ARP 3620 clone project

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

Moderators: Kent, luketeaford, Joe.

Post Reply
User avatar
kogz23
Common Wiggler
Posts: 80
Joined: Sun Feb 27, 2011 12:37 pm
Location: OAKLAND

Re: ARP 3620 clone project

Post by kogz23 » Thu Apr 16, 2020 4:48 pm

Ah, ok. Thanks!

User avatar
kogz23
Common Wiggler
Posts: 80
Joined: Sun Feb 27, 2011 12:37 pm
Location: OAKLAND

Re: ARP 3620 clone project

Post by kogz23 » Thu Apr 16, 2020 5:25 pm

I have a 3604 and am planning on replacing the its circuitry with this 3620 clone. So it looks like I will have to do some sanding and make some slight adjustments to the size of the frontpanel. Kipling, is there any chance you could post the pcb layout for the 3 bus to 2 bus converter?
Thanks.

tempco01
Learning to Wiggle
Posts: 12
Joined: Thu Jan 29, 2015 11:34 am
Location: Midlands, UK

Re: ARP 3620 clone project

Post by tempco01 » Fri Apr 17, 2020 7:03 am

Hi all,
I’m having a bit of a clearout and I have an unused 3620 pcb and front panel for sale. It was from Kiplings second run from a few years ago. It was bought for a second 3620 but I never got round to building it. It also comes with two 470nF polycarb capacitors.
Mint condition, still in original packaging!
Please PM me if your interested.

I’m also selling my TTSH rev 1 and 3620 replica keyboard if anybody’s interested.

I’m UK based.

User avatar
Kipling
Super Deluxe Wiggler
Posts: 1001
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 3620 clone project

Post by Kipling » Mon Apr 20, 2020 8:11 am

kogz23 wrote:
Thu Apr 16, 2020 5:25 pm
I have a 3604 and am planning on replacing the its circuitry with this 3620 clone. So it looks like I will have to do some sanding and make some slight adjustments to the size of the frontpanel. Kipling, is there any chance you could post the pcb layout for the 3 bus to 2 bus converter?
Thanks.
I have actually had a small number of 3/2 bus converter boards made recently so I can post one out to you for a small fee to cover postage.

I also have a Rev3a board design (attached) that fits the 3604-P case as the existing one is slightly too large (though can be safely sanded on all sides to just about fit if you observe where the traces start - see photo below). I've yet to rework the front panel to suit the 3604-P case and Rev 3a PCB, but note that the PCB mounting holes are different from my original design. I hope I've still got the metric measurements from a 3604-P case I was given some time ago.

3604 case  (not mine) with a reduced size 3620 Rev 2 PCB.
3604 case (not mine) with a reduced size 3620 Rev 2 PCB.
Attachments
ARP 3620 Plots V3a.zip
(407.95 KiB) Downloaded 6 times

User avatar
KSS
Super Deluxe Wiggler
Posts: 4800
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 3620 clone project

Post by KSS » Mon Apr 20, 2020 1:08 pm

Kipling wrote:
Mon Apr 20, 2020 8:11 am
I hope I've still got the metric measurements from a 3604-P case I was given some time ago.
The ARP cases were built to imperial dimensions. Can be converted, but sometimes using the original dimensions keeps things simple.

User avatar
Kipling
Super Deluxe Wiggler
Posts: 1001
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 3620 clone project

Post by Kipling » Mon Apr 20, 2020 1:39 pm

Thanks - I've found the dimensions and a front panel design I did to match.

I'm taking the opportunity to update the PCBs to Rev 2b (for my original larger-than-ARP's original size) and 3b (for the ARP 3604-P case size) to use a cheaper and more readily available toggle switch - the NKK M2013TXG13 SPDT with centre-off, which comes with a rocker which can be exchanged for a paddle AT4149B (white) or AT4149A (black) to suit. This will also simplify and hence cheapen the front panel which need not be so thick or have the cutout in the rear face to clear the switch.

