ARP Explorer - Envelope Sustain Issue (Always on)

[JamieOxford]

Hello,

I'm hoping to get some help with an Arp Explorer I'm restoring.

The envelope isn't fully closing due to a rogue voltage, but I'm struggling to find the source of the issue as the envelope circuit looks like one large feedback loop!
Turning up the release slider also increases the output volume when no keys are depressed - not sure if that is part of the original circuit design or a related issue (it could be stuck in constant release).

Attached is a PDF schematic. I've tested most components in circuit/some out of circuit and all seem to be working within a 10-20% value tolerance which seems typical of something from this era.
The circuit is working correctly at point F (the gate input to the envelope circuit). After this, all parts of the circuit are functioning but there's a residual -3 volts at the envelope output when no keys are depressed (at J / Z4-6), which is holding the VCA open.

Any suggestions on how to troubleshoot this?

Thanks!

 

[saint gillis]

Have you replaced the 3.3µ ?

 

[JamieOxford]

Have you replaced the 3.3µ ?

I haven't replaced anything yet, though I'm pretty sure I removed that cap and checked it in a component tester which read the value correctly. (I can test it again tomorrow)

 

[saint gillis]

Leaky caps can read the good capacitance on multimeter and still being bad

 

[k brown]

OT really, but just for grins, here's my vst emulation of the Explorer (with aftertouch added).

 

[saint gillis]

OT too, you did the emulation?

 

[Ike Zimbel]

Hello,

I'm hoping to get some help with an Arp Explorer I'm restoring.

The envelope isn't fully closing due to a rogue voltage, but I'm struggling to find the source of the issue as the envelope circuit looks like one large feedback loop!
Turning up the release slider also increases the output volume when no keys are depressed - not sure if that is part of the original circuit design or a related issue (it could be stuck in constant release).

Attached is a PDF schematic. I've tested most components in circuit/some out of circuit and all seem to be working within a 10-20% value tolerance which seems typical of something from this era.
The circuit is working correctly at point F (the gate input to the envelope circuit). After this, all parts of the circuit are functioning but there's a residual -3 volts at the envelope output when no keys are depressed (at J / Z4-6), which is holding the VCA open.

Any suggestions on how to troubleshoot this?

Thanks!

You might want to test S-1 with a DMM and make sure that you see a very low reading on the switch contacts in each position, like ~0 - 2 ohms. I've seen old switches go high resistance because of dirty contacts, but still "work". This has become my go-to test for old switches.

 

[k brown]

OT too, you did the emulation?

Yes, with Flowstone. I have a site with all my freebies:
https://kbrownsynthplugins.weebly.com/

 

[Hubbub]

That the release slider affects the volume tells me Q6 is probably turning on as it should, so the only thing that can cause Z4 (J) to be at -3V is if Z3 (I) is lower than it should be. Does Q7 collector pull up to ground when it should? Is Z3 working correctly as a voltage follower? For a quick Z3 test use a 10K resistor (which can't hurt anything) between ground and Z3 pin 3, and see what it does to the voltage on Z3 pin 6.

 

[saint gillis]

That the release slider affects the volume tells me Q6 is probably turning on as it should, so the only thing that can cause Z4 (J) to be at -3V is if Z3 (I) is lower than it should be. Does Q7 collector pull up to ground when it should? Is Z3 working correctly as a voltage follower? For a quick Z3 test use a 10K resistor (which can't hurt anything) between ground and Z3 pin 3, and see what it does to the voltage on Z3 pin 6.

That's indeed a much better analysis than my advice to replace the 3.3 cap..

 

[JamieOxford]

Thanks for your replies!

Unfortunately I'm still having issues with it. (I've fixed up another of these that I hoped would help diagnose the issue, but I'm not there yet.)

I've swapped out the 3.3uF Cap, tested the switches (0.5Ω resistance), replaced a couple of transistors (Q5/Q6), Q7 seems to be working, as are Z3 & Z4 from what I can tell (output voltage is following the input voltage).

I also swapped out Z2 briefly as it's a socketed switch. I thought this could be the issue, as the voltage polarity and envelope shape inverts at one part of the circuit when the release fader is turned up about 1/3 of the way (and these are the only NAND gates in the circuit).

From my notes, a change in the release fader sends:
- Z2C - pin 13 high to low & pin 11 low to high
- Z2D - pin 9 from low to high

At R60:
- Release low = -2.2V
- Rel 1/3 up = -12.2v

Any further suggestions before I start removing more components?

Also, the ARP Emulation looks great - I'll have to do a comparison!

 

[k brown]

Thanks for your replies!

Unfortunately I'm still having issues with it. (I've fixed up another of these that I hoped would help diagnose the issue, but I'm not there yet.)

I've swapped out the 3.3uF Cap, tested the switches (0.5Ω resistance), replaced a couple of transistors (Q5/Q6), Q7 seems to be working, as are Z3 & Z4 from what I can tell (output voltage is following the input voltage).

I also swapped out Z2 briefly as it's a socketed switch. I thought this could be the issue, as the voltage polarity and envelope shape inverts at one part of the circuit when the release fader is turned up about 1/3 of the way (and these are the only NAND gates in the circuit).

From my notes, a change in the release fader sends:
- Z2C - pin 13 high to low & pin 11 low to high
- Z2D - pin 9 from low to high

At R60:
- Release low = -2.2V
- Rel 1/3 up = -12.2v

Any further suggestions before I start removing more components?

Also, the ARP Emulation looks great - I'll have to do a comparison!

