Image

If you need any more help, just PM me and I'll walk you (or anyone else) through it.

:tu:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
hsosdrum
Veteran Wiggler
Posts: 737
Joined: Thu May 25, 2017 11:02 pm
Location: Southern California, USA

Post by hsosdrum » Mon Apr 30, 2018 11:43 pm

Thanks, Rex! Now that you've explained it I realize that for a while I had my Q150 similarly hot-rodded, with its output feeding a Q114 channel and that channel's output piped back into the Q150's second input. The next Q114 channel served as the Q150's final output stage. (The Q114 was acting as a distributor, not as a mixer.) With this setup the Q150's second input volume knob served as the "Drive" control. (I had both Q114 channel volume controls set at maximum.) Of course, now that my Q150 is a 1U J150 I can no longer run it like that.

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Wed May 02, 2018 12:47 am

hsosdrum wrote:Thanks, Rex! Now that you've explained it I realize that for a while I had my Q150 similarly hot-rodded, with its output feeding a Q114 channel and that channel's output piped back into the Q150's second input. The next Q114 channel served as the Q150's final output stage. (The Q114 was acting as a distributor, not as a mixer.) With this setup the Q150's second input volume knob served as the "Drive" control. (I had both Q114 channel volume controls set at maximum.) Of course, now that my Q150 is a 1U J150 I can no longer run it like that.
Hmmm ... seems the 1sp retrofit panel trades a couple of features and functions for reducing the Q150's size from 2sp down to 1sp. Oh well, 'spose ya can't have it all! Clever use of the Q114++ on your part as well ... nice!

QX113 MAIN VCO MIXER COMPLETED AND OPERATIONAL:

I got the main mixer done. Turns out I was able to use the 0.250" x 28 Thread Per Inch bolts for more than just "hanging a sign".

I'm also making good use of more .... MilSpec AIRCRAFT WIRE!!!!!!! :trampoline: :loves:

Every place you see a splice with clear heat shrink I used 22 gauge black .. (take a deep breath before you say it) .. 19 strand Teflon Insulated Fiberglass Reinforced Pre-Tinned Military Specification Aircraft Wire (whew!) .. to extend any of the jack and/or pot wires. I was also able to use the same MilSpec wire in 16 gauge that form a bridge from the two outer most jacks' sleeve terminals to a couple of the 0.250" allen bolt. Also put to task was that gas tight goop (zinc crystal impregnated paste) between the eyelets and the panel, as well as between the PCB-end of two pairs of shield/sleeve wires eyelets and the 16 gauge wire eyelets. So the termination stack is as follows - from the bottom up:

*Panel
*paste
*0.250" eyelet with dual runs of 16 gauge stranded wire
*paste
*0.250" eyelet with dual runs (and doubled over) 22 gauge stranded wire
*0.250" internal tooth stainless steel lock washer
*0.250" x 28 TPI stainless steel nut, torqued to manufacturer's spec (aka ~really friggin tight!~).

Wipe off any excess paste using a Q-Tip .... call it good.

Image

When using those rather large eyelets with black 22 gauge wire, there are two wires inside of the eyelet crimp tube, and the bare sections of wires were also bent into a tight "U" to create "4 wires" inside of the crimping tube. This provides for a super secure crimp. On the 22 gauge wires I also crimped first, did the "pull as hard as I can" test after crimping (zero slippage), and then topped it off by soldering the crimp tubes. Just a little, enough to flood the crimped wires and the ~dent~ created by the mega-duty crimp tool.

I also use insulated crimp connectors, but since these are connected to the panel anyhow, I cut the insulators off. When the connectors are crimped on, those plastic insulators become very deformed and look like ass. So I removed the insulators, crimped and soldered the eyelets (just on the 22ga extensions. However on the 16ga grey wires I did not solder them because that would have ~cooked~ the gas tight paste I used on the bare ends of the wire prior to crimping the connector's crush tube.

The tool I use is a Klein F2. It's a super heavy duty crimping tool that I've had since 1994. I used it when we owned our factory authorized warranty repair center for 15 years. It's crimped many (MANY) a crimp connector in the 24 years I've owned it. They're very inexpensive when you consider their value (roughly $20 bucks). And due to their design, they are able to put a crapload of crimping pressure between the jaws (note how short the distance is between the fulcrum and the crimping jaws). Also due to their design, I was able to make sixty crimp connections in one evening lately while making up connection wires for this synth project, and while my hands/knuckle joints got a bit sore, they were FAR LESS sore than had I used a set of the wire stripper/crimpers many of us have on hand for stripping wires. If you're going to be doing a lot of crimping, or using larger crimp connectors , do yourself a favor and invest in a set of Klein F2 crimpers. To provide context, these things are 9.5" from end to end, and heavy enough to double as a murder weapon!!

Image

Here's a perspective side shot of the QX113, use of the 0.250" bolts for more than just attaching the name plate to the main panel is evident here. The front mounted 1.250" standoffs provide ~legs~ for the module to stand on while doing wiring on the back of the panel. Works a treat! The grainy texture of the truck bed liner spray I used is showing it's self in this shot as well.

Image

This next side shot below depicts how the grey 16 gauge wire made a nice (and VERY solid) ~bridge~ over other components in it's route from the 16 gauge bare bus wire to the bolt.

And FYI, the colored spiral loom is 1/4" O.D. I bought it on eBay ... four different colors, 25 feet per color.

I dunna know man, ya think 25 feet x 4 will do it? :eek: :roll: :lol:

Meh, it was priced right ... :despair:

So here's the wire bridge ......

Image

To make things route more smoothly, in this below picture you can see that I went from the right side of the PCB to the left bolt, and the left side of the PCB to the right bolt. It made for a smoother run (not kinked or bent sharply). Those are shield wires that normally connect to the sleeve terminals of the Tip/Sleeve jacks. Rather than connecting those leads to the jacks, I connected them directly to the panel with the use of "Ox-Gard" gas tight paste (as described above).

Image

By the way, that paste may be found all over eBay. It's called "Ox-Gard" and there may be a number after the name ... the number simply describes how much is in the tube (EG; "OxGard 4" means 4oz tube). I bought two 4oz. tubes, so far I've barely touched just one of the tubes. I emptied one full tube into a more useful clear plastic bottle. I also used it on the stripped portion of the 16 gauge wires before inserting the bare ends into the crimp connector tubes prior to crimping them. The zinc crystals ~cut~ into the metal penetrating any oxidation formed on aluminum or copper surfaces and/or wires, creating a more conductive connection. The gas tight substrate paste then prevents oxygen from contaminating the connection over time. I'll be using that stuff on each and every eyelet termination when I go to connect the module power cables to the bus bars. As of now, that will be EIGHTY eyelets just in the main cabinet.

Image

(below) Here you can see how the 16ga. pre-tinned bare bus wire was soldered to the sleeve terminals on each of the Tip/Sleeve jacks. The 3 center-most jacks also have their sleeve wires sent to the PCB, per the factory Dot Com configuration. So there is some serious "grounding" going on! This module may end up serving as a "star point" for ground cable distribution, so I wanted to design things to handle just about whatever I decide to do, or how this module may be used later on (I mean, who knows what the future will bring? ... after all, this is modular we're dealing with here!). "Future proofing" is what all of the kewl kidz call it these days, I believe.

(Geezus I'm an old ass hippie) ... :lol: :lol:

Image

I may
(stress ~may~) end up using one of the normalizing rails as a kinda "grounding bus" of sorts. Connecting large-ish wires from ground points on the three upper row panels to the norm rail, then running a loom of large wires (aircraft wire, of course) from the norm rail to a "Technical Earth" point that would be as close to the main power input ground connector (the wire that runs from the odd-pin of the main power cord to the wall socket) as possible. All in an effort to give the "safety grounding" system of wiring a nice and heavy conductor to provide a means of up to 25 amps to travel on in the event of a failure. I'd much rather do all of that work to provide current with a place to go, instead of it traveling through ME when I touch a panel or patch cable (the metal part of the jack). Unlikely? Yup. Impossible? Nope. Better safe than sorry, they say.

So, all three upper row panels/modules (two Dual VCO voice panels, and the QX113 Main VCO Mixer) could have grounding wires that connect to the lower norming rail in the upper module row. Then connecting the norm rail to the Technical Earth point near the mains power inlet where the ground wire comes in to the synth cab.

Image

Lastly, here's a depiction of the I/O jacks. The Main Out is not used in any way as part of the signal normalizing. It is left unused, but available should I want/need a summed output of all 4 channels. The descriptors on the picture tell the rest of the story.

All that is left is soldering wires from the input points on the On/Off toggles that will connect to the Terminal Strips on the normalizing rails, then back to the VCO's outputs. So ... VCO out to terminal strip, then back to the QX113 inputs. This permits easy installation and removal of any of the 3 panels.

Image

Now I can hook up the one VCO that's been completed to this mixer using a temporary connection lead so I may test out all of this stuff. That's what is next on the list!

Yet one more thing towards the finish line .... :cflag:

Thanx ... C-Ya Next Time!! :headbang:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Wed May 02, 2018 6:23 pm

TESTED THE QX113 MIXER WITH THE QX816 DUAL VCO (ONLY ONE VCO THOUGH)....

AND IT ALLLLL WORKS ... ALL OF IT!:

Hot DAYUM! I'm hopelessly excited! I installed the Mixer with the VCO and a few other modules into a pair of QCR8 Rack Frames. I do not have any LFOs running just yet, so I used the Beatstep once again. I created a short four note sequence that spanned two full octaves, then ran the signal out to a Q105 Slew Limiter ..... instant LFO!

I connected the VCO's own Q141/Q161 selection switch out to channel 2 of the QX113 Main VCO Mixer. The mixer's out went in to a Q108 VCA which I manually manipulated as a simple volume control. Then out to the PA Mixer/speakers. I used the "LFO" output from the Beatstep, moving from CV input to CV input, testing all of the various CV functions (Linear FM, Adjustable Expo FM, and PWM).

IT ALLLLLL WERX!!! And nicely, at that!

Even the little yellow LED I put in the Slew Limiter worked.

I gotta say, that feedback loop mod that I've done to every single mixer works just GREAT. Adding some sine wave to PWM along with a bit of overdrive on just the Sine wave creates some sick bottomy PWM. Adding a bit more overdrive to the mix of those two (in the QX113 Mixer) even does more to make things sound MASSIVE.

So yea .... super thrilled to learn that something that was only in my head, with no previous sonic trials or tests, works. Very well, at that.

You'll see no cables between the VCO and the Mixer since those connections are via the normalizing connections in back ... so even THAT worked out. The orange patch cable plugged in to the mixer is plugged in to BUS A OUT ... since VCO2 is normalized to Ch2 input (which is located in BUS A). See the previous posting for more infos on the way the QX113 is set up.

