ARP 1601 clone Build Thread

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, lisa, Joe.

Post Reply
User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Sun Feb 02, 2020 5:01 pm

elmegil wrote:
Sun Feb 02, 2020 4:31 pm
Width wise they appear to be 8.5mm side to side; there is no sign of problem with that dimension when inserting the switches. So I would guess the resolution of my ruler/eyes is not sufficient to resolve correctly, and that they likely are the right dimension that way.
That’s good. Can you confirm that the 4PDT and DPDT switches fit correctly?

I’ll try to make contact with Jammie to get the spacing sorted for Synthcube going forward, unless they have the ability to change it themselves. Jammie uses DipTrace which I don’t have so I cannot edit the files.

As mentioned previously I can recreate a Rev 5 board from my Rev 4 files should that be necessary, at which point I can also go for a much smaller and cheaper DC-DC converter. Originally I went for the same one as the TTSH as I had no idea of the current load required, which is well below 100mA on each rail.

User avatar
elmegil
Super Deluxe Wiggler
Posts: 2969
Joined: Sun Apr 29, 2012 2:57 pm
Location: Chicago

Re: ARP 1601 clone Build Thread

Post by elmegil » Sun Feb 02, 2020 6:11 pm

Yes the other switches absolutely fit correctly.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Mon Feb 03, 2020 3:35 am

elmegil wrote:
Sun Feb 02, 2020 6:11 pm
Yes the other switches absolutely fit correctly.
Thanks for confirming that.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Mon Feb 03, 2020 12:08 pm

OK, I've created a Rev 6 PCB design from my Rev 4 files utilising the CW Industries PCB-mount switches (as per Rev 5 but without the pin spacing errors) and a much smaller and cheaper Meanwell DCWN06A-15 DC-DC converter, which at 200mA max output per rail is still well above the needs of the PCB. I have a Meanwell DCWN06A-15 DC-DC converter which I'll check is perfectly OK with my original prototype just to be sure there are no issues.

The board is identical to Rev 4 and 5 in all other respects, i.e. the MC14528 and LM3046 are surface mount chips.

Once I've checked out the new DC-DC converter I'll forward the PCB files to Synthcube as they may wish to use them in future PCB production runs.

KSS
Super Deluxe Wiggler
Posts: 3426
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 1601 clone Build Thread

Post by KSS » Mon Feb 03, 2020 8:05 pm

OOPs! Ignore.Wrong thread. Meant for the 3620 PCB and panel.
Last edited by KSS on Tue Feb 04, 2020 7:46 pm, edited 1 time in total.

retokid
Learning to Wiggle
Posts: 6
Joined: Tue Jan 07, 2020 4:10 pm

Re: ARP 1601 clone Build Thread

Post by retokid » Tue Feb 04, 2020 4:44 pm

Puzzled by the idea that SynthCube sells revised boards that apparently have not been tested at all...

SynthCube any comments ?

User avatar
elmegil
Super Deluxe Wiggler
Posts: 2969
Joined: Sun Apr 29, 2012 2:57 pm
Location: Chicago

Re: ARP 1601 clone Build Thread

Post by elmegil » Wed Feb 05, 2020 1:33 am

Kipling wrote:
Sun Feb 02, 2020 9:15 am
I would be more inclined to carefully file down the outer pins until it fits snugly in the board, rather than bending them.
That seems to work pretty well, only takes a little bit of filing; admittedly there are a significant number of switches, but I would expect I could knock out enough for one sequencer in half an hour or less.
IMG_2354.JPG
IMG_2355.JPG
You do not have the required permissions to view the files attached to this post.

User avatar
synthcube
you will be assimilated
Posts: 2133
Joined: Wed Dec 21, 2011 6:42 pm
Location: Boston
Contact:

Re: ARP 1601 clone Build Thread

Post by synthcube » Wed Feb 05, 2020 4:59 pm

Thanks all for the comments feedback and suggestions re: switch fit

We are aware of the issue and of course looking for the best resolution for future runs of PCBs. The issue only recently came to our attention-- so we are doing some homework to figure out if its only this latest run of the Logan PCB files that caused it, or if it may have existed in prior runs with the V5 files. It's an odd sort of error, and we have reached out also to Jim and Jammie for input.

