MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Sick Buchla 216 (Solved. incl: calibration proc.)
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author Sick Buchla 216 (Solved. incl: calibration proc.)
Peake
Have finished a build and the overall output voltages which were working have dropped to a max of about 3V or so, and one 3906 (at Q53 position etc.) starts to become ridiculously hot on the selected key; greater pressure appears to make it worse and effects the same transistor position in neighboring key/pad component groups. (Edit: Could be the Gate circuit causing it, other possibility.) Voltage rails check out fine at the 072s and I'm using the DC converter for +24V; it's output reads +23.x so all's good there. The three independent key/pads were working fine. The last of the 1-8 key/pads didn't select or trim,. Replacing the 3904 did nothing. Removing the internal 24V connection did nothing.

I've checked the parts for the 1-8 key/pad sections as well as for the entire top section of the PCB and confirmed all transistors, diodes and resistors.

Thing is I'm using a variation of the 2N3904, a 3904-U47 (or -047 or D47, can't quite tell) and I'm not sure if these are 100% compatible. I've used Fairchild J202.

Still haven't memorized what the switch positions and functions are or if there's some unintended interaction occurring. Looking at the parts side of the PCB with the switches at the top, the current settings are:

L, Up, R, R; then the pairs for the left group of three, they should all be the same pair to pair so I'll again check their function: RR, RR, LR

The schematic shows +24 is used in the Pressure circuit but not the keys.

Thanks to apetechnology for the hint to use multi-turn trimpots instead of single!!!
Don T
I haven't gotten my 216 up and running yet, no +24V supply, since the DC/DC convertor I used didn't work. Going to build a linear supply with +24V, +/-15V "someday".

But I'll try giving the schematics a look.
Peake
Thanks. I tell people it's as easy as buying a 1A or less 24V wall wart and wiring it to ground and pin 10 of the Distro...

I've lifted and confirmed R88, the one I couldn't see in the bare PCB photo.
On the schemo it's 100R so this isn't likely to be a 1:1 per the schemo...

https://rubidium.dyndns.org/~magnus/synths/companies/buchla/Buchla_216 0_1A_200.jpg
https://rubidium.dyndns.org/~magnus/synths/companies/buchla/Buchla_216 0_1B_200.jpg

Definite additions after the schemo; there are a pair of 1N5240 Zeners off of the oscillator output.

Edit: The 3906 in a key section begins overheating after the key is selected and held. If I just power it on random 3906s in that position (Q53 etc.) immediately begin to overheat. Just set the miniswitches to their opposite positions and no change.
APETECHNOLOGY
sorry to hear you are having problems!
i dug threw old mail and found the switch settings.
i sold mine, dumb.. so i can't make measurements for you.

sw10 select 10-15 V cv range row B
sw 9 select 10-15 V cv range row A
sw 8 select 10-15 V puls row 4-11
sw 7 select 10-15 V finger pressure
SW2, 4, 6: select between 10V and 15V for CVout max sensor 1, 2, 3
SW1, 3, 5: select between 10V and 15V for PULSEout max sensor 1, 2, 3

hope you find out whats the problem is, good luck!
Peake
Thank you and thank you!
Peake
-Undid the int/ext jumper for the 24V converter and no change.

-Have a pretty clean sine at pin 1 of IC1. Scope is set to 50 microseconds.

-Have that same sine at the cathode of D28 (sch: D1) on several pads, equal amounts, but reduced in level from IC1 pin 1

-The transistors which are overheating are Q5 on the schematic, Q53 etc. on the PCB silkscreen.

-The base of Q54 reads +14V then drops to +12V when touched, for the duration of the contact, returning to +14V when released.

I'm drawing too much current somewhere...?
Don T
Peake wrote:
-Undid the int/ext jumper for the 24V converter and no change.

-Have a pretty clean sine at pin 1 of IC1. Scope is set to 50 microseconds.

-Have that same sine at the cathode of D28 (sch: D1) on several pads, equal amounts, but reduced in level from IC1 pin 1

-The transistors which are overheating are Q5 on the schematic, Q53 etc. on the PCB silkscreen.

-The base of Q54 reads +14V then drops to +12V when touched, for the duration of the contact, returning to +14V when released.

I'm drawing too much current somewhere...?