Ouch - the plugin is a 'just-for-fun' thing; doubt it will stand up well in a direct comparison!

 

[Hubbub]

At R60:
- Release low = -2.2V
- Rel 1/3 up = -12.2v

Any further suggestions before I start removing more components?

(On principle you should first make sure +15V and -15V are about right and free of any ripple. If not fix that first!) I'll assume the board matches the schematic, with no accidental shorts. An accidental short at Z3 pins 2 & 3 could cause the whole problem. I always look over the board closely before starting diagnosis.

Are you measuring R60 at the R60 - Z4 pin 3 junction? Do Z4 pins 2 & 6 match Z4 pin 3 within a few millivolts? If not Z4 is bad.

What does Z3 pin 3 measure? After release it should be about the same as Z4 pin 6, maybe a slight bit closer to ground. Do Z3 pins 2 & 6 match Z3 pin 3 within a few millivolts? If not Z3 is bad.

Also, look to see if either of CR9 or CR10 is shorted. After release the voltage across CR9 (or CR10) should be very close to zero, maybe just a few millivolts.

 

[Clintcleveland]

(On principle you should first make sure +15V and -15V are about right and free of any ripple. If not fix that first!) I'll assume the board matches the schematic, with no accidental shorts. An accidental short at Z3 pins 2 & 3 could cause the whole problem. I always look over the board closely before starting diagnosis.

Are you measuring R60 at the R60 - Z4 pin 3 junction? Do Z4 pins 2 & 6 match Z4 pin 3 within a few millivolts? If not Z4 is bad.

What does Z3 pin 3 measure? After release it should be about the same as Z4 pin 6, maybe a slight bit closer to ground. Do Z3 pins 2 & 6 match Z3 pin 3 within a few millivolts? If not Z3 is bad.

Also, look to see if either of CR9 or CR10 is shorted. After release the voltage across CR9 (or CR10) should be very close to zero, maybe just a few millivolts.

For what it’s worth, I just fixed a similar issue in a Doepfer ADSR. Different design, but the issue turned out to be a bad diode at what would be CR9 in this circuit

 

[JamieOxford]

So voltage rails are fine & Z3 & Z4 seem to track the input voltage correctly. I checked all diodes and none are shorted (all measuring a similar resistance of ~ 735Ω in circuit).

Attached are some oscilloscope readings in which you can see a few things:
Z2B is the input pulse which is correct
As the release is turned up, Z3 & Z4 flip polarity and rest around the negative rail
One side of R60 is resting around 0V, the other is pulled down to -2.4V
Something weird looks to be happening on one side of R61

 

[Hubbub]

Is R71 actually 33 Megohms? Probably, but that is too high a value to use in a reliable design - will make the PCB sensitive to moisture and surface contamination. I would have done it differently using lower value resistors for better production consistency.

Do you see Z2D-10 flip when it should? Does Q7 collector pull up to ground (on) when Z2D-10 is low? When Q7 is off the voltage across R61 should always be zero unless Z2B-4 is at -15. But Z2B-4 goes up to ground nicely, so I suspect that is not the cause. With release up does moving attack up (or decay up) change the voltage at Z4-6? (They shouldn't) If they don't and R61 shows zero during release then maybe Z4 is sucking too much input current - because it was damaged somehow.

 

[JamieOxford]

Thanks for your responses! I revisited this at the weekend and it seems some resistors were very out of spec and Z4 was not working correctly.
(The VCO was previously faulty due to a bad resistor, so I suspect they didn't use the best parts and/or they're getting old)

Everything seems to be functioning now from what I can tell, though the old ribbon connectors need replacing as they're failing.

One of the caps on the PSU is also leaking so I'll recap this (I've been using another PSU for testing).
Can anyone tell me what to look for in PSU capacitors besides matching the voltage and capacitance? I read online people recommending Panasonic or Nichicon caps though I can't find the right ones in axial packages. The ARP uses: 250uf 50v axials & 10uf 35v radials (originally 10uf tantalums).

I'm also working on a Polysix which needs new PSU caps: 4700uf & 2200uf 35v axials and some radials: 10uf/25v, 100uf/16v so parts recommendations would be very helpful.

 

[Ike Zimbel]

Thanks for your responses! I revisited this at the weekend and it seems some resistors were very out of spec and Z4 was not working correctly.
(The VCO was previously faulty due to a bad resistor, so I suspect they didn't use the best parts and/or they're getting old)

Everything seems to be functioning now from what I can tell, though the old ribbon connectors need replacing as they're failing.

One of the caps on the PSU is also leaking so I'll recap this (I've been using another PSU for testing).
Can anyone tell me what to look for in PSU capacitors besides matching the voltage and capacitance? I read online people recommending Panasonic or Nichicon caps though I can't find the right ones in axial packages. The ARP uses: 250uf 50v axials & 10uf 35v radials (originally 10uf tantalums).

I'm also working on a Polysix which needs new PSU caps: 4700uf & 2200uf 35v axials and some radials: 10uf/25v, 100uf/16v so parts recommendations would be very helpful.

Vishay (BC) make good axials but they're expensive. If you're not fussy about how it looks, there are ways to make a radial cap either just fit by spreading the leads apart (with heat shrink on the exposed parts of the leads) or by bending one lead along the body of the cap. You will often have to extend that lead with a piece of wire. You may need a zip-tie to hold the lead against the cap, or heatshrink over the cap and the lead...basically laying the cap on its side and bending / extending one lead (usually the positive one as it should be longer) down to the other end.