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

Help me troubleshoot Pharmasonic 704
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Help me troubleshoot Pharmasonic 704
wendallsan
Hi all,

I finally have the gear I need to test and calibrate my Pharmasonic 704 VCA's. I've tried following the directions provided by Pharmasonic to calibrate it and am not getting the desired results as far as I can tell.

I tried to follow the Linear Adjustment step in the calibration instructions:

Quote:
The 1st step is the adjustment of the DC balance.
Set the switch to LIN, all pots to 0 and feed one of the modulation input with 10V.
Probe the VCA output with an oscilloscope (use high gain) and adjust the DC BAL
trimmer so changing the CV amount produces minimum change.


Here is what I did: I've got a Mutable Instruments Shades set up to provide a +10V signal into the 1st of the 2 modulation inputs on the module. All pots are turned fully 'off,' and the switch on the VCA is set to 'linear'. I have the VCA out headed into a voltage meter so I can monitor the output voltage.

If I understand the instructions correctly, I should be watching the output voltage of the VCA while adjusting the voltage of the signal being fed into the modulation input, and then adjusting the DC BAL trim until I don't see the VCA output signal changing as I adjust the input CV level.

What I'm seeing is that the VCA output is at a constant +6.5V regardless of what is fed into the modulation input, and adjusting the DC BAL trim doesn't seem to change this result in any way.

Next calibration step:

Quote:
Once done, feed one audio input with a 10Vp-p SIN wave and set the level to 10.
Feed one of the modulation input with 10V and set the amount to 10.
Probe the VCA output with an oscilloscope and adjust the VCA GAIN trimmer so you get a 10Vp-p SIN wave.


When I send a 10Vp-p sine wave into one of the inputs (knob turned up to 10) and a 10V CV input into one of the modulation inputs (also turned up), I am seeing a square wave output that is clearly heavily distorted. Adjusting the VCA gain trim does not seem to change this behavior in any way.

So my takeaway is that something is probably wrong, and the 1st thing I suspect is my ill-gotten 3080 ICs from China, as the LM301's that I sourced in a similar fashion are bad and folks here are telling me not to order cheap components from China again, and also that there are a lot of bad 3080 chips out there. So I'm hoping someone will be able to help me to troubleshoot one of my VCA's so I can confirm that the 3080's are the root of the problem and I can some replacements ordered. If they were cheaper, I would just order them straight away, but at $5 a pop, I figured it would smart to check this before hitting 'buy', as I'm going to need a lot of them. The good news is that both VCA's are displaying this same behavior, so at least my problems are consistent. I've also done some preliminary tests with my LFO's and have had bad results from those as well, and they also sport the 3080 chips in them, making them even more suspect.

Would anyone be willing to help me determine that things around this chip are good so I can confirm that these 3080's are crap? Many thanks for any assistance.
wendallsan
Here are some notes and measurements based on what folks had asked me to do while troubleshooting the Pharmasonic 707 in another thread.

The green LED on the module lights as soon as power is supplied to it.

Resistance between the IC socket's pins and ground:

legend: '-' = inifinte resistance

1 -
2 549 ohms
3 546 ohms
4 -
5 -
6 -
7 36.7k
8 -

Resistance between pins:

note: there was a lot of 'drift' on these as they started at one value and then stablized at another. I'm guessing this is due to caps in the circuit? Sometimes this took 10 seconds, sometimes this took close to 1 minute. Not sure if this matters at all.

legend: 'x' means ground pin is plugged here


1 x
2 -
3 -
4 -
5 -
6 -
7 -
8 -

1 -
2 x
3 1.09k
4 started at ~150k and dropped to 44k over about 1 minute.
5 -
6 -
7 started at ~20k and raised to 37.2k over about 1 minute.
8 -

1 -
2 1.09k
3 x
4 started at ~55k and dropped to 44.5k over about 15 seconds
5 -
6 -
7 37k
8 -

1 -
2 44.3k
3 44.3k
4 x
5 -
6 173k
7 24k
8 -

1 -
2 -
3 -
4 -
5 x
6 -
7 -
8 -

1 -
2 -
3 -
4 -
5 -
6 x
7 176.7k
8 -

1 -
2 37k
3 37k
4 24k
5 -
6 575k
7 x
8 -

1 -
2 -
3 -
4 -
5 -
6 -
7 -
8 x

Edit: please ignore this next bit, it is meaningless
Voltage between ground the IC pins seems to be zero on all pins, with the power supply turned on and the chip in its socket. The only behavior I notice here is that touching a lead to pins 1, 2, 3, or 8 causes the green LED to to turn off and the red LED to turn on.
batchas
wendallsan wrote:
Voltage between ground the IC pins seems to be zero on all pins, with the power supply turned on and the chip in its socket.