Kipling-- we'd be happy to work with you for an updated PCB file, as we are close to ordering our next run. Since Jammie has been out of touch lately we have not yet been able to fully sort it out. Whether or not we create v6 from updating our current v5 files from jammie, or utilize the Kipling version, future pcb runs will have addressed this...
facebook: https://www.facebook.com/2SynthCube

synthcube webstore www.synthcube.com/cart
music from outer space: www.musicfromouterspace.com
blacet research: www.blacet.com
MOTM DIY Analog Synthesizers: www.motmsynthesizers.com
music from outer space euro smt: www.mfoseuro.com

User avatar
r0main
Learning to Wiggle
Posts: 47
Joined: Sat Jul 26, 2014 3:13 pm

Re: ARP 1601 clone Build Thread

Post by r0main » Fri Feb 07, 2020 8:45 am

Question regarding the companion pcb (Divider): Do you wire the Reset pad to the main sequencer pcb? If so, from which place in the sequencer pcb?

User avatar
elmegil
Super Deluxe Wiggler
Posts: 2969
Joined: Sun Apr 29, 2012 2:57 pm
Location: Chicago

Re: ARP 1601 clone Build Thread

Post by elmegil » Fri Feb 07, 2020 8:54 am

Here's how I did it.
1601-divider-wiring.jpg
Didn't make sense to me to wire it to the 3 pin power header another half of the way across the board, but only for 2 pins, and still having to grab gate bus 2. (using GB2 because that's how the panel is labelled)
You do not have the required permissions to view the files attached to this post.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Fri Feb 07, 2020 11:25 am

synthcube wrote:
Wed Feb 05, 2020 4:59 pm
Thanks all for the comments feedback and suggestions re: switch fit

We are aware of the issue and of course looking for the best resolution for future runs of PCBs. The issue only recently came to our attention-- so we are doing some homework to figure out if its only this latest run of the Logan PCB files that caused it, or if it may have existed in prior runs with the V5 files. It's an odd sort of error, and we have reached out also to Jim and Jammie for input.

Kipling-- we'd be happy to work with you for an updated PCB file, as we are close to ordering our next run. Since Jammie has been out of touch lately we have not yet been able to fully sort it out. Whether or not we create v6 from updating our current v5 files from jammie, or utilize the Kipling version, future pcb runs will have addressed this...
Thanks. I’ll try to check out the cheaper Meanwell DC-DC converter in the next few days and if all is OK I’ll send the files over to you in a PM. From elmegil’s photo above it looks like the descriptive text for the trimmers has been missed off in Jammie’s V5 so I’ll make sure all text is reinstated too.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Fri Feb 07, 2020 11:43 am

elmegil wrote:
Fri Feb 07, 2020 8:54 am
Here's how I did it.

Didn't make sense to me to wire it to the 3 pin power header another half of the way across the board, but only for 2 pins, and still having to grab gate bus 2. (using GB2 because that's how the panel is labelled)
I presume this is for the divider PCB, picking up the Gate Bus 2 signal along with +15V and Ground. If so I can easily add a 3-pin header for that on my Rev 6 PCB to make life a little simpler.

User avatar
elmegil
Super Deluxe Wiggler
Posts: 2969
Joined: Sun Apr 29, 2012 2:57 pm
Location: Chicago

Re: ARP 1601 clone Build Thread

Post by elmegil » Fri Feb 07, 2020 11:44 am

Kipling wrote:
Fri Feb 07, 2020 11:43 am
elmegil wrote:
Fri Feb 07, 2020 8:54 am
Here's how I did it.

Didn't make sense to me to wire it to the 3 pin power header another half of the way across the board, but only for 2 pins, and still having to grab gate bus 2. (using GB2 because that's how the panel is labelled)
I presume this is for the divider PCB, picking up the Gate Bus 2 signal along with +15V and Ground. If so I can easily add a 3-pin header for that on my Rev 6 PCB to make life a little simpler.
Yes, that's exactly what it is for. Thanks!

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Fri Feb 07, 2020 12:11 pm

Kipling wrote:
Fri Feb 07, 2020 11:43 am
elmegil wrote:
Fri Feb 07, 2020 8:54 am
Here's how I did it.

Didn't make sense to me to wire it to the 3 pin power header another half of the way across the board, but only for 2 pins, and still having to grab gate bus 2. (using GB2 because that's how the panel is labelled)
I presume this is for the divider PCB, picking up the Gate Bus 2 signal along with +15V and Ground. If so I can easily add a 3-pin header for that on my Rev 6 PCB to make life a little simpler.
Added!
Capture.JPG
You do not have the required permissions to view the files attached to this post.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Sun Feb 09, 2020 3:06 pm

Good news. The Meanwell DCWN06A-15 DC-DC converter is absolutely fine powering the 1601 and has more than 120mA headroom on the +15V rail (the 1601 runs at 64mA in 16/1 mode and 77mA in 8/2 mode) and more than 160mA headroom on the -15V rail, and gets no more than slightly warm (measured at around 34C after an hour’s use).

