Phase reversing Blend Circuit

[Christoba]

I'm troubleshooting a Ross Compressor blend mod that is just not cooperating.  I have removed the board and soldered up connections to keep the compressor circuit from getting in the way.  The signal going into "CompOut" works perfectly well and the blend seems to work with this.  However, a signal going in from "Q1 emitter" does nothing.  I've pulled all resistors and caps and retested.  They all check out.  I've gone over any bridges or shorts and cant find anything.

Where should I measure to help you all diagnose this problem?

[blackieNYC]

So you are injecting signal at those two points, one works, one doesn't?  Measure voltages at each pin. 

[Christoba]

Exactly.  Measured IC pin voltages with blend knob at 0 and then at 10.  Both readings follow.  Also, does it matter which way the IC socket and socket 'only' is orientated.  The IC chip itself is properly oriented but the socket is upside down.

First blend setting
1-5.4v
2-3.2v
3-3.2v
4-0v
5-9v
6-1.3v
7-6.5v
8-3.2v

Second blend setting
1-2.5v
2-2.5v
3-2.5v
4-0v
5-9v
6-1.3v
7-6.5v
8-2.5v

[GibsonGM]

How are you 'reading' the opamp pins?  They are numbered like this:

1  8
2  7
3  6
4  5

So you might want to re-post the voltages, if you didn't know this.

Socket orientation should not mean a thing, but that chip MUST be oriented correctly, 'notch' at the top...

[Christoba]

My fault on the pinouts.  Here are the correct readings
1-2.5v
2-2.5v
3-2.5v
4-0v
5-2.5v
6-6.5v
7-1.3v
8-9v

Second reading
1-5.4v
2-3.2v
3-3.2v
4-0v
5-3.2v
5-6.5v
7-1.3v
8-9v

This chip is good to go.

Sent from my SM-G900V using Tapatalk

[anotherjim]

I think there is an omission from that blender schematic.

If the "Comp out" is from the 10k/Level output of the compressor, then it is being ground referenced and will pull the blender IC pin voltages low. Pins 3 and 5 should stay about 4.5v whatever the signal or setting states.

In the input to R31, insert a 1uF cap in series. Negative plate toward input. That will keep the blender voltages where it wants them to be. Note that a cap (C20) is already shown fitted in the clean input to R29 for the same reason. R28 is unnecessary if the blender is permanently installed in the same case as the comp'.

[anotherjim]

Just a thought, but you may be expected to remove/re-purpose the existing compressor output level pot which would explain the lack of a DC blocking cap on the input to the blender, since the existing output cap would serve to do that.

[Christoba]

That is correct.  The level pot becomes the new output going to the amp.  For testing purposes I ran level 1 to the amp which seems to work well.  Its just that damn Q1 side that isn't working. 

Thank you for your help on this.  So just to clarify, you don't suggest running the 1µf cap at the "comp out".  What should I do?

If it helps, this is the compressor board that I am using. It is built and has been working great although I plan on doing something about the extra treble that it adds. (side note, think changing C9 would be a good place to attenuate the treble?)https://aionelectronics.com/project/aurora-ross-dyna-compressor/

[anotherjim]

Correct, no need to add another cap - but try one anyway. If the compressor output cap is electrolytic, it may not be polarized properly without the ground path via the original level pot.

You should definitely have 1/2 of the +supply ( if 9v then 4.5v) on the blender op-amp pins 3 and 5. Disconnect both inputs from the compressor and check you have that voltage first.

[Christoba]

I separated the blender from the compressor a while ago and am running a 9v supply directly to the "9v in".  I've measured the voltages and they are 9v. 

I see what you are saying, that resistor should be cutting the 9v in half?  I haven't really figured out calculating voltage drops across resistors as I haven't figured out how to calculate for amperage on a circuit.  Well, at least it hasn't clicked for me yet.  Note, one of the issues that I have with calculating amperage is the effect of grounding a circuit.  All of the intros to calculating use closed circuits as examples.  Any light that anyone can shed on this for me would be highly appreciated.

So, back to the circuit.  The voltage seen at pin 3 and 5 should be constant no matter what is happening with the blend knob as it is only passing through a single resistor on its way.  Is this correct?

[anotherjim]

R26¦R27 make the 4.5v reference for the op-amp +input pins.
Working op-amps, with no other input, will copy 4.5v to their output pins 1 & 7.
Feedback resistors R30, R35, will copy the output voltage to the -input pins 2 & 6.

If the blend control is changing the voltages, then disconnect it and see if you can get the 4.5v

I suspect you have a short circuit somewhere or incomplete track cut on the stripboard.

[Christoba]

Allright. I found the short.  Thanks to your help I was able to fix it and everything was groovy.  I changed the 56k resistor for a few and ended up settling on the 68k as the poster suggested. 

I then did my best to get it integrated with the compressor and packed into the enclosure.  Everything fired up and works properly except when the blend knob is on the clean side. The volume is low and there are no mid and high frequencies.  Son of a.....  Thoughts?

[Christoba]

Done and done!  It's all fixed and boxed up.  I now have a blended Ross style compressor.  Thank you all for the help.  I learned a lot.