What do you mean exactly with voltage between ground the ic pins?
wendallsan
batchas wrote:
wendallsan wrote:
Voltage between ground the IC pins seems to be zero on all pins, with the power supply turned on and the chip in its socket.

What do you mean exactly with voltage between ground the ic pins?


I was asked to take some measurements by another user when troubleshooting another module, and I think I just mis-understood this step. These readings are probably irrelevent and worthless. Hopefully the resistance readings and voltage readings between the IC socket and ground are of some use.
wackelpeter
The DC-blance trimpot is used to eleminate the singal "bleed through", sorry i lack of a better description, from your CV source... Best way to do this is with an oscilloscope.

If you have none, then you can do this by ear, set the pots as described, put a 10V square wave into the CV input (if it has only attentuated inputs, you of course have to turn the attentuating pot of that input fully clockwise) hook the audio output of your VCA to your mixer or amp and turn the trimmer until you have the least noticable sound (therefore you may turn your amp a bit more up) or the smallest amplitude on your scope.

The other step is as described audio into the audio in, steady 10V signal to the CV(modulation) input set the pots as described and turn the gain trimmer to reach an amplitude of 10Vpp at the output. if you had an oscilloscope the input and output then should be the same amplitude. With the pots set at those position you would do nothing less, than trim/set the gain(amplification) ratio input vs. ouput 1:1.

Well, without the schematic i'm not sure what the problem with your steady DC output of 6,5V is. Some VCA's also have an offset-nulling trimmer which is used to center the output level around gnd/0V. But 6,5V seems a lot and despite that your signal would then just appear with an DC offset, meaning instead of with a ratio 1:1, and a 10Vpp signal it wouldn'T go from -5 to +5V but with say 1V offset from -4V to +6V...

How many trimmers are on board, are there any schematics online or is this a clone with the same basic components as the original system 700 VCA, so we could use that schematic as reference?

Perhaps some detailed and close pics of both sides of the PCB wouldn'T harm. there are are enough of people here with a great talent to spot obvious errors.
wackelpeter
P.S. Assuming the first IC you showed your measured resistances, is the CA3080, i'm pretty sure there should be an reading between pin 4 and gnd as it's the -V supply of this IC and you have some trimmers or pots connected between gnd and + and -. Thus you should read there something and the value should change a bit while rotating the trimmers or pots connected between + and -.

Well i'm not an expert and assuming it's pretty close to this schematic, have you checked the direction of the diodes, electrolytes and both trannies if they match with this schematic (D1,2 Q1,2)
Regarding the trannies you should measure between both emitters those 112 ohms (unpowered module) are these matched regular average trannies, a matched pair?

http://www.florian-anwander.de/roland_system700/704_a3.jpg
wendallsan
Hi and thanks for the response!

wackelpeter wrote:
How many trimmers are on board, are there any schematics online or is this a clone with the same basic components as the original system 700 VCA, so we could use that schematic as reference?


There are a total of 3 trims on the board: Exp Width, Exp Offset, and DC Bal. I've been adjusting the DC BAL and not seeing any changes in the output of the amp while feeding +10V into one of its modulation inputs, per the calibration instructions. It seems to be outputting a constant +6.5V when I do this.

Quote:

Perhaps some detailed and close pics of both sides of the PCB wouldn'T harm. there are are enough of people here with a great talent to spot obvious errors.


Here is a photo of the front and back of the board:

https://dansteeby.com/wp-content/uploads/vca-front.jpg
https://dansteeby.com/wp-content/uploads/vca-back.jpg

I do have access to a scope, and am happy to use it SlayerBadger!

I don't have a schematic, and am not sure if this design is heavily altered from the original. I know Pharmasonic has stated that they have swapped out some hard-to-find components for modern ones, but am not sure what that means for the original schematic's usefulness.

The resistance measurements that I provided are indeed for the 3080 at IC1, as this is the one I already suspect is probably bad, and am trying to confirm this as true.

