9700 VCF hot transistor
Page 1 of 2

Author:  bohostudio [ Tue Jul 27, 2010 11:55 pm ]
Post subject:  9700 VCF hot transistor

I've recently assembled a 9700 modular in my own enclosure. It seems to be working, for the most part, however i'm having a few issues. The worst of these is the VCF modulator section. The LED was working properly for a bit, bit now it fails to light up. Transistor Q11 is extremely hot in a matter of seconds, making me afraid to leave it on. I am using a Radio Shack 12V 800 mA AC Wall mount adapter (actually it has a 13V setting as well). I am a newbie to the DIY electronics stuff and i'm not quite sure where to begin to troubleshoot this problem.

Also, I'm having a problem in the VCA section. when the Pan knob is in its central position, neither sound is audible.


Author:  PAiA-Scott [ Wed Jul 28, 2010 10:10 am ]
Post subject:  Re: 9700 VCF hot transistor

Are you using the RS Model: 273-328 transformer for the MIDI2CV8 or one of the 9770 series power supply kits? With the full complement of P9700S modules, you're probably best setting the transformer for the 13v tap to ensure there's the needed 15-18vdc (positive and negative) to the VCO, VCF, and VCA modules.

Check the wire from point V to R91-2/R91-3 on the panel to be sure it hasn't melted against some other terminal, the panel, or the body of the potentiometer, as connections were made. It links Q11 to this R91 Attack control.

Maybe there was a mix-up with the installation of the potentiometers on the VCA and a 5Meg is at the R109 Pan control??? Lately the pots have had the value printed on the top/panel/shaft side, but if you compare markings on the back with another similar valued control in another section/kit, you'll know whether to look closer or not.

Author:  bohostudio [ Thu Jul 29, 2010 12:08 am ]
Post subject:  Re: 9700 VCF hot transistor

hey Scott,
thanks for the reply! I stepped the voltage up to 13v on the RS supply and it seems to have improved performance a bit (now the VCO mod doesn't change pitch when i switch the cycle on).

as for the VCF, i found a bridge that was causing the transistor to heat up and fixed it. now the only problem i'm having in the vcf is the cycle mode failing to activate. the a-r triggers fine with a trigger/ gate activating it. oddly enough, i can put my finger on a couple of the resistors and the cycle works fine. i'm not sure how to fix this permanently though...

haven't had a chance to check out the VCA pot yet. i'm going on the road for a few days, but hope to get everything working in the next week or so.

thanks again!

Author:  PAiA-Scott [ Fri Jul 30, 2010 10:39 am ]
Post subject:  Modulator not cycling

Check the S1 switch for loose/tilted terminals. Heat from soldering can soften the plastic body of the switch allowing the terminals to misalign resulting in no or intermittent contact, or, a short with the metal bushing. It is when wires AB and AC to this switch are connected the cyclic action can occur.

Double-check the color codes for the resistors connected to IC8C, ie it's easy to get a red-violet-yellow-gold 270k and yellow-violet-red-gold 4700ohm mixed-up if not paying careful attention when installing these parts.

Of course it's important the V+ and V- are the needed positive and negative 12vdc (pin 11 of IC8 is negative 12v and pin 4 of this IC is positive 12v). The IC2 and IC1, 7812 and 7912 regulators, on the board in each others place can result in the unit sort of working, but not right.

Weirdness can occur when one of the two grounds to a module is floating: Are the terminals inside the four-pin power supply cable cover securely with their catch in the opening? Did the solders for the associated header flow to both the solder pad and the pin extending into the joint?

Email me photos of the soldering and the parts and panel connections and I might be able to spot and point out possible trouble.

Author:  bohostudio [ Mon Aug 02, 2010 4:39 pm ]
Post subject:  Re: 9700 VCF hot transistor

Hello, thanks again!

I'm back to the VCA for a moment. I still can't figure out why the left and right channels fade to silence in the center position. I checked the pot, it's definately 10k.

Also, the noise oscillator sounds from the A output regardless of Left In normalization and/or Pan Position.