One thing that is a pleasant surprise is that none of the toggle switched functions "pop" when activated. Even the large channel On/Off switches on the Main VCO Mixer don't ~pop~ through the speakers when using them. Not even a little. Very nice to learn that!

Image

Image

Image

Now that I am fully confident in all of the wiring ideas and the labor I put in to making diagrams and planning out circuit routings, I can plow forward with completing the other three VCOs. I also learned a few better methods to do certain things which will speed up the process. It's all going quite well.

One little oversight .... when I get around to building the QX420 Dual Filter I'll be short about 4 or 5 three conductor pigtails ... the ones that connect pots to the Dot Com PCBs. Keep in mind that I bought a number of mixer modules to harvest the PCBs/pots/jacks/pigtails from them. Those mixers only used anywhere from two to six pots. Every single mixer I have put to use in these customized multi-circuit module panels uses SIX pots with SIX three conductor pigtails, and there are SEVEN of those hot rodded mixers in this system (42 pots, 42 three conductor pigtails). I'd be short on 24mm B100K Linear pots as well but a number of those pots were replaced with 16mm B100K Linear pots, so I have pots covered. That said, I'll run four or five short when I get around to putting the filter panel together, plus another SIX when I wire up the Ring Modulators. Said and done, I'll be short about eleven of those pigtails.

The fix will be to order eight of the Dot Com QP-S16 Toggle Switch Assemblies. They come with TWO of those three conductor pigtails on each assembly, as well as one DPDT mini toggle switch. Each of those assemblies are $6.00 bucks each. So ordering eight of them ($48.00) will yield 16 of the three conductor pigtails that I can harvest from those assemblies, along with eight toggle switches (which I can always use for one thing or another). I'm over-ordering these things just to be safe. I look at it this way ... I'm buying 16 pigtails for $3.00 each, and getting 8 DPDT toggles thrown in for free. :tu:

Ok, all done. Just wanted to share my excitement! :hyper: :headbang: :trampoline: :spin: :stardance: :haha: :wookie: :domodance:

Thanks .... 8-)
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Mon May 07, 2018 12:14 am

QX816 "A" IS COMPLETELY UP AND OPERATIONAL - ALSO JUST FOR KICKS I WOKE THE HAMMOND:

Both VCOs are on line. Everything works exactly as designed. When first fired up, it took me a few minutes to ~learn~ my own creation. After about ten minutes, I was using it like I'd had it for years. Removing the giant chickenhead knobs that I originally used for VCO Fine Tune was a necessary change. Those big ass "bird nose knobs" (as my brother calls them) were far too easy to disturb when fiddling with other controls, thereby knocking the VCO out of tune. So, I put 1 inch diameter black *line* knobs in the bird nose knob's place ... turns out I really like all of the extra space surrounding those particular controls. Makes it very easy to tweak them and keeps them out of the way of other controls very well.

There's still one task left to do on this QX816, which is wiring up the CV Input Selector toggles. Each toggle gets two wires (one for CV source "A", and one for CV source "B" ... center position = "OFF" .. as in no CV in). But for playing around and function testing both VCOs, they are not needed. Besides, they connect to a CV input terminal strip in the Main Cabinet .... which isn't ready to accept this module yet anyhow. So for now, all of the CV Source toggles remain in their Center/Off position.

The only ones in that center column being used are HARD/OFF/SOFT sync controls, UP/BOTH sync lock selector (which does some really cool stuff when either hard or soft sync is used). Sync signals are normaled to the opposite VCO, using the Pulse Wave output as the normaled signal. So flip the sync switch to "HARD" and it connects it's "sync input" header on the PCB to the other VCO's Pulse Wave output jack. Flip that switch to "SOFT" and it routes the Pulse wave to the other VCO's soft sync input header. Easy Peasy! Both VCOs are "cross connected" for syncing in either direction ... or both directions.

:hmm:

Wish to use a different waveform for syncing? Simple, just patch a cable from the master VCO's waveform output jack you wish to use and plug the other end into the "SYNC" jack of the slave VCO. The "sync" jack is a switching jack that's wired to defeat the normaled pulse wave connection. Then simply route that plugged in waveform to either ~hard~ or ~soft~ sync headers via the Soft/Hard toggle. The "Up/Both" toggle allows you to have the slave's synchronization to either the upward edge of the master's waveform, or both the upward edge and the downward edge. All of these sync options at the flip of toggle switches. Nice! And the bloody things WORK.

:trampoline: :stardance: :spin:

All of the other toggles that are not in this center column of toggles are 100% on line and operating perfectly! I used Taiway toggles throughout, I really like how positive they feel with engaged, so they were selected for that reason.

I'll post the back side of it soon .... I went with the traditional Dot Com "hairball" wire routing method. It looks like ass, but it certainly keeps crosstalk reduced to total null. I tried to do a few runs in the twisted pair style, but to be honest it looked worse than the bird's nest technique. All I know is it works excellently.

Used the Beatstep again as both a 1v/oct source and as a fakey modulator by using the Q105 Slew Generator to mess up the stepped voltage. I used a mult to divvy up the 1v/oct feed (the ~mult~ used to be a Q146 Normalization module that I had removed all of the pigtails from) .. so now it's a nice little 2x 4jack mult.

Balls like a friggin RHINO when the feedback loops are used properly. The key word is ~moderation~ and using them in the best places depending on the patch/sound/waveforms. Note that there are 6 feedback loops (two per VCO and two on the Main VCO Mixer). Every red knob is a feedback loop control (aka Drive Level). Judicious use pays huge dividends.

Image

Image

In the above picture, note the use of Pulse Waves, along with Hard Sync of VCO2, and look at the feedback amounts. Also look at the Main VCO Mixer (QX113) and how the first 2 input levels are set as well as the feedback level for their bus (Bus "A"). This worked out great.

HAMMOND A102 released from the crypt:

I became inspired by a sequence used with the QX816/Beatstep, so I shut down the synth (easily done, just flip the VCO on/off toggles on the Main VCO Mixer off for the 2 channels I was using), and dug the 1962 Hammond A102 out fum-unduh it's covers. Patched it up to the attenuator, then into an MXR-108 10 band graphic EQ, then into a Boss RE20 Space Echo, then into the Marshall 2061 amp powering the 1965 Leslie 251D that I modified to accept power from the Marshall. Holy SHIT ... I'd forgotten how powerful that ensemble is! The Marshall 2061 "Lead/Bass" is the perfect match with a vintage Hammond and Leslie. It's not Emerson's 1967 C3 with 2 or 3 Hiwatt amps powering a few Leslies ... but gottum plenty big balls and can SCREAM with the best of them.

Image

For folks unfamiliar with the A102, it's one of the A100 series organs (the "2" designation is for the furniture style, which is French Provincial). The A102 has precisely the exact same innards as the mighty mighty B3/C3 of yore. The difference being the cabinet style, as well as the A100 series having internal speakers and internal self contained reverb amp and tank.

Of course I just couldn't leave well enough alone. I modified the hell out of it, removed the speakers, power amp, reverb amp, and reverb tank. So now it's just another C3 ... (with a 1/4" output modification of my own design) ... ~ahem~. I made up a variable output signal attenuator so it may be easily connected to instrument level, line level, or pro level outboard gear ... just set the output level with a pot in the attenuator.

The Marshall 2061 has also been modified to produce the correct frequency response for use with a Hammond.

Ok .... so next up is QX816 Dual VCO "B". I'll use lessons learn from "A" and hopefully it will only take about a week or two (probably two) to get it up and running. I've yet to order the filter panel from Front Panel Express just yet. But there's no hurry, I have puhLENTY to get done before I am ready to tackle that thing! As a refresher, it's a Q107 state variable filter, normalized to a Q162 Filter State Mixer for the Q107, and lastly a Q150 Transistor Ladder Filter .... with an outboard Slope Jumper Panel (for selecting any of 4 filter slopes for the Q150) normaled to the Q150. There's a switch to swap the filters' order in there as well as a number of other goodies. Flip that toggle, and the Q150 is first in series with the Q107/Q162 ensemble, flip it the other way and the Q107/Q162 ensemble is first. Plus MOAR!!!!!!

Later Skaters!!! :rock: :chug:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Boogie
Veteran Wiggler
Posts: 507
Joined: Thu Nov 10, 2016 12:56 pm
Location: Belgium

Post by Boogie » Mon May 07, 2018 10:13 am

Nice system Rex

:banana:
-------------------------------------------------------
“Digital generated ‘analog’, is missing its analog soul and therefore isn’t analog any more.”
Jupiter 8 clone: https://www.muffwiggler.com/forum/viewt ... p?t=209322
Minipops 7 clone: https://www.muffwiggler.com/forum/viewt ... p?t=199360
ClapTrap clone: https://www.muffwiggler.com/forum/viewt ... p?t=202248
www.nabla-instruments.com

Dr Gris
Wiggling with Experience
Posts: 369
Joined: Tue Feb 14, 2017 5:14 pm
Location: Sweden

Post by Dr Gris » Mon May 07, 2018 1:51 pm

Now we're getting somewhere :tu:
Too bad about the chicken head knobs, I liked the look those.
Well, it has to be practical to use...

Since Hammond is my main instrument, I salute you!!! :party:

//Magnus

RussiaZero23
Wiggling with Experience
Posts: 324
Joined: Fri Aug 23, 2013 12:53 am
Location: Clearwater ID 83552

Post by RussiaZero23 » Mon May 07, 2018 2:23 pm

Very awesome work Rex. I have been Following from afar quietly for some time now.

You gave me and helped me out with a few Dot com mods I want to do as well.

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Tue May 08, 2018 2:29 am

Boogie wrote:Nice system Rex

:banana:
Thanks man ... keep on with the boogie!!
Dr Gris wrote:Now we're getting somewhere :tu:
Too bad about the chicken head knobs, I liked the look those.
Well, it has to be practical to use...

Since Hammond is my main instrument, I salute you!!! :party:

//Magnus
I may go back to the large chickenheads ... I'm still working out what kind of knobs to use on this stuff. I too, like those big "bird nose knobs"! And here's to ya, brother Hammond player. There is nothing like a vintage Hammond console organ. Nothing! The sound, the keybed action (fantastic!), and the very cool electro-mechanical engineering. Not to mention how friggin solid a 350 pound instrument feels under your hands. Right on, Brother!! :tu:
RussiaZero23 wrote:Very awesome work Rex. I have been Following from afar quietly for some time now.

You gave me and helped me out with a few Dot com mods I want to do as well.
I absolutely LOVE reading that. My primary intention of posting all of this project progress is to inspire and help others that have ideas about what they wish to do, but don't have visuals and written explanations of how to go about getting to a working piece. The secondary intention is just to document this entire ride. I'm really glad to hear you're getting something useful from my project progress entries.