I believe that all oriented components such as diodes, polarized caps, IC's and transistors are properly oriented according to the silkscreen printing on the boards at least.

Quote:
Regarding the trannies you should measure between both emitters those 112 ohms (unpowered module) are these matched regular average trannies, a matched pair?


I'm not sure how or what to measure between the emitters of these transistors-- please let me know what you're looking for (voltage, resistance, etc) and I'll try. I'm very new to circuit troubleshooting and haven't done much with testing tranistors yet. The transistors are not matched pairs.

Thanks very much for your input and advice! Please let me know if anything looks wonky in the photos or if I can provide any other details.
guest
ok, so what are the voltages on the CA3080 when its powered? you should have 0 on most pins, with 7 at 12V, 4 at -12V, and 5 at -11.4V. if thats all good, then measure the voltage drop across the 10k resistor connected to pin5 with the VCA gain knob turned full up. this should be 6V or so.
Modulart_JP
The answer to your problems is in the thread about your module 707 issues...
wendallsan
guest wrote:
ok, so what are the voltages on the CA3080 when its powered? you should have 0 on most pins, with 7 at 12V, 4 at -12V, and 5 at -11.4V. if thats all good, then measure the voltage drop across the 10k resistor connected to pin5 with the VCA gain knob turned full up. this should be 6V or so.


I measured the voltage between the ground pin on my power supply and the IC pins, here are the results:

1 9.71
2 0
3 0
4 -11.85
5 5.11
6 5.88
7 11.89
8 9.72

I measured the voltage drop on the 10k resistor connected to pin 5 (first time I'm trying to measure a voltage drop, so please let me know if I'm doing something wrong, I'm just following along with the first article on the subject I found ... ) by placing my meter leads to either side of that resistor with the meter set to measure DCV. The output appears to be either 0 or 0.1mV regardless of the position of the gain pot next to it. Power was not supplied to the circuit during this test. Here is the resistor I tested, in case I'm messing with the wrong one:[/img]



Thanks again for the assistance!
wendallsan
Modulart_JP wrote:
The answer to your problems is in the thread about your module 707 issues...


I'm sorry, I'm not experienced enough to see the solution based on the answers given in that thread. I'm trying to go through the steps taken in that thread to provide what details I can for this one. Hopefully this saves a little time and saves folks from having to answer the same questions all over again for this thread to hopefully move things closer to a solution without wasting a lot of time or being a nuisance.
guest
your 3080 is dead, sorry. the bias pin should be 0.6V above the negative rail, and its at 5.1V.
wendallsan
guest wrote:
your 3080 is dead, sorry. the bias pin should be 0.6V above the negative rail, and its at 5.1V.


Thanks very much for confirming this suspicion. New IC's have been ordered and I'll resume calibrating/testing when they arrive.
Modulart_JP
wendallsan wrote:

I'm sorry, I'm not experienced enough to see the solution based on the answers given in that thread. I'm trying to go through the steps taken in that thread to provide what details I can for this one. Hopefully this saves a little time and saves folks from having to answer the same questions all over again for this thread to hopefully move things closer to a solution without wasting a lot of time or being a nuisance.


Experienced ?
No need to be.
We already told you in the other thread that you should not have bought those parts and they were probably fake.
Throw away all the Chinese junk you bought and get reliable parts.
All your problems come from those fake components.

These days the only reliable CA3080's are the Rochester ones.
Available from Modular Addict or Thonk.
wendallsan
After a 1 week break from this project for a trip out of town, I got home last night and was able to swap out the bad 3080 IC's in my VCA's with new ones from Jameco. Sure enough, they are now working well, although I have not yet fully calibrated them.

On top of that, I was also able to quickly complete the ADSR, Noise/Ring Mod, and Mixer, which brings me to a total of 8 of 16 modules for the project working! I do have an output level issue with the Noise source, but will post about that in a separate thread.

It's peanut butter jelly time!

All that is left are the double-layered board modules, which include the LFO's, Filters, and Oscillators. I started in on one of the LFO's, but it did not immediately work after swapping out the bad IC in it. I should have some more time to troubleshoot this circuit tonight, and will reach out if I'm not able to get it working.

Thanks again to all that offered help with this, I'm feeling some serious momentum now that I'm out from under those crappy Chinese IC's.
guest
glad its all working. what a difference an IC can make!
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group