Author:  PAiA-Scott [ Tue Aug 03, 2010 11:13 am ]
Post subject:  VCA pan, noise source leak

If the trouble with the VCF is an open-circuit on G or SG, then the VCA further down the chassis is going to be missing this too. Try the VCA with its power connecting cable into the MIDI2CV8 instead of the VCF. If this does the trick, reconnect the VCO to the MIDI2CV8 and plug the VCA into the VCO. If things are still OK, inspect the solders on the VCF power header, the terminals in the cover at the end of the cable from the VCA, and finally, the solders for the wires of this cable at the VCA. Wire D connects the circuit board ground up to the panel at R109-3 (pan) and it distributes to other panel points from there via bare wire links (see the assembly manual illustration supplement figs. 3 and 4).

You'd mentioned finding a solder bridge and a short-circuit like this is another way things can go wrong. The joints are closely spaced on the printed circuit boards of the modules in a P9700S and it's important the joints don't have excess solder forming blobs and that the wires extend mostly straight up from the joint and are clipped at the top of the joint instead of being bent over close to the board and with lengthy excess which could extend over and contact adjacent but separate circuits.

A DC voltage test can be made on the Pan control to see that it's doing the job of providing 0 to 12 volts dc to wire C on its middle terminal as it is rotated from R to L. IC3B inverts and offsets the voltage to range from 12v to 0v (the circuit node including IC3pin7, R5, and R9) for the same R to L rotation of the Pan control.

Inserting a plug into the L VCA input should work to move the tip contact away from its normally-closed connection with the shunt contact. The Noise Source signal runs up to the connector shunt terminal (X) via wire P. When there is not a plug in this connector the tip contact rests against the shunt contact and the noise is 'normalled' on through to the L VCA input via wire O on the tip terminal (T).

There are two other ways to interrupt the noise in its normal path to the main VCA (A), the Left VCA output connector and the VCA A / balanced-modulator-carrier input connector. When a plug is inserted in the VCA L Output connector (another closed-circuit connector with the shunt terminal for a normally closed connection with the tip), the VCA L output is diverted away from the passive mix via resistors R116-117 to the L+R output connector and the VCA A/Carrier Input. When a usual mono (tip-sleeve / TS) plug patch cord is inserted in the stereo (tip-ring-sleeve) connecter, this works to connect a source to the VCA A/Carrier input but also the ring contact of the socket contacts the shaft of the plug which is also in contact with the socket sleeve (ground/circuit-common) and the L+R passive mix is shunted away or diminished at this J3 end of R115 (but not at the L+R output at the other end of R115) (see figs. 3, 4, & 7).

When making these checks and tests, be careful to avoid accidental shorts between un-mounted powered modules and the chassis metal. A rag or magazine or piece of cardboard can help keep things separated.

Author:  bohostudio [ Fri Aug 06, 2010 12:16 am ]
Post subject:  Re: 9700 VCF hot transistor


i've disconnected the VCF completely from the power chain (it was already last in my chain). It was working, for the most part, but I wanted to focus on the VCA issues without wondering if it was somehow involved. I've tried everything i know how to do, but admittedly i don't really know what i'm doing.

The L&R signals are still disappearing in the center of the pan knob. I checked DC there and its definitely sweeping from 0 to 12V right to left. I'm not really sure how to test from there, i.e. where to put the probes and what readings i should be looking for. The schematic is starting to make some sense to me, I just don't know what to do with it...

Maybe this is related to the pan issue, but the A output makes a hissing sound for the duration of the envelope cycle. in the pan knob's center position, it is all I hear. I completely disabled the noise oscillator section via wire "P" and this noise still persists, leading me to think it must be another sound entirely. The noise sound itself is a bit faint, but it appears to be normalized correctly.

what should i try from here? thanks again

Author:  PAiA-Scott [ Mon Aug 09, 2010 5:40 pm ]
Post subject:  DC voltage testing on the 9710 VCA

When checking dc voltages with a multitester/dvm/vom, it can be a lot easier if you have an alligator-clip test-lead, or a set of clip-on probes for the tester. Then, at least one probe, the black common/minus can be clipped on to the ground/circuit-common/0vdc point (ie the bare wire making point to point links for the panel controls and connectors.