35 THOUSAND VIEWS!!! - POWER CABLES - A BETTER WAY TO DO CRIMP ON EYELETS:

Just a bit of celebrating 35,000 views of this project. Granted it's been zig-zagging along in fits and starts with a number of distractions, (as well as a roughly one year break) for over five years (I started this thread on Monday, February 25th, 2013) ... but still ... 35,000 views, man!! Cooly! THANK YOU ALL VERY MUCH FOR THE SUPPORT!!!!!!!!!
:hug: :loves: :headbang:

On to business...

POWER CABLES: I wanted to share the recent order I received just today from Hinton Instruments of 20 Euro power cables. Very high quality, made well, designed well, priced decently. In fact, these are the HIGHEST QUALITY Euro power cables that are commercially available. Far and away better than using ribbon cables (which use 32 gauge wires and crappy connection headers with very low connection cycle life). The high quality of the Hinton cables is due to a few factors, which I'll cover in this entry. Also, Graham makes cables for many formats including 5U/Dot Com, Wiard, FRAC, MOTM, Modcan, Serge, Bugbrand, and others.

Also, I special ordered them at 18 inches long with no distribution system ends installed, Graham kindly made them precisely as I wanted them. On the Hinton Instruments website, they are spec'd as 16/0.2 wire. Link = http://hinton-instruments.co.uk/paprod/psu/cables.htm (be sure to scroll down to the Euro cables section).

The online AWG to Metric wire conversion charts say that 16/0.2 is 16 strand 24 gauge wire. Said another way, this is 16 strands of wire bundled together to create a 24 gauge conductor. So each Euro power cable is ten conductors of 16 strand 24 gauge wiring. Hinton offers Euro cables with both 10-way headers and 16-way headers, I ordered twenty cables with 10-way header connections (which are ten wire cables).

EDIT: Hinton's web page says that 16/0.2 wire converts to 20 gauge AWG. The conversion charts I have located on the web says that 16/0.2 equates to 24 gauge AWG. Hinton's chart makes a bit more sense since the wires look larger than 24 gauge to me. So I must be reading those internet charts wrong.

Hinton's Page Link (scroll down once on the page) = http://hinton-instruments.co.uk/paprod/psu/psuinfo.htm

Screenshot of the chart on Hinton's webpage linked above ....

Image

And yet, here are two other charts, located on the web ... one says 16/0.2 = 22 gauge, and the other says 16/0.2 = 24 gauge. ......

Image

Image

So Hinton's chart says 16/0.2 is 20 gauge, another chart says 16/0.2 is 22 gauge, and yet another chart says 16/0.2 is 24 gauge. This inconsistency is consistently all over the web. EVERWHAT!! It's fekking fat ass wire, that Hinton cable is. Leave it there.

Moving forward .... here's the cables ...

Image

Hinton Instruments uses a very high quality header connection, stuffed with very high quality female socket pins. The individual socket pins are what really make the difference between what Graham makes and what I made. The ones I made use female crimp sockets that only have three contact points inside of the socket. That means that the male pins on the module's power connection header are only in contact with three itty bitty little contacts. Those contacts tend to dig in to the male pins, and wear them out prematurely. The pin sockets that Graham Hinton offers have much better pin contacts and offer more connection cycles before either the female sockets or the male pins wear out to a degree of unserviceability. Better said, the pin sockets I used wear out the connection sooner. And they don't make nearly as good a connection in the first place (higher connection resistance). The Hinton header connections spec at a contact resistance of 30mΩ max, per his website. That's 30 milli-ohms!

Besides, buying them instead of making them saved me from cutting, stripping, and crimping on the female pin sockets for 200 wires.

Here's the Hinton connections (right) vs my DIY connections (left) .... the Hinton Instruments stuff even looks like better gear.

Image

I'll be cutting them individually to length after I construct the control cabinet that is 100% Euro (it's 36 inches wide, with two 3U rows of Euro using Vector rail ... comes out to 180HP per row, 360HP total space available). Just like the two 5U cabs, there will be bus bars used in the Euro controller cabinet (three bars ... 12v pos - 12v neg - 0volt). I also need to mount a Bel Power HCC15-3-AG which is a 3 amp x 12v Linear PSU. It will be wedge shaped and fit between the 52" wide 5U cabs and the Dot Com QKB61 5 octave keybed (wonderful FATAR 9P keybed action!). I have preliminary dimensions all worked out, but I've yet to actually cut wood.

With all of that having been said, I have no real idea how long the power cables will need to be. So I special ordered 18 inch long cables without any terminals installed on the bus-bar end. To top it all off, from order payment to arriving in my mailbox (from the UK) the whole thing took only 8 DAYS!! Ordered/paid on Monday, received them the next Monday. Once Graham wrote to provide the tracking number, five days later they were in my mailbox. And I didn't have special high speed/low drag shipping used, either. Some killer service right there, especially since they had to be made when I ordered them. Excellent, simply excellent.

I also cannibalized a Dot Com QDH40 power cable ~squid~ to create 20 individual 5U power cables that I can cut to length and terminate with crimp on eyelets and attach to the bus bars in the 5U cabinets. Doing so yielded ten 24 inch long power cables, and ten 36 inch long power cables. All of them appear to use PVC insulated 24 gauge wire in their construction. As far as I can tell at this point, the longest cable will end up having one wire (the longest wire) that will be about 12 inches long. The rest of them will be shorter.

Image

CRIMP ON EYELET INSTALLATION METHOD: I've done something different here. I've got plenty of crimp eyelets on hand in red, blue, and yellow (the different colors on the insulators describe how much wire will fit into the actual crimp connector ~barrel~ with Yellow being the largest and Red being the smallest). I've also worked out a method of installing them that I feel is superior to just using the colored plastic insulator, which involves adhesive filled clear heat shrink tubing.

(re; image below) ... here's that picture of those forty 20ga normaling wires all set to go into the 5U project (they're part of the control voltage normalizing setup) This is for reference to the description of how/why I did what I did when installing the #8 eyelets on the wires. I'll be using the same technique when I finish up all of the power cables ... both Euro and 5U.

Image

I took the time to remove the stock insulators after crimping since that process deforms them so much, instead I used clear heat shrink with adhesive inside. It seems as though using adhesive filled heat shrink tubing creates a better strain relief system for the wire, reducing the chance of the wire fraying or breaking where the edge of the insulation and the crimp on connector meet. The wire's insulation takes over as the strain relief, the cable wants to bend a few mm away from the end of the insulation and edge of the connector. Intuitively, it just seems to be a better way of doing things. Time consuming, yes, and too inefficient for commercial production. But for DIY projects I personally feel it's worth the extra effort.

The stock colored plastic crimp connector ~insulator~ is usually crushed by the tool during crimping (as is the wire's insulation) and makes for a poor support, no matter how well the crimp is executed. Even if the insulator is also crimped (as a second crimp) around the wire's insulation, which is a commonly used method of installing insulated crimp on connectors.

The use of adhesive filled heat shrink tubing forces the wire to bend more gracefully, rather than the wire bending right at the junction where the insulation ends and the crimp connector meet up. Added all together it makes for a better strain relief system. At least it seems so to me, being the shadetree engineer that I am using redneck teck to build a synthesizer!

It may appear as though insulation was crimped along with wire, but it is not. Everything was done carefully, and each one passed the "pull on it as hard as you can test" after crimping.

Quite technical, that. :lol:

This technique is made even better with the use of Fiberglass Reinforced Teflon Insulated Mil Spec Aircraft Wire ("FRTIMSAW" for short). Use of wire with that type of reinforced Teflon insulation is a step upwards due to that insulation's ability to deal with heat, and the fiberglass reinforcement helps to support the wire even more when the adhesive filled heat shrink tubing is used. Again, the purpose is to use the insulation as a strain relief to help make the wire bend in a graceful curve a few millimeters away from the point where the end of the insulation meets the opening of the crimp connector (the actual metal connector). That way the wire won't tend to bend right at that junction, but rather where the adhesive is flowing out of the heat shrink tubing. It will hopefully prevent the wire from fraying or even breaking at that critical location.

Whether the "FRTIMSAW" is used, or the more popular PVC insulated wire is used, this method of installing eyelets seems to be an improvement over using the stock plastic insulator on the eyelets. Many times, while crimping the connector, not only is the crimp barrel crushed but also the insulator AND the wire's insulation are also crushed. That can't be a good thing.

Why do this? Because I'm there. And while I'm there, I may as well do what I can to improve the synthesizer as a whole construct. Takes more time? Yes, but less time than it would take to move the entire synth, remove the power cable, repair the broken termination, reinstall the cable, and put the synth back into place.

Thanks once more for all of the support, everyone. 35 thousand views. Just wow.

Off to bed. :tu:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Wed May 09, 2018 10:37 pm

QX816 TEST WITH LED/RING MODULATOR - ERNIE BALL VOLUME PEDAL AS EXPRESSION PEDAL TEST:

Ok, so while I have this operational QX816 Dual VCO module up and on it's wings, it a great time to test a few other things that are reliant on the QX816s. I still have that funky ring modulator test panel I made up a while back. Just to go over it once more, it is a "CGS Passive Ring Modulator" circuit built on a ready made PCB of the same name. The boards may be purchased from Modular Addict (like $7.00 bucks). After purchasing transformers and diodes, you'll have ~about~ $15.00 to $20.00 in each one. I am building FOUR of them into this synth.

** Go back to page 8 (link provided below) and scroll down to the post titled "OUTPUT STAGE PUT TOGETHER, and RING MOD PROTOTYPE BUILT:" and scroll down about halfway through that entry for details on these rings mods, and the modifications I made to them.

** Then, scroll further down Page 8 until you see the entry named "Center Strip Controls - Dual Ring Modulators - T-Nuts - Gorilla Glue:" for more information about the quad ring mod setup.

** Then scroll a little more down Page 8 until you get to the entry named "RING MODULATOR TEST PANEL TEST (WAIT --- WHUT??):" for more info on these RMs.

** Then go to the first post of PAGE 9 for further information regarding the use of LEDs as diodes in these rings modulators.

Link to Page 8 (just to get you there) = viewtopic.php?t=78836&postdays=0&postor ... &start=175

This most recent test is the first time I've patched up the LED ring mod to the actual QX816, however. This is an important test, since I've never tried these ring mods with mixed waveforms or waveforms that have been mixed with the feedback loops involved. So I sortof held my breath as I went headlong into this.