I hope to complete this over the next couple of days then depending on demand I can get a batch of 10 of each made and/or post the gerbers for others to do likewise. I still have 10 sets of 470nF polycarbonate caps and the 6-pin panel & line connectors and 10 of the 3/2 bus adapter PCBs for using an original 3604-P keybed.

User avatar
solar1
Common Wiggler
Posts: 63
Joined: Fri Jul 31, 2015 5:14 pm
Location: USA

Re: ARP 3620 clone project

Post by solar1 » Mon Apr 20, 2020 1:56 pm

Hi Kipling, I may be interested in a couple front panels depending on the price if you set up a group buy.

I'd love to get one with white graphics on the grey paint.
Thanks again for your efforts.
Please keep us posted.
BTW, I PM'd the 3604 panel case opening dimensions on March 4, 2017

User avatar
KSS
Super Deluxe Wiggler
Posts: 4800
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 3620 clone project

Post by KSS » Mon Apr 20, 2020 2:17 pm

solar1 wrote:
Mon Apr 20, 2020 1:56 pm
BTW, I PM'd the 3604 panel case opening dimensions on March 4, 2017
ARP panels were pretty sloppy fit to case dimensions. 3620 panel is exactly 6-1/8" wide by 8-5/8" tall. The PCB has identical dimensions. With 4 corner cutouts.
Case opening can be as high as 8.85" Width opening is similarly loose-sloppy. Depends how well the pine ply KBD rear cover was made.

Jimhyak
1-Post Wiggler
Posts: 1
Joined: Sat Oct 19, 2019 4:20 pm
Location: Portland

Re: ARP 3620 clone project

Post by Jimhyak » Tue Apr 21, 2020 7:32 am

Kipling
I’m interested in a set - PCB, face panel, connectors. I have caps.
Thanks

PPE
Learning to Wiggle
Posts: 9
Joined: Wed Apr 22, 2020 9:41 am

Re: ARP 3620 clone project

Post by PPE » Tue Apr 28, 2020 1:14 pm

tempco01 wrote:
Fri Apr 17, 2020 7:03 am
Hi all,
I’m having a bit of a clearout and I have an unused 3620 pcb and front panel for sale. It was from Kiplings second run from a few years ago. It was bought for a second 3620 but I never got round to building it. It also comes with two 470nF polycarb capacitors.
Mint condition, still in original packaging!
Please PM me if your interested.

I’m also selling my TTSH rev 1 and 3620 replica keyboard if anybody’s interested.

I’m UK based.
PM’d

User avatar
kogz23
Common Wiggler
Posts: 80
Joined: Sun Feb 27, 2011 12:37 pm
Location: OAKLAND

Re: ARP 3620 clone project

Post by kogz23 » Tue Apr 28, 2020 11:56 pm

I would be like a pcb a front panel for the 3604 sizing. Also the bus adapter. Thanks Kipling for this!

User avatar
kogz23
Common Wiggler
Posts: 80
Joined: Sun Feb 27, 2011 12:37 pm
Location: OAKLAND

Re: ARP 3620 clone project

Post by kogz23 » Wed Apr 29, 2020 12:06 am

ARP 3620 panel with graphics.fpd
(1.55 MiB) Downloaded 13 times
Here is a FPE file I changed to be the dimension to fit the 3604 case. I changed it to white and Tonus label.

User avatar
Kipling
Super Deluxe Wiggler
Posts: 1001
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 3620 clone project

Post by Kipling » Thu Apr 30, 2020 5:40 am

kogz23 wrote:
Wed Apr 29, 2020 12:06 am
ARP 3620 panel with graphics.fpdHere is a FPE file I changed to be the dimension to fit the 3604 case. I changed it to white and Tonus label.
Thanks for this. I have produced the gerbers for a reduced size PCB to fit the 3604-P case (my original is oversized by a few mm), which has different mounting holes to the front panel. I'll adjust it to suit and post.

I have also taken the opportunity to change the +/- 2 octave paddle switch such that it will no longer need the cutout in the rear of the panel, and the panel can hence be thinner, making the overall cost lower by about $15 per panel.