To check the voltage regulators, just measure for DC volts on the 324 op-amp power supply pins. Pin 11 should have the negative DC and pin 4 the positive DC. Look for -12 and +12 readings at each of these points with one probe on ground. If the tester is analog, reverse the polarity so the needle doesn't try to swing the wrong direction. Note the regulators are two different parts even though they look similar. IC2 is a 7812 and IC1 is a 7912. The DIP IC pins count-up around the part in a ccw direction from the notch as viewed from the top-side of the board

I think your L and R VCA sections might be OK, but there's some sort of trouble in the main, VCA A section affecting the L+R mix. To confirm this. Take the output from the L and R VCAs direct from their outputs, Lout and Rout The Pan control should work like a volume adjust for the signal patched through these vcas and appearing at their outputs (Noise is the default for the VCA L and some signal would need to be connected as a source to VCA R.

When making the VCA R test, try making a dummy patch into the VCA L output or one that doesn't go anywhere but simply amounts to a plug in this socket. This will disable the VCA L output from the L+R passive mix. It is possible that DC voltage could be present in VCA L and affecting L+R and VCA A (backwards C21, solder bridge, etc.).

Following are dc voltage tests that can be made in the VCA A section going sort of left to right from IC6c to IC6D according to the schematic. Set ADSR controls to 9:00 and modulator to Cycle.

IC6pin9 0V
IC6pin10 0V
IC6pin8 4.9

R34 adjusting disk (appx. calibrated setting) 4.4V

IC4pin10 5v
IC4pin8 5v
IC4pin2 -0.73
IC4pin3 -0.074 to -0.091
IC4pin1 -0.01 to -0.02
IC4pin4 -0.02 to -0.20
IC4pin14 -12
IC4pin5 -10.47
IC4pin6 6.33 to 6.35
IC4pin12 6.35 to 6.37

IC6pin13 4.75 to 4.90
IC6pin12 5.35 to 5.37
IC6pin14 -0.04 to 0.23

P9700S tips and suggestions
P9700Sts.pdf [62.99 KiB]
Downloaded 1357 times

Author:  bohostudio [ Mon Aug 09, 2010 8:18 pm ]
Post subject:  Re: 9700 VCF hot transistor

Hey Scott,
thanks again, sorry to be such a pain. I've tried the tests you mentioned.

voltage regulators appear to be fine.

The outputs from L & R appear to work fine. the pan does work like a volume knob for the outputs on their respective sides, with the signal disappearing completely by their midpoint positions...

All of the IC pin checks seem to be pretty close to the settings you mentioned...

Author:  PAiA-Scott [ Thu Aug 12, 2010 7:20 pm ]
Post subject:  Monitoring the 9710 VCA L + R and VCA A Outputs

It finally occurred to me that it might be the device connected for monitoring the passively-mixed L+R output that is resulting in the loss at the mid-setting of the Pan control. For a significant amount of the signal to appear at this output, the impedance of the connected load should be 10k or more. Headphones into this connector would certainly be a lot less than this.

With the normalled connection of the L+R signal into VCA A, there is about 180k at the end of the passive mix.

I was scrutinizing the one I have here and saw that noise source bleed into the VCA A output could be shut down by just gently touching a shorted plug into the L input. This doesn't move the tip contact serving to couple the L VCA input away from the shunt contact and it grounds the noise source output, muting it.

Here is a mod for a better way of working this:
9710NoiseMute.pdf [259.77 KiB]
Downloaded 1371 times

Instead of making a Switch-Plug, you could also use a TRS plug with the ring and sleeve terminals wired together (and the sub'ed TRS L VCA input connector). When plugged into the L VCA Input, the noise would be muted. A second TRS plug could be made for use when Noise is needed--the tip and ring of this would be jumpered at the plug terminals.

Page 1 of 2 All times are UTC - 6 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
./cache/ is NOT writable.