I patched the Modulator and Carrier straight off of the "Q161" outputs from each of the 2 VCOs. Then I routed the RM's output to a Q114++ mixer because I already knew the RM's output was going to need a bit of a boost to get the level anywhere even close to the level of 2 VCOs mixed together. To be able to test how it all sounded by adjusting INPUT levels to the Carrier and Modulator inputs of the RM/LED rig, I routed each VCO outs to individual input channels of the Q114++, then straight out of each Q114 channel out to the RM's inputs. I then sent the RM's output back to the Q114 on it's own input channel, hit the "100x" switch for that RM output signal, and patched that to the AUX A input of the QX113 Main VCO mixer so the output of the RM is mixed with the output of the QX816. First image captures the whole patch .... second image captures more of a close up view of the patch and provides a little better detail on how the Q114++ is patched up and set up. Refer back and forth to both images to get a better grasp.

Image

Image

That Q114 is one handy gizmo! I was able to use two of the channels as completely independent signal attenuators, and a third channel as a completely independent 100x booster for the output of the RM.

Here's a couple of diagrams .... the first one shows the VCOs all by themselves without any RM going on.

Image

This next diagram shows how the signals from the VCOs going to the ring modulators totally bypasses the QX113 Main VCO Mixer. These signals are tapped directly from the Q161 Waveform Mixer or the Q141 VCO Aid (both of which are built in to each VCO) ... there is a toggle switch on each Ring Mod that selects which source the rings modulator will use as it's input signals, and there is another switch on each ring mod that determines which source will be the carrier and the modulator. But this diagram is more about the basic signal flow of the waveforms from the VCOs into the Ring Mods, and where the Ring Mod outputs are sent.

Image

Now it is also more plainly seen where these signals end up at that 3-switch CP3 input panel. So now you can see how easy it is to switch each source on or off (Switch 1 = Voice "A", Switch 2 = Voice "B", and Switch 3 = Ring Mod Mixer output). Those switch's outputs are routed to the CP3 mixer's first 3 input channels, leaving channels 4 and 5 unused (which can then be used to add positive and negative offset, since that is the way the STG CP3 Mixer is designed).

RESULTS? Yes. There were results. :lol: :lol:

Ok, let me try that again .... yes, there were VERY GOOD results! The LED fitted passive ring modulator produced the SAME results that it did when I performed the first set of tests back a few months ago. It adds a super nice set of upper end harmonics to the regular plain old oscillator waveforms eminating from the QX816's outputs. I kept messing with the RM's output level via the Q114 level knobs (both input levels and it's output level) in an effort to see how the overall sound changed. It's just right!!! It's not really the type of ~ringmod~ "screaching" sound you'd expect, it's more of an increase in fidelity added to the normal oscillator sounds (all mixed together it almost has an acoustic quality).

If I hard sync'd the VCOs, removing the RM signal made the oscillators sound almost DULL!

Now the output level of the unboosted ring mod is pretty low. Low enough that 100x gain was just enough to provide what was needed to add it to the VCO output. This is why I've added a feedback loop to each ring mod mixer bus, and use the Q118 Instrument Interface to add gain via the 10x/100x/1000x toggle switch.

In short, it works! But it WOULDN'T work without boosting the output. Not even a little bit. Those diagrams should help you with understanding the signal flows of the VCOs, and the Ring Mods. I'm quite happy with the way it all worked out. This test was absolutely necessary, now I know I can move forward with wiring up the Ring Modulators in that 32 inch long center strip. Only thing is, I'm not totally sure using Germanium diodes will work out. The overall output level of those particular passive ring mods is pretty low, even though I was using LEDs as diodes. The LEDs produce a louder output level than the Germaniun diodes do. I suppose since the ring modulator I used is that ~test bed~ ring mod which has sockets for the diodes, I can fairly simply swap out the LEDs for Germaniums and give it a try.

OH! ... one other thing .... just cuz, I changed all 4 LEDs around (changing all four of their installed polarities to completely opposite of what the instructions called for) and it made absolutely NO DIFFERENCE in anything ... not output level, not sound or function, not tone. Nothing. It worked exactly the same no matter which way I oriented the LEDs. As long as they all in "quadruple asymmetric parallel" (???) it worked the same way either way.

Like this ....
-->
<--
<--
-->

Or like this ...
<--
-->
-->
<--

.... It worked exactly the same.

ERNIE BALL VP Jr. VOLUME PEDAL AS AN EXPRESSION PEDAL (USED WITH THE DOT COM Q181FJ):

I've discovered certain volume pedals are highly suitable for use as expression pedals. As long as the expression pedal input requires a Tip/Ring/Sleeve plug you're good to go. This is important, since locating a really high quality expression pedal isn't easy. The Ernie Ball VP series of volume pedals are built like a shovel! They may be totally rebuilt, rebuild kits for the string system and the long-shaft pot with the sintered brass pulley are both readily available and quite cost efficient. The pedal itself and the housing are made of 1/4 inch thick extruded aluminum, the internal workings (electrics) may be easily modified, there's plenty of internal space to add an offset pot if you wish (like the pot on the side of the Moog EP-2 pedal ... which is all plastic). And the Ernie Ball pedals may be purchased new for less than $70.00 bucks if you look around. They'll easily outlive you.

So ... how does one go about using a volume pedal as an expression pedal? All that is required is what is known as an "insert cable". Sometimes called a "Y cable". There is a Tip/Sleeve plug on each end of the double cables, and a Tip/Ring/Sleeve plug on the stem of the ~Y~. Insert one of the Tip/Sleeve plugs into the "INPUT" jack of the volume pedal, and insert the other Tip/Sleeve plug into the "OUTPUT" jack of the volume pedal. Then, plug the Tip/Ring/Sleeve plug into the input of the expression device. In my case, I used a Dot Com Q181FJ module. Set up as described, with the Tip/Ring/Sleeve plug inserted into the Expression input of the Q181FJ ... it worked PERFECTLY!

Here's the hitch .... if you connect it all up, and when you go to use it things don't seem to be sweeping properly (it's just all fuddup sounding) ... stop, unplug the tip/Sleeve plugs, swap their positions in the IN/OUT jacks, and try it again. That totally fixes things, and you end up with a combat proof expression pedal, made of H.E.A.V.Y. extruded aluminum, that has rebuild kits available for it, and will outlive you! No more plastic expression pedals. The Ernie Ball VP Jr. is slightly smaller in length and width than the regular VP pedal. The VP Jr. is also a bit less costly. Either one has a really smooth and long pedal throw so they're easy to use. Make sure you use the 25k ohm version!!

Image

Image

Image

Image

This pedal is what I used to test out doing hard sync sweeps and soft sync sonic dehoyting during testing of the QX816. It worked fantastically! It should also be noted that this same pedal with the same insert cable works perfectly with the Dot Com Q142 Pedal Interface (I tested it myself). So it works with the Dot Com Q181FJ and the Dot com Q142 Pedal Interface.

Ok, I know what yer thinkin' .... "when can we hear these Ring Mods and this 4 VCO/Dual Voice setup, Rex?" ......

I'm going to make a VERY CRUDE video, shot from my wife's smartphone pretty soon. I just have to figure out what most 12 year old kids already know how to do. I'll also need to create a TooYube account (yuk!) and learn how to post videos. I may also just chicken out and make a crude audio MP3 of a few things. I know in my heart this needs to be done, so please be patient with me (as if you already haven't!).

Ok .. we're outta here! Class dismissed.

:tu:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Thu May 10, 2018 12:56 pm

MAJOR CORRECTION REGARDING THE CGS RING MODULATORS (FAULTY TRANSFORMERS!!):

Ok time for me to don the dunce cap. All this time I've been using this one single ring modulator as a test bed, I've been very leery of why it sounded .... well ... not like a ring modulator! After posting the entry just prior to this one here in the project thread, I felt even more unsettled about it.

I spent some time last night watching various videos that demonstrate this exact ring mod circuit board. Those sounded like what would be expected from a ring modulator. Mine never has. It has always required a shit ton of boost, both going in and coming out. And even then it just didn't sound right.

:bang:

I triple checked all of my wiring ... it dead matched all schematics I could locate, all was just as it should be. So, as a last resort I bench tested six of the transformers that this build calls for that I have on hand as "repair stock". Just to get it said, I've been collecting spare parts for this project synth as I go along, for just this type of situation. And boy am I glad I have done that!

I desoldered one of the transformers from the test *RM (*ring modulator). Bench tested it ..... far FAR different readings from the six spares. Hmm ... so I pulled the second one from the RM board ... bench test again showed far different values against the six spares.

:hmm:

All I can figure is that the vendor I bought them from had mismarked them, or perhaps they were mismarked at the factory. All I know is that they are nowhere even close to the six spares.

:despair:

Tired and ~done~ with messing with it, I went to bed. Straight away this morning I replaced the two oddball transformers with two of the spares. Connected everything up on the operating QX816, took a deep breath .... pressed a key down on the keyboard ..... and ......

Holy Radioactive Coyotes Batman!! It friggin WORKS!!!
:yay: :trampoline: :nod: :headbang: :spin:

Suddenly, the LED diodes are now lighting up as signal is passed through them (they never had previously), and it needs exactly ZERO boosting from the Q114++. Matter o' fact, I took all of any ~boosters~ out of the signal chain, and it works great!!! It ... at last ... sounds like a ring modulator is expected to sound. Without any additional gain boosters other than using the feedback loops in the onboard Q162 waveform mixers. It actually doesn't need them, but adding some feedback boost creates a nice warm distortion. I suspect the LEDs are clipping a bit ... which is totally fine.

The LEDs look really cool .... if you set up the VCOs to slightly ~beat~ from detuning them a wee bit, the LEDs alternate brightness, gently lighting two while two others dim, then as the ~beat~ swings around the brighter ones dim down and the darker ones brighten up. They continue with this gentle back-n-forth dim/darken routine as long as the VCOs are detuned enough to cause the wonderful beating sound.

:tu:

It looks so cool that I am wanting to remove the center strip (the part of my synth with the ring mod controls in it which is between the upper and lower rows of modules) and drill eight new holes in it to accept four LED holders on each side (4 for each RM pair). That will look sick! The counter-flashing LEDs also confirm there is sufficient signal levels passing through them, and can assist in patch-troubleshooting. If things aren't set up properly the LEDs will either be super dim or totally unlit.

So now I am excited to test out Germanium diodes in it. Easy enough to swap them in the test RM since I used sockets for the diodes.

This is all EXCELLENT news!


One thing to add, the two transformers I used in the test RM were not obtained from a well known vendor. The six spares are from Mouser, and all six bench tested to proper values. So there's a little lesson.

Total transparency and full disclosure here .... I failed to follow one of my own cardinal rules .... "Test all components prior to assembly." I always test parts before I solder them in .... I preach about it, I admonish and lecture others for not doing it ... and the one time I don't follow my own rules ... it bites me square on the ass! I deserved that. It was the Universe clobbering me on the head with a lead sledge hammer!

:doh:

That won't happen again, I honestly learned here.


