G-4500 Volume Wierdness

[Overundr1]

Another way is to run down to RadioShack, if you still have one and pick up a pair of 8ohm 20watt cement resistors which will be ok as long as you dont max out the output for long. Hook them across the speaker teminals, also at the same time hook your dmm across the 8ohm leads, set the meter scale to ac and measure how fast the voltage ramps up. Watch to see where the volume control position is when you hit the 1vac level and if you are at 0vac when the control is fully counter clockwise. Just for reference, driving a sine wave your reciever should reach approximatly 18vac at clip. It is not good practice to drive an amplifier to clipping with square waves, rather use a sine wave. Square waves usually are used to measure other amplifier parameters which we do not want to drag into the mix at this time.
You can also use a scope for this however since we are looking for voltage rise the dmm method (unless your scope includes a built in dmm display function) is more relevant to what we are trying to determine.

 

[Jeff Warner]

Wow! Lots of questions and thoughts to digest.

I did eventually hook up my scope to both the pot and just the cable to see what I could see and what I saw was too much going on. I may have had the synth set for square wave but all I saw was a bunch of overlapping sine waves that peaked on the left. I'm wondering if I'm picking up AC interference on the 50' RCA run. I could try taking a picture. I'll need more info on what I'm looking for when trying to identify clipping or output jumping to midpoint on the trace, but for now, I think I'm not sending a very clean signal. I have already purchased a function generator so I may just wait for that before going too much further. I'm going to have to seriously rearrange my bench. The big CA,TU, BA 2000 tower in the middle has got to go!

While I wait for the signal generator, I'll work on getting some resistors and your dmm test.

I can tell you that on my meter the voltage increase measured on the pot was not linear. It took a while to start increasing and really didn't start to dramatically increase until I was past the 5 position on the knob (12 o'clock). Seemed logarithmic for certain. And yes, I did confirm that the part number is correct. Also, I forgot to mention that the aggressive cleaning of the pot while it was out seemed to help quite a bit with the imbalance left to right. Strange that the cleaning actually increased the resistance slightly on both sides.

Additional damage is certainly a concern. I've looked quite extensively at both boards. The impact did not impact them directly. It really didn't get the transformer or power supply board either as it was right in the middle in an open section of the left side where the transformer mounts. The worst thing I've seen so far was a little chip out of a board mounting tab on the power supply board where it popped out. Hardly noticeable. I'll go ahead and get after it again with the magnifying light to see what I might see.

Watch to see where the volume control position is when you hit the 1vac level and if you are at 0vac when the control is fully counter clockwise.

This is very doable once I get the resistors but what output voltage do I want to start with on the RCA cable if I'm looking for where I get 1vac on the pot. Wouldn't it be important to have that known first since I can adjust it with my mixer? (I just reread your earlier post where you mention 750mvac. I'm guessing that's what you were talking about. I can dial that in for sure. Please correct me if I'm wrong about this assumption.) I'll go ahead and switch to a sine wave even though that's what I'm seeing on the scope anyway and try to take further steps to clean up the signal. Oh and I should state that I checked the DC Offset a few days ago and had Left side 4mV, right side 0mV so that's pretty darn good and low.

 

[Jeff Warner]

Signal from end of RCA cable at 780mvac, A above C solo clarinet sine. At least 3 of these waves are present when the probe isn't even connected. Probe set at 1x.

 

[Jeff Warner]

Inspected the boards on both sides and these two shots are the only things I found that jumped out at me. Their both on F-2711 and you can see this first one is right next to one of the bias VRs. The cap is C25.

Here's the second one which was quite a glob. I carefully cut it and pulled out as much as I could. What's the proper way to remove this glue anyway? The Diodes are D01 and D02 and the bigger cap is C45.


These two areas are on either side of the large caps.

 

[Jeff Warner]

Continuing my conversation with myself or repair diary. I finally decided to look at the obvious which I should have done long ago. Checked the resistance on the Balance pot and saw 1000 ohms on one side and 1200 on the other. Okay, so I thought that might explain some things so I cleaned it. Then I was seeing basically 0 ohms on both sides no matter how I turned it. Well that sure didn't make any sense to me. It is supposed to be a resistor right? So, after being sure it was dry, I fired it back up and found that it was indeed still working. Turned it off and checked it again and saw 3900 ohms on one side and 2200 ohms on the other. So all I can figure is there must be some caps involved in feeding the function of this pot and probably some resistors that influence what I see from those caps. I don't understand it but at least I can see that there is an imbalance and that at least supports the idea that there is a problem.