So I will finalise the Rev 6 design in the next day or so and send it over to Synthcube.

The changes from Rev 5 are:

1. The CW Industries switches PCB pin spacing has been corrected to manufacturer’s specification. These are 7.92mm between each position, and 8.64mm between each row, except for the 4-pole switch which is 8.64mm between the two inner rows and 5.59mm between inner and outer rows. Note: I do not have any of their switches to physically measure them. Perhaps someone with said switches and a metric dial caliper can confirm this for me.

2. The original large and expensive DC-DC converter has been changed for a much smaller and cheaper Meanwell DCWN06A-15. Note: requires change to the BOM.

3. A 3-pin header has been added for the divider add-on providing +15V, Ground (0V) and Gate Bus 2. Note: requires change to the BOM.

4. Descriptive text has been reinstated for all the trimmers and anywhere else it may be missing on the Rev 5.

5. Reference to JMSL has been deleted as this is entirely my design.

EDIT: I am contemplating changing the single turn trimmers for 10-turn for finer setting accuracy, and also adding trimmers for the gate bus outputs to make it more eurorack friendly - the standard gate voltage is around 14V which may be a bit on the high side for eurorack. Your thoughts please.

KSS
Super Deluxe Wiggler
Posts: 3426
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 1601 clone Build Thread

Post by KSS » Sun Feb 09, 2020 8:49 pm

Kipling wrote:
Sun Feb 09, 2020 3:06 pm
I am contemplating changing the single turn trimmers for 10-turn for finer setting accuracy,

Code: Select all

0 0
000 000 000 0 0
 0   0       0
If this aligns correctly the first -left- pad pattern accomodates both the flat single turn bourns style trim and also its upright 10T version. Keeping the trim adjust location in the same spot for panel to PCB alignment. There are actually a few different bourns style trims which will fit in this pad layout. The second pad patern is not as universal, but still allows more than one trim P/N to be used. The third and fourth are IMO abominable pad alyouts as they restrict trim choice unnecessarily. (In the first and second pad layouts the vertical pairs of pins are joined.)
and also adding trimmers for the gate bus outputs to make it more eurorack friendly - the standard gate voltage is around 14V which may be a bit on the high side for eurorack. Your thoughts please.
Since these are simple two resistor voltage dividers, trimpots seem huge overkill. Just calculate a single or pair to give wht you desire, and put that as an option on the bom. Adding 3 resistors to a synthcube kit to accomodate a lower GBus out option is IMO far better answer than using trims here. For long-term function and use also better to have fewer than more trims.

Edit: even better, just add the 2nd resistor location -per output- and a two pin 'jumper' HDR. Then one can use a shorting block or external switch to activate or allow either of two different levels. A small section of the build manual can describe how to choose a resistor for other than those 'provided'. And of course the two pin HDR could also interface to a pot, for add-on adjustability. Cheap and easy. Simple to ignore for those who don't need it. Simple to apply for those who do. /edit

User avatar
FactoryDefault
Common Wiggler
Posts: 164
Joined: Tue Mar 29, 2016 12:37 pm

Re: ARP 1601 clone Build Thread

Post by FactoryDefault » Tue Feb 11, 2020 11:07 pm

elmegil wrote:
Wed Feb 05, 2020 1:33 am
Kipling wrote:
Sun Feb 02, 2020 9:15 am
I would be more inclined to carefully file down the outer pins until it fits snugly in the board, rather than bending them.
I twisted the 4 outer pins on my switches by 90 degrees. It worked great.

User avatar
elmegil
Super Deluxe Wiggler
Posts: 2969
Joined: Sun Apr 29, 2012 2:57 pm
Location: Chicago

Re: ARP 1601 clone Build Thread

Post by elmegil » Tue Feb 11, 2020 11:31 pm

FactoryDefault wrote:
Tue Feb 11, 2020 11:07 pm
I twisted the 4 outer pins on my switches by 90 degrees. It worked great.
I'm having a hard time picturing what you mean. Photos?

Filing is working, but is time consuming.

User avatar
Jim the Oldbie
Ultra Wiggler
Posts: 887
Joined: Mon Jun 10, 2019 12:16 am
Location: Midwest USA

Re: ARP 1601 clone Build Thread

Post by Jim the Oldbie » Tue Feb 11, 2020 11:45 pm

FactoryDefault wrote:
Tue Feb 11, 2020 11:07 pm
I twisted the 4 outer pins on my switches by 90 degrees. It worked great.
There ya go!