Ok, so if anyone was sketchy about using the CGS Passive Ring Modulators from Modular Addict (which are based on the age old Ken Stone design) due to my testimonies about it ... let not your heart be troubled any longer. They work GREAT .... and they will work with signals straight out of the VCO waveform outputs without any special or extra ~boosting~ inolved. Not going in to the RM, and not coming out of the RM.

Now .... I need to hunt up the "LED as diodes in Ring Mods" thread I authored several months back and add this correction. It was really bugging one of our members that the LEDs weren't lighting up ... it really threw him. So now he will be vindicated.

Can't wait to test Germs and different colored LEDs!!!!!

Thanks .... :tu:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

Dr Gris
Wiggling with Experience
Posts: 369
Joined: Tue Feb 14, 2017 5:14 pm
Location: Sweden

Post by Dr Gris » Thu May 10, 2018 2:28 pm

Great!!
We wanna hear it!!!!!!!!!

Transformers rarely fail, are you sure it's not a wrong ratio thing?
What ratio is it supposed to be in this ringmod (too lazy to look it up...)

//Magnus

Dr Gris
Wiggling with Experience
Posts: 369
Joined: Tue Feb 14, 2017 5:14 pm
Location: Sweden

Post by Dr Gris » Thu May 10, 2018 3:30 pm

Nevermind, the transformers are supposed to be 1:1.
The Ken Stone/CGS variant points to a low budget transformer, maybe they impart a cool sound?
I have some old chunky Western Electric transformers I don't use.
Maybe I should try them, after all, big is beautiful!! But I think their impedance is way off...

While waiting for your demos, hint...hint... I'll have another listen to CZ Rider's brilliant YouTube video.

//M

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Thu May 10, 2018 5:15 pm

*Dr Gris ... one of the transformers has an intermittent short on the primary side (I'm not totally certain but it may be between the center tap and ground ... "the irons" ... it comes and goes if you flick at it with your finger. The other one is just plain old wrong for the application ... about 6:1. Both have much higher impedances on the primary and secondary windings than the spec calls for as well. I don't think they failed, I think I bought them like that.

Spec calls for 10K:10K. All six of the ones I bought from Mouser are precisely that. I ordered the same part numbers from both suppliers .... Mouser got it right, the ~other place~ blew it.

If you want to build some like *CZ's, use Edcor transformers. They offer 10k:10k, the difference being they spec at 20hz to 20khz, compared to the ones suggested by CGS/Modular Addict are 300hz to 3khz (I believe it's something like that). But to be honest, I don't hear a problem with the ones I got from Mouser, even though they're a "lower spec" than Edcors. The Edcors are also fairly massive ....

Image

.... so I'd probably just do "point to point" with the four diodes rather than using a PCB if I were going to use Edcors. A mounting method and location would also need to be worked out.

These CGS ones sound ~about~ the same as the Dot Com Q116 Ring Modulator .... there's a video by "Synthmania" where that guy uses a Q116 to make some really nice sounding bass stuff. I can do pretty much the same thing with the CGS units. So I don't think I'm going to end up putting Edcors in my synth. Besides, my cabinet is fairly filled with normalizing rails, bus bars, terminal strips, piggy-backed PCBs on panels, and of course Ring Mods ... along with the associated wiring that goes with all of that stuff.

In any case, I located the trouble .... that's the important part of all this.

Ok, I'm eating supper, then I'm going to refit the test RM with Germanium diodes. We had to go out for an appointment, we just got back ... just before we left I located all of the diodes I had matched up several months ago (silicons, Germaniums, and other colored LEDs) so that's a big help since I won't have to go through matching diodes again!

So I can sit down and test out various diodes in the refitted test bed ring modulator.

And ... I think I have found a place to drill four more 5/16" holes for the eight LED holders on the center strip (four on each side). I'm pretty ~set~ on doing that!!

Later!! 8-)
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Sat May 12, 2018 4:01 pm

No Update. But I am going to set up a YT channel:

... ok ... I've started a "HELP ME!!" thread on getting some help to set up a Tou Yube channel....

I'd appreciate any input.

Link = viewtopic.php?t=201231&highlight=

So I guess I'm actually gonna do this.

:w00t:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Sat May 19, 2018 9:53 am

ASSEMBLY OF QX816 No2 (DONE MORE GOODER):

Got started with the second QX816 Dual VCO Multi Circuit module. Refining the construction this time. The first one was all about getting it to work properly (which it did, straight off of the bench). On this second one, I'm using a few refinements that will help move it along quicker, as well as (possibly) making it a better module.

I began with installing the CV Input jacks. Those are "Normally Closed Switching Jacks", just like the first one. This time, I've installed a "ground rail" (ground buss? ... ground wire? ... ground thingie?) along the aligned shield connector tabs. I'm hoping this will assist in making it easier to construct since I can tie the wires that need to connect to the shield tabs on to this rail.

I also used the Zinc Crystal Impregnated Gas Tight Paste (brand name of "OxGard") on the panel side and the jack side of the internal tooth lock washers that I use on 1/4" jacks. In the image below, the red line depicts the synth panel. I wasn't able to use the extra thick stainless steel flat washers on the 3mm thick FPD panel, so I just used the ~stock~ nickel plated flats (same thing I did on QX816-No1). But I did use the stainless steel lock washers shown in the image on both dual VCO panels.

Image

(below) Not much to see here, just some neat looking perspectives. The bare wire is called "pretinned solid copper bare bus wire" and is in 16 gauge ... really nice stuff to work with, and sold by weight instead of length. I bought the 1/4-pound spool which works out to about 32 feet.

Link to buss wire vendor =
https://www.allelectronics.com/index.ph ... =buss+wire


Image

Image

Image

(pic above) Look at where the arrows point to, you'll see the gas tight paste. Rubbed between the fingers a grit may be felt within the grease-like grey paste. That ~grit~ is the zinc particulates. The zinc crystals assist in cutting through oxidation (which can be as thin as a single molecule) which forms a few milliseconds after a fresh cut is made into the surface of the aluminum. The paste substrate is the "gas tight" part of the mixture. Once the zinc crystals cut into the aluminum (along with the teeth on the lock washer) the paste prevents oxygen from creating the molecule-thick oxidation layer. This process helps to create and retain a solid electrical connection between the two components. The use of that aggressive internal tooth lock washer cuts right through the anodized surface of the panel. It may not be "perfect" but it is as much attention to these details that I can give. Believe me whan I tell you "Deez Nuts Are TIGHT!" regarding the jack nuts .. I use a lot of torque when tightening them down .. so as to give the teeth on the lock washers enough pressure to really dig into the aluminum panel and the barrel flange on the jacks. The lock washers were carefully selected, so they really have a very nice fit around the jack barrels with very little ~slop~. This ensures that the teeth are solidly digging into the barrel flange and not just barely hooking on to the flange's edge.

The long "grounding wire" (for lack of better words) soldered to the twelve shield tabs creates a place to ~tap~ into for connecting any/all chassis ground wires ... it will also serve as a place to connect a heavy conductor (two parallel 16ga aircraft wires) that will connect the chassis to the normalizing rail inside of the cabinet, which does double duty as a kind of "grounding bar" .... the rail on the left of this image below (this image was used in a previous entry, so where it says "I'm thinking of..." means that "I have decided to" ... heheh).

Image

That normalizing rail will use parallel runs of 16 gauge aircraft wire which will themselves connect to a single "Technical Earth" termination point that will be as close as possible to the "Safety Ground" wire that comes into the synth from the main power cable.

Image

So .. from the synth panel via the shield tabs of twelve jacks ... to the normalizing rail suspended inside of the cabinet ... to the "TE Point" on the power control module. The same setup will also be used on the lower row of the synth, where the QX420 Dual VCF Multi Circuit panel lives, along with the VCA and it's buddies.

Why go about all of this? Since this synth will be very normalized, there won't be very many patch cables used. That said, something needs to be done to handle chassis grounding since sometimes there will be no patch cables used at all. This setup will make sure that all of the panels have solid chassis ground paths even though there may not be any patch cables connecting all of them together. It may turn out that I won't need to attach a wire run from the normalizing rails to the Technical Earth point on the power control module, however it will still require that all of the panels are connected together via the normalizing rails themselves. So, even if it turns out that running wire to the TE point isn't needed, every other single detail I've laid out in this entry will still be required to connect all of the panels' shields together, just as if I were using handfuls of patch cables.

Why wouldn't a wire run connecting the panels to a TE point be needed? Because in these unbalanced synth, the Zero Volt wires in the module power cables carry that, and the Zero Volt rail is connected to Earth at a point very close to the ground wire on mains inlet. I'll need to do further research on whether using a wire run from the normalizing rails to the TE point on the control module is required. But for now, doing the rest of this chassis grounding scheme rolls forward.

That's all for this entry. Next will be more QX816-No2 progress, but I'm not going to post the entire progress of QX816-No2 because other than just a few changes it will be the same as No1, which has already been documented.

Hopefully I'll get the 12 Cab bus bars mounted which I'll also document here. All of the wood and bus bars are cut and ready to mount in the 12U cab, just need to get in there and get moving on it.

Thankies!
:tu:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Eric the Red
Veteran Wiggler
Posts: 737
Joined: Tue Jun 02, 2015 12:43 pm
Location: Detroit, MI

Post by Eric the Red » Sat May 19, 2018 6:24 pm

Keep on trucking along Rex!!
Abandon all hope, ye who wiggle.

User avatar
Dr. Sketch-n-Etch
Super Deluxe Wiggler
Posts: 7777
Joined: Thu Jan 21, 2010 1:46 pm
Location: Vancouver, BC, Canada

Post by Dr. Sketch-n-Etch » Sat May 19, 2018 10:05 pm

You are my hero, Rex Coil 7 !!!

:hail: :hail: :hail: :hail: :hail:
We've moved fast, we need love; a part we offer is our only freedom

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Sat May 26, 2018 1:18 am

Eric the Red wrote:Keep on trucking along Rex!!
One bite at a time, I'm eating that elephant! Thanks Eric. :tu:
Dr. Sketch-n-Etch wrote:You are my hero, Rex Coil 7 !!!

:hail: :hail: :hail: :hail: :hail:
Damn ... thanks Dave. You're no slouch yer own dam self! Every time I see your name pop up somewhere with a new post I think to myself "Cool! ... wonder what he's cooked up THIS time!". I've never seen ANYone produce something from idea to tangible working construct as quickly as you. Thanks for the compliment .... and .. right back atchya!

THE QX500 - Multi Circuit Dual VCF:

I have put off and put off and put off ordering the dual VCF panel from Front Panel Express .. something about the QX420 was bugging me, but I wasn't sure what was making me so pensive about ordering it. It was done. It all checked out. It looked, well, ok-ish I guess. But something held me back.