Quick hunch: That particular touchpad isn't shorted, is it?
Peake
The 1-3 independent keys worked fine and calibrated well. Keys 1-8 showed, after doing the +0.2V cal at the +V side of the 1uF SMD cap for all keys, that key 8 didn't have the cal range available and didn't select when played. I don't think it's shorted; at least the keypad itself has no visible cuts or marks like that.

I'm fine leaving the non-responsive key 8 for after finding whatever is causing the transistors to overheat in all of them...I first noticed it on key 8. This is all after seeing that 1-7 would calibrate and it -was- producing 12V output, which is now 2V or so max...something must have shorted or burned.

If you meant pad 7 from which I posted the reading the transistor base, I was told it should read 15V then drop to 14V or so.

The (schematic) Q5 of every key will overheat. I don't know if I've adequately described that this transistor on every key overheats when the key is selected, sometimes its neighbors as well.

Thanks Don. I appreciate any and all help.
Peake
If all pots are set to above minimum (ground), nothing heats up. Aha...

Edit: Diode installed backwards at pot 2 row B. Even though I'd gone over everything and stared at it. More than once.

If you do a build and are seeing this same overheating issue, set all pot values to half then watching the voltage output of each row, check for a pot which interacts with the voltage set by others. That has a reversed diode and you're shunting everything to ground.
Don T
Peake wrote:
If all pots are set to above minimum (ground), nothing heats up. Aha...

Edit: Diode installed backwards at pot 2 row B. Even though I'd gone over everything and stared at it. More than once.

If you do a build and are seeing this same overheating issue, set all pot values to half then watching the voltage output of each row, check for a pot which interacts with the voltage set by others. That has a reversed diode and you're shunting everything to ground.


Excellent detective work!
APETECHNOLOGY
great news!
w00t
Peake
Thank you both. Can't believe I didn't catch the reversed diode despite having more than triple-checked polarities.

Module is working except Pressure, set to +15V range, idles at +2.1V and maxes at about +11V.
Peake
In the other forum, someone had indicated successful calibration in trimming each key's voltage to +0.2V, read at the left side of the 1uF SMD caps. I've found the Pressure voltage has an offset tied in with the trimmer settings of the 1-8 row of keys. I've found a setting (+0.384V) for each key which results in an overall Pressure idle offset of around 5mV. However, the maximum doesn't reach +15V, only goes up to a bit over 11V with maximum finger contact on a single key with SW7 set to a +15V output range.

Note that any currently selected key/pad will read over +1V at the cap; adjustments must be made to deselected keys.

Paralleled a 47K with the 47K at R97 to gain a max Pressure output level over 15V when an entire finger is applied to a key. This raised the Pressure offset a bit.
sanders
I’m just finishing up my first 216 build. I was pretty surprised when I finished assembling and found that none of the 4-11 section was doing anything right, and I couldn’t trim down the voltage (at the 1uf cap) lower than about 6v on any of the 4-11 stages. Stages 1-3 work great.

Sensor 6 alone wouldn’t trim down past 10v or so, and was the only stage where front panel pots did anything. Holy moly.
After and checking and re-checking component values and continuity everywhere and not finding anything, I took a cue from one of the posts here that mentioned some 2n3904s didn’t work well in this circuit.

This was the issue. I replaced the 3904 on stage 6 with a socket, put a new 3904 in, and the entire circuit came to life. I’ve now got most all of the stages trimmed to 0.2v, except 2 or 3 that won’t go lower than about .8v. I think I have to replace the 3904’s on these stages as well. The stages with .8v work and respond, but their range and response is not as good as the others.

I’m taking all of the stage CV from the “B” row, pulses, and pressure (actually quite good pressure range here). But my “A” row is stuck very high, and no A pots respond. I have a feeling that this again is going to come down to one off-spec component in the A voltage chain.

Last thing, I only measure about 17.5v from the 24v regulator. It’s a brand new part, just in from Mouser; the same one linked in the BOM.

I’ve only just started focusing on these last issues, and I’m hoping maybe someone here might have some further insight. I’ll post again when I find the solutions.
Lento_Zoom
I'm gonna be completing my build here this weekend hopefully. Are there any specific 3904s that are known good or bad?
sanders
I ordered mine from Mouser to be on the safe side, even though I already had some. But there were still several that didn’t work well (as described above).