On a positive note I finished rearranging my bench and got the tower out of the middle so everything is more accessible and I have more room to work and no more cords hanging in my face.

 

[Overundr1]

Scope picture looks like a lot of background noise, for what we are working on its a not to be concerned issue. Another member just posted about a G4500 he just acquired so I asked in his thread to comment. Amplifier sections only amplify the level of signal they receive, in other words the volume control per se has no bearing on the amp itself other than attenuating the input signal. A real puzzler here, not that folks don't want to chime in, rather that you are in sort of uncharted territory. If there was a way to track down the +/- wires that feed the amp from the preamp section I would at this point have separated them and measured the ac rise with a constant sine wave input to the pre amp verifying the rise slope is not excessively rapid. If it was I would be really looking hard at the volume control being of the correct value. If that volume control is not obtainable and is the culprit you could attenuate the input level by adding a resistor to those input wires to tame down the level somewhat.

 

[Hyperion]

These two areas are on either side of the large caps.

To explain.
You are looking at Sansui 'glue' - that's the brown 'goop' - nasty stuff Sansui used to secure large components for shipping. The bad new is it goes corrosive AND conductive with age, and should all be removed.

But don't worry about that now - you have bigger fish to fry.

 

[Jeff Warner]

The bad new is it goes corrosive AND conductive with age

Right and that is why I was wondering if in being conductive it might be causing the volume issues. But with what Overundr1 just said, it looks like I should be looking at the input side not the amp side. Still I think I'll take a whack at cleaning up some of it where it is contacting leads.

 

[Jeff Warner]

If there was a way to track down the +/- wires that feed the amp from the preamp section I would at this point have separated them and measured the ac rise with a constant sine wave input to the pre amp verifying the rise slope is not excessively rapid. If it was I would be really looking hard at the volume control being of the correct value.

The signal generator will be here Friday so maybe I can get somewhere with that. As for the vol. pot. When tested off the board it looks great. 260k or so on both sides and a very slow rise until it starts to eventually take off for zero or vice versa.

What do those resistance numbers on the balance control tell you?

 

[Jeff Warner]

All of the wires from the pre-amp (two sets of three) lead to the Loudness board where the Balance is also located and then directly into the volume pot on the amp board. Portion of schematic as seen above from top down.

 

[Overundr1]

Can you provide a clear picture around IC01 and IC02 input differential pair?

 

[Jeff Warner]

 

[Jeff Warner]

I greatly appreciate your interest and assistance. Keep the requests coming. Since it seems like the problem is coming from the input side, I'll start tracing back that way and checking things where I can.

 

[Overundr1]

Those are the tuner ic's, was looking for the two 2sa798 input differential ic's. Looking at the schematics it looks like there is no easy way to modify the dual fet input stage so we will have to work on the connections between the balance pot and volume pot as we dig deeper.

 

[Jeff Warner]

Okay, so I shot the wrong ones. But those were IC01 and 02 on board F-2924. Should I be looking somewhere else or are you saying they don't exist. I'm also thinking that since the balance pot is giving me the weird resistance numbers after being on, that the problem exists before that and that would be F-2924 right? Assuming it's not the Loudness board to begin with and I'm not assuming that.

 

[Jeff Warner]

I only see one other IC and that's IC03 and it does appear to be associated with the tuning capacitor.

 

[Jeff Warner]

Okay so this is interesting. The backside of those two ICs. IC01 has a capacitor soldered onto it and three of the pins were heavily gooped in flux.



Gooped up pins were the last three at the far left top on IC01.



Here's the cap from the other side and the three ex-dirty pins are the bottom three at the far right next to the cap. I believe this is the #1 pin end of the IC.

 

[Overundr1]

Those appear to be mounted to the backside of the tuner board, quite common with Sansui products. If you look towards the back of the reciever you should see two 5 legged critters, those are dual inline differential pairs. Only reason I wanted to see a picture of that area was to determine a safe point to detach the input signal from the main power amp. Have never been in a small G, only the big ones so visual reference will help.

 

[Jeff Warner]

Okay, I'm assuming we're talking about the tuner / pre-amp board but I'll take a couple shots of the whole thing and look for what you're talking about. Thanks! Should have probably started with this to begin with because I know these units are not well known.

 

[Jeff Warner]

Okay, whole G first. Tuner / inputs/ pre-amp on the right, amp on the left.



Tuner / Inputs / Pre-amp side



Amp side



Glue problem areas are on left and right side of main caps. In the middle, left of the small diodes next to the smaller caps (under the red wire) and between the lower Bias VR and smaller black cap on the right.

I could not find any 5-legged beasties.