I put it aside for a bit while I worked on figuring out the Pre-VCF Voice Mixer (a quad-VCA mixer that feeds into a Moog CP3 discrete transistor distortion circuit, which sends it's output into the Dual VCF panel). After several days of bugging the hell out of the 5U subforum folks about the Quad-VCA/CP3 thingy ... the PROPER rendition of the Dual Filter Array panel (QX420) finally hit me.

I present to you, the QX500. I broke from tradition of adding up all of the Dot Com module model numbers and using the sum of the numbers to name the panel. QX500 just works.

Following the signal flow for this description, it includes a Q147 Distributor, Q107 State Variable Filter, Q162 Filter Mixer, Q150 Transistor Ladder Low Pass Filter, Q109 Envelope Generator, and the "RC7-SJP" (Slope Jumper Panel). It has expanded the original 4MU QX420 to 6MU, however it replaces 7MU of Dot Com modules and 1MU for the RC7-JSP. And it is GLORIOUS! I've designed it to match the QX816 Dual VCO Voice panels (panel color, lettering typeface, knob design, and basic overall look and feel).

Here's a screenshot of the "Master" file (large holes to depict knobs and/or flat washers ... whichever takes up the most space on the panel is what is depicted).

Image

Diving straight in to it's features and functions, here's some FPD screenshots followed by a mockup diagram of Dot Com modules that make up the QX500, along with a few written descriptions.

I'll begin with SERIES MODE. This mode puts the two VCFs in series. The toggle switch marked "ORDER" is for swapping the two filters' order ... switch UP = Q107 into Q150 ... switch DOWN = Q150 into Q107. The switch marked "Neg/Pos" is the output control for Channel 3 of the Q147. I've replaced the attenuverter pot with a toggle switch. Switched to "Pos" and the first filter is sent a positive signal ... switched to "Neg" and a negative signal is sent. Both settings send the signals at 100% volume, and each filter's input level knob is used to set the input levels. And obviously, the switch marked "PARA/SERIES" (placed in SERIES mode in the diagram) selects whether the filters are in series or parallel. Any controls that are relevant to this mode are colored in.

Image

Here's how the signals are patched together, represented by the Dot Com modules that I've used in this panel.

Image

Next, we go to PARALLEL MODE. Same rules apply (relevant controls are colored in, switch settings are circled, and so on .. just like the first pair of images above). FPD screenshot first, Dot Com modules patch second.


Image

Image

Next is the Onboard Envelope Generator mode. This makes use of the onboard Q109 Envelope Generator. The toggle switches marked "ENV" engage the EG for each filter. When UP, the onboard EG is used, when DOWN, then the outboard normalized CV signals that route to the CV inputs that are controlled by attenuverters are sent to the filter(s). So "UP" = onboard EG, and "DOWN" = whatever is selected by the "FRQ" toggle (position A, Off, or B) that selects whatever sources are connected to trunk A or B at the time.

A couple of notes on this; the two small circles at the extreme low/right of the panel are the Gate LED, and the manual Gate button. Also, notice the "Sustain" control is a 60mm B100k slider pot (Alpha). When I saw this ADSR configuration on the new Moog Grandmother, it inspired the hell out of me. It really makes a lot of sense to me. So, I swiped it! It takes up less space on the panel than another 3/4" knob would, and it's the "oddball" function of an ADSR envelope ... it's the only one that deals with signal level, the A,D, and R all deal with time. It just makes sense to make it a slider. I really like how it looks on this panel as well. Dimensions used were taken directly from the Alpha datasheet (screenshot of those dimensions posted beneath the patch image).


Link to Mouser for 60mm B100K slider = https://www.mouser.com/ProductDetail/312-9302F-100K

Image

Image

Image

Next is the Filter Bypass routing. Channel 4 of the Q147 Distributor is not attenuated in any way. Whatever comes in is precisely what is sent out of Ch4. It is routed directly past the entire dual filter array and is sent straight to the VCA Input Mixer Ch4 Input where it may be mixed with the filtered signals (or run by itself, or completely cut from the VCA Input mix altogether).

Image

Next up, here's all of the controls that are relevant to the Q107 State Variable Filter and the Q162 Filter Mixer. The ~dark red~ dots depict the CV controls, the other dots depict the filter output and output controls. I drew a line to visually connect the toggle switch (A/Off/B), the CV level knob (largest circle), and the CV input jack. Whatever the label above the toggle switches say are what also apply to the knob and the jack. This method of labeling reduced panel clutter and overall cost of the panel (the more lettering on the panel, the more money you must send FPE).

The upper row of dark red toggle switch circles are functions that are otherwise changed through the use of jumper links on the PCB. "ENV" engages the onboard EG when UP ... "BW" is like installing a jumper link on the "Constant Bandwidth" jumper pins ... "REZ" engages the jumper pins that make the resonance LOUD AS HELL on the Q107. When flipped up, it is very easy to get that ~squelch~ by turning the REZ knob up to about 11 o'clock. In the DOWN position it is set up with the "factory setting" for the resonance level function.


Image

Here's the Q150. Same set of rules apply. Same lines connecting the toggles with the knobs with the CV input jacks to cut down on panel labels and costs. The two BIG blue dots are the Cutoff Freq and Resonance Level knobs, the six blue dots grouped together on the upper right corner are the six Slope Jumper jacks used to change the secondary slope selection to either 18db/3-pole, 12db/2-pole, or 6db/1-pole. When the "4P" toggle is in the UP position, the filter has a 4 pole slope, in the down position the slope is determined by the jumper plugs inserted into the slope jumper jacks. The "LVL" toggle (green dot) is like installing the signal level compensation jumper that compensates the reduction of signal when higher resonance settings are used. And of course you already know what the "ENV" switch does.

Also note that one knob has a different glyph above it than all of the other CV level knobs ... it's reads "O - <" ... which describes that pot as being a unipolar control. All of the other CV level knobs/pots are bipolar (hence the "> 0 <" glyph above them).


Image

Lastly, I'll explain what I'm doing with the VCA Input Mixer. You've seen a few references to "switches" on the VCA Mixer in some of the diagrams above. Those are references to the three large toggle switches on the panel that is connected to the Suit and Tie Guy CP3 Mixer I have .... this beasty shown below (along with the main output VCA).

Image

NOTICE SERVED! .... The STG Mixer shown in the picture has numbered days. I'll be building my own Moog CP3 mixer for that position which will have a few tricks of it's own. I am unable to normalize the STG Mixer to anything (either inputs or outputs) due to the way it was designed and constructed. Don't take me wrong, the STG Mixer isn't a BAD thing, in fact it's fantastic sounding. But it was not designed to be normalized to anything so the I/O jacks must be used. This is a problem for me. It's why I built that panel with the toggle switches and the red wires coming out of it in the first place, so I could have a means of connecting signals that are behind the front panels (normalized signals). I call that panel the "Spark Plug Panel" because the red wires look like spark plug cables and the right angle jacks look like spark plug caps.

ONE LAST ITEM: I finally put together a knob scheme I prefer on the first QX816 Dual VCO. After a few weeks of using it, I settled in on a knob and color combination that works for me and works for the synth as well.

Image

I went back to the chickenheads on all of the rotary switches, and GIANT "Ravenheads" on the VCO Freq fine controls. The two 1/2" knobs in each lower corner look like they're a foot long in this picture, no idea why, it just came out like that. But they're not extra long, they're just standard 1/2" inch Oberheim clone knobs in medium blue. I'm also short a few toggle switch covers. The green ones are almost always normally UP, the red ones on each side are the all-important output selectors that select between the Q141 and the Q161 outputs sent to the Main VCO Mixer in between the two QX816 Dual VCO panels. Now I need some yellow ones and a few others to finish up the panels.

Ok, that is all for tonight. I'll be sharing the new Pre-VCF mixer and options panel (4MU wide) in the next few days. But only the design aspects, not the actual panel. I'm working out a few details in private discussions with Tony of Oakley Synths for the design elements (hint hint). So we'll see what comes!

G'night folks! :nod:
Last edited by Rex Coil 7 on Sun Jun 03, 2018 10:53 am, edited 2 times in total.
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Tue May 29, 2018 1:00 am

QX7SJ - VOLTCON - 2am WIRING DIAGRAM:

I finally got the dual VCF panel pretty much done. Renamed to the QX7SJ. The "SJ" part refers to the "Slope Jumper" in the panel. THe "7" is due to this being the SEVENTH iteration of this panel. Yep ... seventh! Changed some of the lettering from capitals to lower case, moved a few things just a little, moved around a few of the I/O jacks on the bottom. I've gone over it dozens of times, making certain that some things look just like the dual VCO panels (QX816). Below is the "big holes" MASTER file .... under that is the "FINAL" file I'll be sending to Front Panel Express to be made into a real panel. Ready to go, this Dual VCF came out to $152.69 (ouch!).

Image

Image

On to the VOLTCON. This is what happens when I'm unable to sleep due to "really bad thoughts" interrupting slumber. It's that time of year for me ... on May 29th 1992 a good friend of mine and coworker was assassinated by an MS13 member. Contrary to what the politicians will assert, MS13 has been trouble since the late 1980s, especially in the south east Los Angeles area. My friend was shot 5x in the back and once in the back of the head. I was unable to protect him, or help him, because I arrived too late. So this time of year is difficult.

So last night I got out of bed at 1:37am because the demons were having a fucking party in my head. This was an excellent distraction, so I just let my ideas flow like soothing water. It outlines the signal flow of everything that goes on before the signals hit the Dual VCF input.

This is part of the VOLTCON Pre-VCF mixer. That mixer includes three separate circuits, with six instances of one of those three circuits. The diagram signals flow begins at the 4 VCOs, as well as the Q118 Instrument Interface (which is where my Kurzweil PC3A6 comes in as another "voice"). There are 4 ring modulators (RMA, RMB, RMC, RMD).

* RMA and RMB are the two ring mods that are between the VCOs in each of the 2 "voices".

* RMC goes across the two output busses of the MAIN VCO Mixer.

* RMD uses the signal from the Kurzweil (post Q118) and then it also uses the two Voice bus outputs, with a toggle switch to choose which one will be the carrier ("AxB"). The "MxC" toggle is the "Modulator/Carrier flip flop" switch. So RMD has the Kurzweil as one signal input, and either the Bus A or Bus B signal from the MAIN VCO mixer as the other input. The switch marked "AxB" is an A/B switch to swap between the two Voices, and the "MxC" then flip flops between the Kurzweil and either of the two voices swapping their modulator/carrier assignments.

Whew!

This is the diagram that I worked up to keep all of these signal routings square in my head. This diagram does not include the PCB connections or any of the controls such as pots, jacks, switches.

Image

Once I got the signal flows worked out (everything from the VCO outputs to the Pre-VCF Mixer inputs), I then sat down at the computer and started working up the actual Pre-VCF panel. In honor of Robert Moog's early efforts on a VCA fed CP3 mixer, which he named the "volt. con. mixer" (obviously meaning "voltage controlled mixer"), I sortof tweaked the heading Dr. Moog put on the drawing into the word "VOLTCON" as the name of this module.

