Audio signal path - leaking

[samplebias2]

Asking for general ideas here since nobody probably has an EMS vocoder-

As you know, a vocoder imprints the sonic signature of one source, onto another i.e. robot voice (speech onto synth tone).

My EMS 2000 vocoder has an "output select" switch that lets you hear the carrier, modulator, or vocoder output. Only 1 at a time.

After the unit warms up, my human voice starts fading into the vocoder sound intermittently, then stays on..

I have replaced that 3-position switch, and the "pause stuffing" switch ( this feature blends those two sources, voice and vocoder if that's what you want to happen)

My question is, in general- what WAY could two separate signals merge once the unit warms up- corrosion or some way the signal can "jump" from one to the other? IC chip? Unit is from 1989 and was barely used. Hoping to arm my tech with more info when I bring it to the shop when it's my turn in 3 weeks.

 

[Bo Deadly]

Just from glancing at what looks like an old service manual, the output section appears to be something like this?



Again, I'm just spitballing here but I do see that if the signal pin of J8 is not being shorted through the normalling contact properly, the signal "EE" will bleed into the mix.

Try connecting signal pin on J8 to ground using an alligator jumper. If that fixes it, try cleaning the normalling contact, check the sleeve to ground connection, broken solder joint to ground, etc.

Although that does not explain why it only happens after some time. That sounds like a transistor being used as an analog switch is not functioning properly after it heats up maybe. Look for transistors that look a little roasted. Semiconductors that get subjected to a lot of heat can get burned up inside and start to loose their ability to turn completely off or completely on. Or a FET is just busted.

If that doesn't fix it, you really need a scope and much preferably a schematic.

 

[samplebias2]

I really appreciate the insight- this is above my pay grade, and will print your reply as addditional info for the electronics shop tech to digest- anything to cut down hours- I feel this could be a LENGTHY process to isolate this gremlin!!!!

 

[user 130681]

Three things come straight to my mind;
1/ If the +-12volt lines are not stable, the DC offset levels will vary and cause the pre set levels to vary & in turn cause more or less carrier, modulator or vocoder signal levels. This could be due to dried out heatsink compound or just old regulators.
2/ There are a number of very high value resistors controlling the FETS. These hace a habbit over time to change value or become unreliable.
3/ Bad connections on screens or controls.

Just a note, CA3080 ICs have not been available for many years if not decades. Don't trust fleabay knockoffs!

 

[samplebias2]

Three things come straight to my mind;
1/ If the +-12volt lines are not stable, the DC offset levels will vary and cause the pre set levels to vary & in turn cause more or less carrier, modulator or vocoder signal levels. This could be due to dried out heatsink compound or just old regulators.
2/ There are a number of very high value resistors controlling the FETS. These hace a habbit over time to change value or become unreliable.
3/ Bad connections on screens or controls.

Just a note, CA3080 ICs have not been available for many years if not decades. Don't trust fleabay knockoffs!

Really appreciate this tip! This was originally a 220v unit sold in the UK. I thought it was great that it has almost zero use, but I'm thinking now that was not a good thing. But I'm grateful the guy spent his severance check on it, as a trophy, therefore bringing it into the world... less that 150 units ever made.