This idea occurred to me as well, but I was unable to try it out since I haven't even received a ship notice on my 1601 yet. :waah:
Glad to hear it works.

retokid
Learning to Wiggle
Posts: 6
Joined: Tue Jan 07, 2020 4:10 pm

Re: ARP 1601 clone Build Thread

Post by retokid » Wed Feb 12, 2020 4:47 am

So I will finalise the Rev 6 design in the next day or so and send it over to Synthcube.
Kipling, is it on purpose that I can find no PDF of the PCB (as I found easily for the 1601 sequencer for example), that would be of great help in building and fault finding.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Wed Feb 12, 2020 6:14 am

retokid wrote:
Wed Feb 12, 2020 4:47 am
Kipling, is it on purpose that I can find no PDF of the PCB (as I found easily for the 1601 sequencer for example), that would be of great help in building and fault finding.
There are PDF's of the 1601 layout on post #1 on page 1. Rev 5 boards are essentially identical in appearance to Rev 4 (with the SMD chips).
Capture2.JPG

Or are you referring to the 3620 keyboard? If so I will add them to the relevant thread.
You do not have the required permissions to view the files attached to this post.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Wed Feb 12, 2020 7:06 am

KSS wrote:
Sun Feb 09, 2020 8:49 pm
Kipling wrote:
Sun Feb 09, 2020 3:06 pm
I am contemplating changing the single turn trimmers for 10-turn for finer setting accuracy,

Code: Select all

0 0
000 000 000 0 0
 0   0       0
If this aligns correctly the first -left- pad pattern accomodates both the flat single turn bourns style trim and also its upright 10T version. Keeping the trim adjust location in the same spot for panel to PCB alignment. There are actually a few different bourns style trims which will fit in this pad layout. The second pad patern is not as universal, but still allows more than one trim P/N to be used. The third and fourth are IMO abominable pad alyouts as they restrict trim choice unnecessarily. (In the first and second pad layouts the vertical pairs of pins are joined.)
and also adding trimmers for the gate bus outputs to make it more eurorack friendly - the standard gate voltage is around 14V which may be a bit on the high side for eurorack. Your thoughts please.
Since these are simple two resistor voltage dividers, trimpots seem huge overkill. Just calculate a single or pair to give wht you desire, and put that as an option on the bom. Adding 3 resistors to a synthcube kit to accomodate a lower GBus out option is IMO far better answer than using trims here. For long-term function and use also better to have fewer than more trims.

Edit: even better, just add the 2nd resistor location -per output- and a two pin 'jumper' HDR. Then one can use a shorting block or external switch to activate or allow either of two different levels. A small section of the build manual can describe how to choose a resistor for other than those 'provided'. And of course the two pin HDR could also interface to a pot, for add-on adjustability. Cheap and easy. Simple to ignore for those who don't need it. Simple to apply for those who do. /edit
Thanks for your feedback, which I have considered carefully.

I have gone for the upright multi-turn Bourns trimmers type 3296Y (the fourth pad layout in your list) to replace the eight existing trimmers. Four-off 652-3296Y-1-102LF (1K) and four-off 652-3296Y-1-104LF (100K) are required and replace all the existing 3362P types. These are in fact 25-turn as Bourns do not offer 10-turn in this style. The pad spacing is exactly the same as the 3362P single-turn trimmer but for neatness I've updated the screen-print to suit. I am sure there are other manufacturers offering similar products and footprints though I have not investigated these.

Regarding the variable gate and clock pulse voltages, I have stuck to my intention and added six upright side-adjust single-turn trimmers type 3362M (six-off, all 10K, Mouser part# 652-3362M-1-103LF) with in-line pads as there was limited space to add extra resistors and header pins for voltage selection or to have the offset pin spacing without too much track and component moving. These replace 10K resistors R139, R141, R148, R150, R156 and R159. It will be a simple case of leaving all six fully clockwise for normal operation (14V outputs) or setting each one accordingly to achieve 5V or 12V as necessary. Just measure the output on Gate Bus 1 with the sequencer static on position 1 with its slide switch set to Gate Bus 1 and adjust R139 (GB1 ADJ) to your required voltage, then set the other fiver trimmers to the same position.

An example of how these all look on the board is shown below:

Capture5.JPG
Capture4.JPG

Hope this is OK for everybody.
You do not have the required permissions to view the files attached to this post.