User avatar
solar1
Common Wiggler
Posts: 63
Joined: Fri Jul 31, 2015 5:14 pm
Location: USA

Re: ARP 3620 clone project

Post by solar1 » Thu Apr 30, 2020 7:38 am

Thank you Kipling! We appreciate your continuing efforts on this important project.

Fernandozarone
Learning to Wiggle
Posts: 19
Joined: Sat Jun 20, 2020 3:28 am

Re: ARP 3620 clone project

Post by Fernandozarone » Sat Jun 20, 2020 3:37 am

Hi all,
I have a problem with my original ARP 3620: when I play in a fast sequence, e.g., 5 adjacent notes, only the first 3-4 play, after which, playing again the fifth one, it works. It is like the kbd CV does not “reach” the last notes when played rapidly. The problem is NOT present when I switch the portamento switch ON, it appears only in the OFF position (in any case, portamento time = 0)! What the problem could be due to?
BTW, yes I thoroughly cleaned both the bus bars and the j-wires after this problem, and checked that CV contacts the bar before gate/trigger
The thing is, the problem arises only when the PORTAMENTO switch is in the OFF position
Please help me...
THX
Fernando fernandozarone at mac dot com

User avatar
KSS
Super Deluxe Wiggler
Posts: 4800
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 3620 clone project

Post by KSS » Sat Jun 20, 2020 4:44 am

Welcome to muffwiggler, Fernandozarone!

I though you already posted about this in a thread and ended up replacing the 3954 dual FET to fix the problem?

I know in that thread someone said the big yellow caps were special and not to replace them. But they *do* go bad, even if they *are* special. They are not made anymore, but can be found still and even replaced with a different type with no change in performance.

User avatar
Kipling
Super Deluxe Wiggler
Posts: 1001
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 3620 clone project

Post by Kipling » Sat Jun 20, 2020 11:55 am

The 0.47uF capacitors should ideally be polycarbonate but as mentioned they are not made anymore (the factory making the polycarbonate film closed down some years ago). They can still be found online if you’re lucky. Search for mkc capacitor - Philips made them amongst others. I have about 20 of them in 22.5mm pitch radial.

Fernandozarone
Learning to Wiggle
Posts: 19
Joined: Sat Jun 20, 2020 3:28 am

Re: ARP 3620 clone project

Post by Fernandozarone » Sat Jun 20, 2020 12:27 pm

Thanks,
yes on AH list I wrote that the problem had been solved replacing the dual FET, but after some time, the day after, the problem is back. I shall try to understand if there is also a "mechanical" problem (faulty contact or so on) but do not believe that the key contacts are to be blamed. I cleaned and aligned them well, and first of all the problem ONLY appears with the Portamento switch in the OFF position, working well in the ON (slider all the way down). And the problem becomes worse after many times I play 5 adjacent notes rapidly, I mean at the beginning the last note played is the 4th out of 5, then it stops at the 3rd, although my last fingers have pressed the two keys ...
could be a capacitor? I tested the two yellow .47 ones and the DVM gives a correct .47 measurement. Does this mean they are ok or could them be faulty as well?
I ordered two of these... https://www.ebay.com/itm/254570758582
Are they good in your valuable opinion?
Thanks for the help, I am sure that in this forum there is the competence for solving the issue!

User avatar
elmegil
Super Deluxe Wiggler
Posts: 3029
Joined: Sun Apr 29, 2012 2:57 pm
Location: Chicago

Re: ARP 3620 clone project

Post by elmegil » Sat Jun 20, 2020 12:49 pm

If the FET fixed it but only for a short time, is there something that is damaging the FET?

User avatar
Kipling
Super Deluxe Wiggler
Posts: 1001
Joined: Tue Dec 10, 2013 3:31 pm
Location: South-East England

Re: ARP 3620 clone project

Post by Kipling » Sat Jun 20, 2020 12:52 pm

Fernandozarone wrote:
Sat Jun 20, 2020 12:27 pm
Thanks,

I ordered two of these... https://www.ebay.com/itm/254570758582
Are they good in your valuable opinion?
Thanks for the help, I am sure that in this forum there is the competence for solving the issue!
They will be correct for the 3620. Whether that fixes your problem remains to be seen of course but fingers crossed 🤞🤞