It includes:

* 1ea Dot Com Q113 Mixer pcb.
* 1ea Manhattan Analog "DTM" (Discrete Transistor Mixer) which is a CP3 clone.
* 3ea Oakley Dual VCA boards for a total of 6 VCAs (one for each input channel I'll be making for the Manhattan Analog CP3 mixer).

The Manhattan Analog board and the 3 Oakley Dual VCA boards will easily fit in the 4MU panel's space. The Dot Com Q113 mixer board will be mounted "piggyback" on to the Q118 Instrument Interface module which is mounted directly next to this Pre-VCF input mixer panel. So the Q113 board will mount on the module next to the Pre-VCF mixer, but it's pots and jacks will be mounted on the VOLTCON panel.

Here's a simple mockup of the Oakley and Manhattan Analog boards sitting on a 4MU blank Dot Com panel. The fakey boards are to exact scale. All of the boards will be mounted on standoffs parallel with the panel. The 3 Oakley Dual VCA boards will be separated by 1" long standoffs, for a total of 3 inches in height. 3 inches is the same height as putting one Dot com board on top of another Dot Com board when 1.5" standoffs are used. 1.5 x 2 = 3" .... 1 x 3 = 3" ... same same.

Beneath the mockup picture is a screenshot of the VOLTCON Pre-VCF mixer panel. The VOLTCON panel is at $98.34 so far.

Image


Image


If tonight is anything like last night, I'll be up in about 3 hours from now, working on the second VCO panel some more. It's coming along nicely.

A few "wish-Ida-duns" at this point in the project:

** Used stainless steel T-nuts in the cabinet.

** Used Oakley "1 of 3" VCOs since they're more Moog-like than the Dot Com Q106s (maybe, maybe not ... this one could go wither way I think).

** Used CP3 mixers in the VCOs (where I used Dot Com Q161s).

** Used a CP3 mixer in the Dual VCF panel in place of the Q162 Filter Mixer (the Q162 may still work out really well, what with the feedback overdrive loops I've built in to it).

Things I may still do:

* Add a VCA to each VCO ... this can be a helpful thing for FMing VCOs together, as well as other functions.

* Make another CP3 mixer that goes after the filters (feeding the main output VCA).

Off to bed. Here's to hoping my little friends ~the demons~ don't give me this shit tonight ... :msnsmack:
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Sat Jun 02, 2018 12:41 am

MORE TIME ON THE SECOND VOICE PANEL - VOLTCON 1970 - RING MOD TESTS:

1st thing; I've spent more time on the 2nd QX816 Voice panel. It's getting there.

2nd thing; I think I have the VCA/CP3 module about all worked out (named "VOLTCON 1970"). I've ordered 4 Oakley Dual VCA boards. I got lucky, the day they were paid for was a day when the global stock market went all weird due to something that went on in Italy ... however/whatever happened changed the British Sterling to US Dollar ratio in my favor, so they ended up costing me a bit less. I've also ordered 4 of the Manhattan analog "DTM" (Discrete Transistor Mixer) PCBs. I'll be able to complete work on the design of the VOLTCON 1970 panel in Front Panel Designer when all of those boards arrive (I need them to properly measure the PCB standoff hole locations). 3 Oakley Dual VCA boards and 1 Manhattan Analog DTM board are required to construct the VOLTCON 1970. The extra boards are just for ~whatevers~. (It might be nice to work out a way to put a VCA attached to each of the 4 VCOs ...makes for nice FM patches as well as other little tricks). Here's a screenshot of the "so far" panel design (I know it may seem chaotic, but it's not like I'm selling these or something, as long as I know what all of the lettering means that's really all that matters);

Image

3rd thing; I spent the better part of all day today and a good deal of the previous day testing different LEDs in the workhorse ring mod. As a reminder, that ring mod is a CGS Passive Ring Modulator that I have installed sockets where the diodes are supposed to go. This modification permits easy installation and removal of different diodes and LEDs for testing purposes.

I tested 5 different types.

Image

To gain as much data as possible, and to establish working comparisons, I used a few different test methods. Before doing audio tests, I did some bench measurements. I used a cheapo digital meter, as well as a 9v battery with a 1k resistor soldered in line with one of the test leads.

VOLTAGE DROP OBSERVATIONS:
The measured open circuit voltage of the battery was (with the 1k resistor included in the circuit) 6.80vdc. I connected each different LED to the powered test leads and measured the voltage drop across the LED's leads and recorded the observed drop.

1.) Rectangular ~RED~ LEDs voltage drop = 1.93vdc.

2.) 5mm Round ~Yellow~ LEDs voltage drop = 1.89vdc.

3.) 3mm Round ~Water Clear BLUE~ LEDs voltage drop = 2.78vdc

4.) 5mm Round ~Dark RED~ LEDs voltage drop = 1.85vdc.

5.) 5mm Round ~Standard RED~ LEDs voltage drop = 1.7vdc.

Note the relatively high voltage drop figure for the 3mm Water Clear Blue LEDs (#3).

METER "DIODE TEST" OBSERVATIONS:

I then used the cheapo digital meter, set on "diode test", took measurements and recorded my observations. The numbers presented here are the average of four LEDs of each type.

1.) Rectangular ~RED~ LEDs = 1790.25

2.) 5mm Round ~Yellow~ LEDs = 1715.75

3.) 3mm Round ~Water Clear BLUE~ LEDs = (meter unable to provide measurement, unknown reason - multiple attempts made).

4.) 5mm Round ~Dark RED~ LEDs = 1655.25

5.) 5mm Round ~Standard RED~ LEDs = 1535.75

AUDIO TESTS AND OBSERVATIONS:
Lastly, I gathered up my toys and set about going into the studio. I patched the test Ring Mod rig up to VCO #1 (Modulator) and VCO#2 (Carrier), and ran the output of the Ring Mod to a Dot Com Q108 VCA, then out to my desk mixer (12ch Carvin). The VCOs are Dot Com Q106s, a sawtooth waveform was used for both Modulator and Carrier. Channel volume of the Carvin mixer was set at 0db, and the Main Output was set to -12db. The gain level of the Q108 VCA was set to 9 o'clock.

These settings went completely untouched throughout the audio tests.

This arrangement produced a certain number of LEDs on the Carvin mixer to light up (those are the output level meter LEDs). Here's my observations:

1.) Rectangular ~RED~ LEDs = 4 solid LEDs lit on the output meter, when VCOs set to a slight detune the beating caused the output level to fluctuate a little. Nicely rounded distorted sound in unison tuning, aggressive sound when tuned a perfect 5th apart. Nice looking alternating lights on the LEDs during beating.

2.) 5mm Round ~Yellow~ LEDs = 4 solid LEDs lit on the output meter, when VCOs set to a slight detune the beating caused the output level to fluctuate a little. Nicely rounded distorted sound in unison tuning, aggressive sound when tuned a perfect 5th apart. These yellow 5mm produce a dimmer light, and are not as dramatic looking when VCOs are beating. Nice sound all around.

3.) 3mm Round ~Water Clear BLUE~ LEDs = 4 solid LEDs lit (and also just barely lighting the 5th row of LEDs) on the output meter, when VCOs set to a slight detune the beating did not cause the output level to fluctuate. Much more sharply distorted sound in unison tuning, VERY aggressive sound when tuned a perfect 5th apart. Nice looking alternating lights on the LEDs during beating, but only if the output level of both VCOs was reduced beforehand, otherwise. When VCOs are at max output level these LEDs are down right RUDELY BRIGHT when viewed on-axis.

4.) 5mm Round ~Dark RED~ LEDs 4 solid LEDs lit on the output meter, when VCOs set to a slight detune the beating caused the output level to fluctuate a little. Nicely rounded distorted sound in unison tuning, aggressive sound when tuned a perfect 5th apart. Alternating lights on the LEDs during beating not as dramatic as the rectangular LEDs. These appeared more like the Yellow 5mm LEDs as far as their alternating flashing goes.

5.) 5mm Round ~Standard RED~ LEDs = Pretty much equivalent to the Rectangular Red LEDs. 4 solid LEDs lit on the output meter, when VCOs set to a slight detune the beating caused the output level to fluctuate a little. Nicely rounded distorted sound in unison tuning, aggressive sound when tuned a perfect 5th apart. Nice looking alternating lights on the LEDs during beating. Very much the "mama bear/just right" set of LEDs.

SUMMARY:
Note all of the observations on the 3mm Water Clear BLUE LEDS relative to the others. MUCH higher voltage drop, diode test so high the meter went ~TILT~, RUDELY bright light output level, very aggressive distortion sound (especially with interval tunings between VCOs), and higher overall output levels seen at the desk mixer. I even tested these LEDs with boosted triangle waveforms as well as a mix of sine and triangle waveforms (with feedback boost added to that mix) ... super killer sounds produced. I can only imagine what adding a CP3 mixer into the works would do with this set of LEDs in the ring mods.

Other than that, all of the other LEDs produce fairly equal results in all categories. The 3mm Water Clear BLUE LEDs were the definitive SNOT MEETS FACE - RUDE ASS PUNKS of the group.

FINAL THOUGHTS:
I find that all of these observations align perfectly with the discoveries I made using different types of diodes (and LEDs) as clipping components in various guitar overdrive and distortion pedals. The higher the voltage drop, the louder the output. And the higher the voltage drop, the more ~sharp~ the distortion "granules" are.

This would also align with how the use of Germanium diodes in passive ring mods makes the output level of the ring mod so much lower than when using LEDs ... precisely the same thing happens in distortion pedals.

Silicon diodes fall right in the middle between Germanium diodes and LEDs in all regards. This is true in both passive ring modulators and distortion pedals.

Personally, for my preferred use of ring mods in my own synth, I find the Water Clear LEDs to be my favorite. 3mm or 5mm, it doesn't matter, they're the same when it comes to electronic performance (nothing to do with the amount of light produced). The only problem is, the way I'm mounting LEDs in the VOLTCON 1970 mixer panel, those little "lazer bright" water clears will be right at my eye level, and pointed directly at me when I am at the keyboard. That said, I think I'll put the water clear LEDs on the ring mods that are mounted in the Center Strip (and hidden from view), and install "standard red" LEDs in the LED holders of the VOLTCON 1970. They'll be subjected to higher signal levels there (fully adjustable, of course) and they'll be easier to handle as they shine their light. They "flicker" nicely when the Modulator and Carrier are detuned a little. They do that so well, I can actually create perfect tuning without any volume (no sound) ... just using the flickering to determine when the sources are tuned properly. This also applies to perfect intervals as well.