KSS
Super Deluxe Wiggler
Posts: 3426
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 1601 clone Build Thread

Post by KSS » Wed Feb 12, 2020 7:14 pm

This does not make any sense. At least add a fourth pad in the middle, and based on the layouts you've shown every single position can accomodate the 6 hole UNIVERSAL layout shared. Three hole layouts are the worst, as they severely limit the available options. Even adding only the fourth center hole expands the potential candidates greatly.

Why does this matter? Because by adding pads to the layouts you're insulating against shortages and part changes for both present and future. We've all seen popular parts undergo short supply due to sudden popularity and we've also seen even long time available parts simply disappear. And where's the rub?? There is almost NO down side to implementing these additional holes. Perhaps a few cents more per PCB, and literally only a few cents IME. Aside from that, it's ALL good.

Next, do you really think multi-turn pots are necessary for setting gate level? Unless high quality multi-turns are provided they will have more backlash and take a long time to set. With no real gain to offset these two downsides.

And to again re-iterate, the six hole pattern used with bourns upright multi-turns and bourns flat leave the adjusting hole/point in the same position.

Finally, it's generally bad practice to use only two pins of a trim. Connecting the third pad ensures there will always be at least some resistance at that point in the schematic. It can't go open circuit. Without this , it can. And with the ever poorer qualkity and specs of todays trimpots, this open circuit result is to be expected over a trims life. So this becomes a support issue too. Because it will usually go intermittent first.

I appreciate your having taken a look at what I wrote before. Hopefully these further comments will help you to see the value in implementing more pads in each trim position. And connecting the 'loose end' traces.

User avatar
elmegil
Super Deluxe Wiggler
Posts: 2969
Joined: Sun Apr 29, 2012 2:57 pm
Location: Chicago

Re: ARP 1601 clone Build Thread

Post by elmegil » Wed Feb 12, 2020 9:16 pm

I think you're stating your case a little "strongly" perhaps, but your logic is good. And I would agree, I didn't look at the layout closely before, it is standard/best practice to join the two pins on any pot wired as a rheostat/variable resistance rather than a voltage divider.

User avatar
Kipling
Ultra Wiggler
Posts: 970
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 1601 clone Build Thread

Post by Kipling » Thu Feb 13, 2020 3:53 pm

KSS wrote:
Wed Feb 12, 2020 7:14 pm
This does not make any sense. At least add a fourth pad in the middle, and based on the layouts you've shown every single position can accomodate the 6 hole UNIVERSAL layout shared. Three hole layouts are the worst, as they severely limit the available options. Even adding only the fourth center hole expands the potential candidates greatly.
I have added a fourth pad to each type to make them more suitable for each person's preference.
Why does this matter? Because by adding pads to the layouts you're insulating against shortages and part changes for both present and future. We've all seen popular parts undergo short supply due to sudden popularity and we've also seen even long time available parts simply disappear. And where's the rub?? There is almost NO down side to implementing these additional holes. Perhaps a few cents more per PCB, and literally only a few cents IME. Aside from that, it's ALL good.
As above.
Next, do you really think multi-turn pots are necessary for setting gate level? Unless high quality multi-turns are provided they will have more backlash and take a long time to set. With no real gain to offset these two downsides.
I think you've mis-read or misinterpreted my post - "I have stuck to my intention and added six upright side-adjust single-turn trimmers type 3362M"
And to again re-iterate, the six hole pattern used with bourns upright multi-turns and bourns flat leave the adjusting hole/point in the same position.
I trust 4-hole is sufficiently universal. To add another row would mean stretching the part outline to avoid screen-printing over some of the pads. It's already quite crowded in many parts of the board.
Finally, it's generally bad practice to use only two pins of a trim. Connecting the third pad ensures there will always be at least some resistance at that point in the schematic. It can't go open circuit. Without this , it can. And with the ever poorer qualkity and specs of todays trimpots, this open circuit result is to be expected over a trims life. So this becomes a support issue too. Because it will usually go intermittent first.
Any pad that appears not to be connected is either connected to ground or +15V (just like every other component that appear not to show tracks on all pads), it's just that it doesn't show the two inner planes on the four-layer board in the design view - check out Q31, Q36, SK11 and C7 in the second shot below. Some ICs obviously have unused pins but everything else is connected to something.
I appreciate your having taken a look at what I wrote before. Hopefully these further comments will help you to see the value in implementing more pads in each trim position. And connecting the 'loose end' traces.
Thank you for your critical review. I trust all the above is now satisfactory. :tu:

Here are updated views of the two types:

Capture7.JPG
You do not have the required permissions to view the files attached to this post.

Post Reply

Return to “Music Tech DIY”