I’d highly recommend using transistor sockets (I use edge connectors) for the per stage 3904’s, that way you can sub out new ones for any that don’t initially work. I’m sure there’s a way to select for “good” ones, but I don’t have any idea what characteristic you’d be measuring for.
Lento_Zoom
Thanks for the tip! I'll go that route to avoid any headaches down the road.
sanders
Lento_Zoom wrote:
Thanks for the tip! I'll go that route to avoid any headaches down the road.


Yeah, these 200r PCBs are really nice, high quality boards, but the plated thru-holes are quite narrow and I find it especially difficult to cleanly remove and replace parts. It’s gonna be a serious pain replacing the other 3904s that need replacing.

So, I would highly recommend first reading through all the posts for each module before you start it(I should have done this myself for the 216) and then using sockets for any and all parts that may potentially need substitutes.
Lento_Zoom
Yeah. Not nearly as re-workable as the 208 boards. I finished my build this weekend and it is almost there. I clear coat sprayed my touch plate before final assembly and wound up having problems with the pulse out on the keyboard part. My pressure range was also really small, like 3V. Anyway, turns out I did wayyyyyy too much clear coat. Was able to remove it all with acetone, and immediately found the pulse to be more reliable and responsive. This also made my Pressure range acceptable and made the decay envelope outs louder. Now my only issue is that there is couple volt offset on the Pressure and the pressure response is too sensitive. Right now I'm planning on re-calibrating and doing a sensible single coat of clear enamel. Thank you all for the discussion here, there isn't much info on the 216 build out there and y'all's experience has helped me greatly.
sanders
Lento_Zoom wrote:
turns out I did wayyyyyy too much clear coat. Was able to remove it all with acetone, and immediately found the pulse to be more reliable and responsive.


Interesting. I made about 3 or 4 light spray passes over my 216 touch plate, let it dry an hour, and made a couple more passes. It works well, but there’s no visual indication that it’s been coated. I’ve seen a few examples of really glossy touch plates and rather liked the way the gloss looked; So I’ve been planning to do a couple more coats; but maybe I’m better off leaving it as, staying on the safe side.

I also scratched up a couple plates while trying to figure out these electronic issues. : (
It’s so hard to work with after it’s assembled and all the wires are soldered in.

On the upside, I solved the rest of my issues— the constant A voltage and low 24v supply line were both due to a backwards diode. I was so happy that’s all it was. It’s all good now.

I believe if one doesn’t make any assembly errors, the only tricky part of this build is the 2n3904 selection; but it’s an inexpensive part, and only the odd 3904 seems to give a problem, so even that’s not a big deal.

I’m really happy with this controller.
Lento_Zoom
I'm glad to hear that you're up and running! Seems like I spoke too soon. I'm having a problem with gates holding high. I've spent most of the day messing with the trimmers and trying different techniques. Seems like everytime I'm satisfied and screw the panel into the boat, one or more of the pads will hold high after touching it.

Does anyone have a clue what this can be or what I can do to stop it?

I've swapped around 3904s, but that doesn't seem to make any difference.
Lento_Zoom
Took another look at it this morning and was still having trouble. It seemed like it was getting worse over time. Was going nuts and was about to shorten the sensor wires. Then I got the bright idea to try cleaning the touch plate. It wasn’t visibly grimy, but after a wipe down with alcohol, the gates stopped holding! Ran it through a series of quick taps, long taps and slides and it held up. So I guess the lesson is that if you are a greasy fingered individual like myself, keep your sensors clean!
christie
On the right hand section, are the row a and row b output voltages supposed to hold when you release? I'm testing this as I type and on some keys the voltage holds and on others it resets upon release to whatever voltage level is set by the right most knob in that row.

Pulses work fine on it, those don't hold when I release, and I get a nice range of pressure from each pad.

Is this miscalibration as well?
Lento_Zoom
Yeah. The eight pads should latch on the last touched one. It’s just a matter of adjusting the trim pots. I ran into a similar issue.
christie
Thanks!

Totally worked. Very small sweet spot in the trimmer where it latches and hold the voltage, but pulse doesn't stay high. Pressure range still seems pretty good too.
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group