Fernandozarone
Learning to Wiggle
Posts: 19
Joined: Sat Jun 20, 2020 3:28 am

Re: ARP 3620 clone project

Post by Fernandozarone » Sat Jun 20, 2020 1:09 pm

I'll go on with the test after I go back home and keep you informed about what happens... 🙏🏻

Fernandozarone
Learning to Wiggle
Posts: 19
Joined: Sat Jun 20, 2020 3:28 am

Re: ARP 3620 clone project

Post by Fernandozarone » Sat Jun 20, 2020 3:54 pm

PS From a nerd... ("Unfortunately", I am only a COVID era physician... :-) if I test a .47 capacitor and the DVM confirms the correct capacity, I guess, is it sufficient to say that the cap is ok?

User avatar
KSS
Super Deluxe Wiggler
Posts: 4800
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 3620 clone project

Post by KSS » Sat Jun 20, 2020 4:15 pm

if I test a .47 capacitor and the DVM confirms the correct capacity, I guess, is it sufficient to say that the cap is ok?
It's not fully sufficient, but it's enough to be worth soldering in to test in-circuit. As a physician, I'm sure you understand the concept ;)

The 3620 is known for having these caps go bad. It's also true that the 3954 dual FETs sometimes have problems. One simple thing to remember is that when this was designed, the dual FET was the best-cheapest way to get a high impedance opamp input. So many ARP repairs and clones now attach some kind of special mojo to the dual fet. In a couple places of the ARP universe, that's true, But mostly, it's not.

Something like an LF412, with the FET resistors removed and the FET pads jumped gets you a far better impedance and opamp, while costing orders of magnitude less.

Glad you found the PC caps. But there again, although PC (polycarbonate) truly does have some great properties, using PP (polypropylene) or PS (polystyrene) or ideally PPS (polyphenylsulfone) is not going to hurt one bit. No one, and I mean NO ONE is going to be able to tell if you have all of these mods in a 3620.
Last edited by KSS on Sun Jun 21, 2020 2:21 am, edited 1 time in total.

User avatar
KSS
Super Deluxe Wiggler
Posts: 4800
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 3620 clone project

Post by KSS » Sat Jun 20, 2020 4:17 pm

elmegil wrote:
Sat Jun 20, 2020 12:49 pm
If the FET fixed it but only for a short time, is there something that is damaging the FET?
I'd pay attention to this excellent point.

Have you checked the power rails for spikes, noise and stability?

User avatar
KSS
Super Deluxe Wiggler
Posts: 4800
Joined: Mon Jan 25, 2016 7:28 am

Re: ARP 3620 clone project

Post by KSS » Sat Jun 20, 2020 4:47 pm

I thnk you asked about the 4392 in that other forum thread. Yes, it's the correct part. The original 5459 was updated to the 4392 in a service update.

There are really only a few components in the chain here. The OPA between the CV bus bar and the FET. The fet itself and the FET-OPA combo (or replacement Hi-Z OPA). The capacitor.

Don't forget you have TWO of these CV memory chains in a 3620. We'll assume for now that input OPA of the 2nd voice CV memory is fine, since you didn't say anything about general note problems.

You've mentioned that the problem only happens with the Port. SW ON.
That means there's another FET possibly in play. The one shorting the Port. pot. Q3. When Port. is ON this FET -is supposed to be- fully OFF. If it's leaky or not being gated correctly, it's going to affect what C4, the 1st voice CV memory capacitor 'sees' voltage wise.

Q3 has resistor R20 and the diode attached and the diode anode goes to the Port on/off SW and perhaps notably, some 4xxx logic. The 4011 is another possibly faulty or potentially destroyed component. <--See what I did there? ;)

But you haven't said anything else like the interval latch, or general triggering are having problems. I'd check the full performance of every function to narrow down where the real problem may be.

Start with the power rails. They are notorious anyways in ARPs, and this PCB sits at the end of a long cord and a couple of possible dubious connections to begin with!

The last thing you want is to replace some expensive parts and be back here too soon asking what is wrong again.

Post Reply

Return to “Music Tech DIY”