man, im sorry guys..synthex mod help PT. 2.

[rjd2]

sorry to bring up old stuff, but a new problem has arisen. thanks in advance for any help/leads/advice:

-i am now getting a large amount of jitter/voltage-jumping on the master synthex pitch bend. it is to the point that its unuse-able. mind you, this is with the slave synthex NOT EVEN ON, and the cable to send info DISCONNECTED. these results are happening ONLY INTERNALLY ON THE MASTER SYNTHEX. the joystick goes in 4 directions, of course. here's how they seem to be shaking out:

joystick down(pitch bend down)-the worst jitter. this is opamp input/output 1 going from 0 to +2.7v.
joystick up(pitch bend up)-slight jitter. this is opamp input/output 1 going from 0 to -4v.
joystick right(fliter mod)-slight jitter. this is opamp input/output 2 going from 0 to -11v.
joystick left( osc mod)-very very slight jitter. opamp input/output 2 going from 0 to +11v.


here is the signal path of the pitch/mod ONLY REGARDING THE MASTER SYNTHEX, or INTERNALLY:

pitch/mod board output pins>>>internal destination board AND inputs 1 and 2 of the master's opamp>>>opamp outputs 1 and 2 to output jack, of course.


so somehow, i seem to be getting some crosstalk(backtalk?) from the opamp to the internal destination board of the pitch/mod output board. as i bend the joystick, the voltage jumps around; it seems to do it at the exact same range each time(worst at around +1.5v of input/output 1).

any ideas? as usual, i cant thank you guys enough.(i dont want to scrap all the work YOU GUYS did on this thing, but in its current state, its so bad that i would have to disconnect the whole modified circuit if i cant fix it.)

[The Scum]

So this is with the outputs of the master disconnected?

And it behaves just fine when the slave is there?

Common sense things to check first:
-You used some small caps on your board for the power rails, right?
-Power is still getting to the opamp board? Check the DC and AC voltage levels right at the opamp power supply connections.

Where are you probing the voltages? At the joystick outputs, or somewhere else?

And just for curiosity/discussion, do you have a digital camera? Can you take a macro-mode photo or two of your handywork? I'd like to see what we're talking about.

[rjd2]

-it does this jitter whether the slave is connected or not.

-yes, used 22mfd/25v caps on the opamp +12v and -12v power leads, as well as ceramic caps across the same leads as the elec caps.

-when you say probing, do you mean reading w voltmeter, or where i tapped into as an output source for the pitch/mod control?

back in a bit with pics.

[earl parameter]

and you are sure the elecs are not backward? positive rail (+) cap (-) ground (+) cap (-) negative rail - also you could try to add more or use a higher value. maybe 100u


also did you add the 1M resistors on each of the positive inputs to ground on the new opamp?



steven

[rjd2]

crashsick-yes, i did both the elecs like that, and the 1m resistor to ground on both V+ and V-. i can try 100uf caps instead, see if that works.

still working on photos.

[earl parameter]

its pretty doubtful that the larger caps will clear it up. especially since they are just buffers. just thinking out loud


steven

[The Scum]

I agree, if there are some caps there, that rules out one line of thinking...bigger caps shouldn't change things much.

-when you say probing, do you mean reading w voltmeter, or where i tapped into as an output source for the pitch/mod control?

Probing means using a test tool of some sort: you're sticking the probes (of a multimeter or oscilloscope) into the device. So, yeah, where do you measure the unsteady voltage, where do you see it?

And what do the power supply voltages look like when it's happening, right at the opamp pins?

You were using TL074's, right? What have you done with the 2 unused sections in each of them?

And how long do you hold out before getting an oscilloscope?:lol:

Pix will be a great boon! Thanks!

[rjd2]

-tried 100mfd caps. no change.

-crashsick- believe the resistors you had me using to ground were across the opamp input pins and ground correct? if so, those are in place. i wired it up exactly as your last diagram showed.

-i am getting my readings from two places-the input pins 1 and 2 of the opamp, and the output pins of the pitch/mod board. as there is essentially nothing in between them, of course i am getting the same thing on both. so the problem is NOT downstream of the opamp. its really got to be the opamp interfering with the voltage that the pitch/mod board is sending out somehow.

-yes, i am using TL074 TYPE opamps. the two inputs(3 and 4) i am not using are getting tied into ground, per someones request. it was either the input or output, whichever i was told to put in before.

-a little more info: i notice that there is a wierd "dead spot" on ALL 4 of the voltages that are sent from the pitch/mod board(pitch +, pitch -, mod +, mod -). they ALL go a little funny right around 1.8V. more specifically, they reset to 0. so, if i slowly push down on joystick, voltage readings kinda go 0,-.1,-.5,-1,-1.5,-1.8,0!!!!!!!!!!,-.1, -2, -2,7. they should obviously be gradually going from 0 to -2.7.

heres a link to pics, although they probably wont really tell you anything unfortunately...

http://www.megaupload.com/?d=CPZ4BBMT

[brianroth]

I'm a dumbass <g>...what are we discussing? Lead me back to any original threads so I can at least look and ponder , since "wiggling voltages" sound vaguely analog to me.

OTOH, if this is actually DigiProbs, all I can suggest is 1. Throw it away and buy a new one. Or 2. Hope the OEM is still in business.