~Fin~.
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
Rex Coil 7
Super Deluxe Wiggler
Posts: 6582
Joined: Thu Nov 29, 2012 11:29 am
Location: Captain Of Outer Space

Post by Rex Coil 7 » Wed Jun 20, 2018 4:03 pm

REDONE RING MODS - PRO2000 - LOCK WASHERS - QX675 - PRE-VCF MIXER - PRE VCA MIXER - ADDITIONAL PANEL/CHASSIS GROUNDING METHOD - EXTERNAL SYNTH AS FM MODULATION SOURCE:

Lots to cover, let's get to it.

REDONE RING MODU-MONKEYS: After mucho testing of the passive rings modulators in many different situations with many different diodes/LEDs, I have settled into a solid hard core choice. When I tested Blue Water Clear 3mm LEDs as diodes in these passive ringers, "I found the one!" .... finally! So I set about the task of removing all of the other diodes/LEDs from all four passive ring mods used in my synth and replacing all of those with Blue Water Clear 3mm LEDs. It wasn't ~hard~ or difficult, it just took time and patience. So now, all four passive ring modulator-olas are outfitted with the same LEDs as their diodes.

There's one RM between each VCO in "VOX A" and one RM between each VCO in "Voice B". Now there's also an RM between the two "VOX" outputs (so an RM between VOX A and VOX B). And one more between either Vox A or Vox B .. and .. whatever I have coming in to the synth (just before the VCF input) via the Q118 Instrument Interface. Four total. Each one has it's own output level since I've installed a separate Ring Modulator Mixer who's main output goes in to the Pre-VCF mixer.

Whew! That's a mouthful!

Image

Pro2000: For going on about six year now, I've lusted after the Kenton Pro2000 MkII MIDI-CV converter. With rack ears, it sells at most US retailers for roughly $600.00 BIG ... ASS ... BUCKS. That was a lot of money for me to kick out for a MIDI-CV converter, especially since I already had a Kenton Pro Solo MkII for about 5 or 6 years. A few weeks ago I had to hit the Kenton website while looking for an operator's manual I needed. Just cuz, I looked at the Pro2000 page ... just to drool and lament over the fact I'd most likely never own one. I looked at the price and damned near fell over! $279.00 British Sterling. What the absolute FEK? I quickly checked the British Pound-to-US Dollar conversion for that day ... and the total came to just barely over $300.00 US for that particular day. With rack ears it came to $325.00 USD.

I told my wife about it, we checked our expenses for the month and figured Mac-n-Cheese would do for the rest of the month. ~Click~ ... went the "BUY" button. We then looked at a device that she has needed for about six months ... a MIDI/USB Hub for her Yamaha arranger synth that we bought her last year for our wedding annivesary. It's not one of those multi-thousand dollar setups, I think it was $279.00 new. It only has USB I/O with a headphone audio out ... that's it. so we needed a MIDI/USB Hub to connect that Yamaha to her other synths and FX. The Kenton was $79.00 British Sterling, and the Dollar-to-Pound conversion made it $99.00. ~Click~ went the "BUY" button again.

We saved a total of nearly $325.00 BUCKS buying directly from Kenton.

So I ended up with a MIDI-CV converter that really REALLY simplified the Super Mini-Mod project. With that Pro2000 I was able to eliminate almost $1,500.00 worth of support and routing modules that were on the "need it" list! Fifteen Hunnuds O'Buckity Bucks!!!! The addition of this wonderous converter has also changed the way that I am engineering the CV signal trunk lines and interface panels as well. In short, it's all much simpler now, with far less gear involved. AND ... AND AND AND ... I can haz four note polyphony (well, paraphony actually) just like my old prized Korg Mono/Poly! Not to mention 24 separate configurations saved into the 24 separate memory slots of the Pro2000.

My sweet sweet little kitty "CooCoo" voiced her approval by putting her "light bulb tail" on display for the photograph.

Image

STAINLESS STEEL 3/8" INTERNAL TOOTH LOCK WASHERS:
I suppose it appeared as though I was so busy with the yakkity yak in forum that I wasn't working on my synth project. Well, that isn't totally true (there has been a LOT of progress on ancillary tasks other than assembly and soldering on the Super Mini-Modular). But, you'd be right saying I haven't hardly touched the second Dual VCO panel in about two weeks. Reason had nothing to do with ~burn out~ or laziness ... it had everything to do with running out of internal tooth lock washers! When we ordered more from our local supplier, we got the usual "four days" story on when they'd arrive. Four days became nearly three weeks! Finally, they've showed up ... 100pcs ... wahoo! So last night I got ta bizzy and started back into production on the second VCO panel (QX816B).

I got sixteen more jacks installed, with gas tight goop. I also got the second "chassis grounding bus wire" soldered on to all sixteen additional jacks.

Image

Image

If you've ever wondered how well this internal tooth lock washer gig I use for making chassis ground connections through an anodized panel works out ... I went ahead and applied full assembly torque to a jack with one of those lock washers and then removed the jack to see the depth of the cuts into the anodizing (just to see if I'm wasting my time with that effort). Here's the result .... nice DEEP cuts ... ten cuts per lock washer (each lock washer has ten cleats). There are TWENTY EIGHT jacks on each Dual VCO panel ... twenty eight x ten = two hundred and eighty cuts into the anodized panel, with gas tight paste added to protect the freshly cut connections.

Image

ADDITIONAL METHOD TO CONNECT A CHASSIS GROUND CONNECTION POINT TO A PANEL:

This seems to work out pretty nicely. The 1/4" crimp on eyelets fit very nicely over Taiway toggles. Why Taiway? Taiway toggles have a larger barrel outer diameter than Mountain mini-toggles do. Mountains measure at 0.230" and Taiways measure at 0.240". This fact makes the Taiway mini toggles a solid candidate for mounting an eyelet.

Image

Image

Image

QX675 MULTI CIRCUIT DUAL VCF PANEL ARRIVED!!!: Looky looky what showed up in today's Fed Ex truck! Right on mens!! I have every single little tidbit to complete this panel on hand SAVE TWO. I need the 60mm B100K slider pot (and a knob for it) ... and eight more three conductor ~pot~ pigtails that are used to connect pots to Dot Com PCBs. The slider pot is a Mouser thing, the knob is a Small Bear thing .... and the pigtails are a Dot Com thing. I'm buying TEN of the Dot Com toggle switches that already have TWO of those three conductor connectors with pigtails attached to the toggle switch. Each one of those assemblies has two pigtails, and cost $6.00 each. So that makes each pigtail $3.00 each ... and buying TEN brings home TWENTY of the pigtails with connectors.

Other than that, I can get started on this panel lickety split. I'll start with the jacks first (all THIRTY of them ... twenty four along the bottom and six more for the SLOPE JUMPERS Tip/Ring/Sleeve jack field in the upper/right corner).

Image

Image

PRE-FILTER MIXER PANEL AND POST FILTER MIXER (MAIN OUTPUT?) PANEL:

Here's a couple of screenshots of two panels that are in the "WIP" (work in progress) folder. The blue goes on the LEFT of the QX675 Dual Filter panel, the red one goes on the RIGHT of the QX675 Dual Filter panel. Both the blue and the red panels are 4MU wide. The Dual VCF panel is 6MU wide. This makes for a total of 14MU wide which is exactly how wide the cabinet is.

Image

Blue one is the master mixer, it has SIX VCAs (one per input channel, all six are Oakley Dual VCA boards) and one Manhattan Analog clone of the R.A.Moog CP3 mixer. There are to be several mixing options, as well as routing options. It ultimately feeds the sum total of all signals into the Dual Filter.

Image

Red one is the final mixer that mixes the two filters (when they are in parallel) and sends it's sum total out to the final output VCA (a Dot Com Q108). This panel has TWO channel VCAs (one Oakely Dual VCA board) and another Manhattan Analog CP3 mixer. There's also some re-routing silliness going on with the filters, so they can be run in series while modulating their signals as well as modulating filter feedback signals through the use of the Oakley Dual VCA. There's some other CV routing going on with this panel as well. More to come!

Image

EXTERNAL SYNTH USED AS A MODULATION SOURCE: I traded a used overdrive stompbox that I paid $60.00 bucks to obtain, for a nearly new Volca FM. So far, testing has been very encouraging ... I plug the Volca FM into a Dot Com Q118 Instrument Interface (or a Dot Com Q114++ mixer/distributor) to boost the Volca FM's signal level, then that signal becomes a wonderful modulation source. Think about it... the external synth has all of it's own modulators (the Volca FM also has a sequencer and an arpeggiator) and all of it's own triggers, as well as MIDI clock sync on it's own LFO/Arpeggiator/Sequencer. It's stupid simple to reduce the waveform down to a super smooth and very stable sine wave on the Volca FM. Folks, this works! I can't wait to test it all out on the completed and operating Super Mini Modular when it is fully prepped for combat!

This beasty also replaces what I was going to use for ~FM~ sounds which was a pair of modified Q107 State Variable filters (used as sine wave "oscillators" only) and a Happy Nerding FM Aid module. All of that stuff totals to about $420.00 if I bought used Q107s. And that rig would still need support modules to work (LFOs, EGs, and so on). This silly little Volca FM does all of the sounds I was hoping to squeeze from the Dual Q107/FM Aid rig ... AND it has 32 preset memory slots!

I'm ultimately going to use two of these Volca FMs ... one as external modulator and one as sound source to be mixed (or not) with the VCOs of the modular for various additional tones (sortof like how a Hammond uses a percussion generator to add to the drawbar tones). I'll have no more than $180.00 in the pair of them.

Image

Ok .. that's us for this entry.

8-) Later Skaters!!!
Last edited by Rex Coil 7 on Wed Oct 17, 2018 10:21 am, edited 1 time in total.
Never Quit, Die Falling Forward
5U PROJECT - (skip pages 4 through 6, boring junk) ... https://www.muffwiggler.com/forum/viewt ... highlight=

User avatar
ranix
Super Deluxe Wiggler
Posts: 1760
Joined: Wed Jul 13, 2016 9:39 pm

Post by ranix » Mon Jul 09, 2018 10:14 pm

cool dude. The Volca will probably be more capable than 1 fm aid and 2 q107. 2 fm aid and 2 q107 is pretty hard core though.

Check out the Dexed program for making patches on the Volca, I think it's compatible if you send sysex. There are hundreds of hours of fun to be had with just a couple oscillators once you have all the controls available right in front of you.

The thing that happened to me was I got tired of not having real knobs for the envelopes and volume levels. I tried setting up a BCR2000 but the interface was still too clunky to do what I really wanted. It took a couple years to finish playing with it though.

Panason
is banned
is banned
Posts: 3324
Joined: Sat Feb 03, 2018 7:56 am

Post by Panason » Tue Jul 17, 2018 3:44 pm

Holy shit.

Image

Are you getting paid for this!!? :yay:

Post Reply

Return to “5U Format Modules”