Bri

[rjd2]

thanks for your interest, brian. here's the previous thread(problem is actually less complicated than all that was covered there):

http://messageboard.tapeop.com/viewtopic.php?t=48877

in a nutshell, here's what we are dealing with:

1-all the mods covered in that previous thread were completed and functioning properly, but it introduced ONE PROBLEM.

2-MASTER SYNTHEX has a pitch/mod board(lets call this 8503) that sends output voltages that run to an INTAKE pitch/mod board(lets call this 8504).

to complete this mod, i tapped into 8503 for a source of voltage control. so now, 8503 sends voltage to both 8504 AND the opamp configuration(inputs 1 and 2). the problem is that introducing the opamp into the circuit causes the voltage output of 8503 to flutter. so instead of a smooth bend, the pitch jumps around, rendering the point of a pitch bender on a synth virtually unuseable.

trying to figure out how to make that voltage NOT flutter. fyi, this happens with the slave synthex not even on or plugged in, so this is a 100000% internal problem of the master synthex.

thanks brian!

[rjd2]

UGH. BAD NEWS.

just to see what could be going on, i disconnected the power leads to the opamp, and the output taps i had put on the pitch/mod board(8503) that were going to inputs 1&2 of the opamp. so the entire master synthex mod was DISASSEMBLED, not connected to anything.master synthex was totally back to "stock".

the problem was still there. furthermore, i actually think its getting WORSE the longer i leave it on. when i noticed it at first, it seemed minor. now, it seems very noticable.

im going to write an experienced synth tech i know and see if he has any ideas. but i obviously screwed something up. FLEW TOO HIGH ON WINGS OF PASTRAMI. ugh.

thanks for your help guys; at this point im assuming its something that only someone who is versed in the machine is going to have leads on.

[earl parameter]

not sure how old the elka is but could this be the old smoothing caps failing? if its progressively getting worse over a short period of time i'd say that would be the one of the first things i'd replace.

the huge caps around the transformer. i forget what the were but maybe a pair of 2200u and 100u caps on either side of the regulator.

also if the regulator was already being pulled on to its limits by the original circuitry and we added the new OPA to it then that could be giving some odd behavior too. that is possible but pretty rare i'd say. especially since we just added a buffer and it shouldn't be drawing that much current and i'm sure the designers would have over shot the draw.

you should try to replace the filter caps anyway. if its that then its probably the easiest thing to do right now and it can only help.


steven

[rjd2]

cant be any of the PSU caps, i replaced all those this year.

here's a possible theory:the joystick is essentially 2 potentiometers, one vertical, and one horizontal. is it possible that the resistance rating for these potentiometers would be such that drawing twice the amperage(right? cause i had an extra line tapped into these outputs that ran to the opamp inputs)from the outputs would cause the pots to malfunction/get screwy/wear out in certain places?

im NOT 100% sure its getting progressively worse. that could just be perception. i AM 100% sure its not a temperamental thing, as its done it every time ive turned it on since i installed the mod.

im looking at the pathway, and it seems like the problem would HAVE to be in the joystick board,(the joystick board has no caps/resistors/anything other than a joystick on it) cause i ONLY get the jitter/flutter when the joystick is activated.

[earl parameter]

if you are just pulling lines straight off of the joystick you are probably getting just raw voltages. i guess we never covered this? the joystick's output actually goes through a tremendous amount of processing before its used in the rest of circuit and i can only guess that the designers wouldn't have done that if it wasn't needed. shit just that alone board is more complicated then most standard VCO's. there's more happening then just the joysticks but still...


if you look at page 12 of the scheme along the right side of the dotted border there are two points labeled JOY -VCF and JOY -VCO

JOY -VCF is pin 1 of a 1458 labeled 2D

JOY -VCO is pin 7 of the same 1458 labeled 2D


figuring out which is which 1458 might be tough if the board is not labeled. there should be six in the same area around a 3080 and 13600

the board label should be CST5801 and CST5802 is the joystick board


try seeing if that clears anything up at all. i guess it could be that the carbon in the joystick pots are fading but it might be hard to find a properly fitting replacement. you could pull it out and try cleaning the wafer with distilled water and a q-tip and making sure it 100% dry before testing it. due to its age and how hard it might be to replace don't use the deoxit shit or anything else it will just speed up death. if you see obvious wear then i guess that's a sign to look for a repayment. but none of that will work if it won't let you open the casing.

the elka is a complicated fucking synth for sure. 51 page sheme!!!


steven

[rjd2]

thanks alot, crashsick.

i heard back from the synthex tech who has been my point guard through all this. he suggested that i may have damaged 4 of the opamps on the 5801 board, and suggested replacing them. i plan on sourcing them and replacing on monday(with sockets, in case i have to do it again).

so crashsick, as someone whos looked over the schem and has an idea how the signal path works, do you think that it would be reasonable to hope that there are OUTPUT pins from the 5801 board that could possibly be used to run the opamp mod configuration, instead of straight voltage leads off the joystick board(5803)?

yes, youre right-the synthex is a complicated and bulky beast, and handles like a bmw 745. 12 cylinders of pure grace.... fyi, the guy who designed it, mario maggi, is apparently a revered dude in the world of synths because of this machine.