Fisher 500c Channel Imbalance?

baconbadge

Active Member
Hey folks,
I was getting a slight channel imbalance on my 500c that I finally managed to track down with my scope. The culprit appears to be C31 & C33, which are the first coupling caps before V15. I get perfect overlapping waveforms going into the caps and about a 20mV difference coming out. The Fisher parts list calls for “Ceramic, 24pf, 5%, N150, 1000V”. Mouser has this 27pf ceramic, but it’s a N1500 dialectric. Is that a good replacement? Anything else recommended?

http://www.mouser.com/ProductDetail...=sGAEpiMZZMt1mVBmZSXTPI4RnmSwHRBuByu5ipyWAuo=


kLoqRux.jpg


Thanks!
-Keith
 
Ken - Would you mind explaining how you used a scope to track down the problem? I've been wanting to learn how to use my scope to do exactly that. Thanks, Thorne
 
Keith -- depending on what frequency you were using as a test signal, it is doubtful that those caps are the problem. I'm not questioning your reading, but may I give you an alternate analysis?

The caps you are questioning are not coupling caps, but called "compensation" caps, and are primarily effective at frequencies over say 5-8 kHz, acting to counteract Miller in V15 and also some of the shielded cable used throughout the receiver (due to the capacitance they add to the circuit). If you are intent on replacing them, its value is best determined using as high as a 10 kHz square wave, with your scope attached to the reverb jumper for the channel in question to monitor the signal at that point. While the tone controls are not in the circuit at this point, be sure they are centered none the less to ensure there is no abnormal load on V15. C31 and C33 are then adjusted for best square waveform without overshoot. If you do adjust them, don't be surprised if it takes a different value than specified, and a different value in each channel as well for best waveform.

Down in the real world however, where most gain measurements and troubleshooting use 1 kHz sine waves, these caps have a very limited effect. In this frequency range, more than likely what you're really fighting is an imbalance in value between (using left channel component designations) R36, R40, and R41, and their counterparts in the right channel. These components set (among other things) the basic gain for each section of V15, and is most likely where an imbalance between the stages is occurring.

All of this assumes of course that both sections of V15 are good, and reasonably matched. As a base then before you start, and using the reverb jumpers as a measuring point, when an 250 mV 1 kHz sinewave is applied to the Aux input of both channels, what is the difference in output level you are measuring at the reverb jumpers?

Dave
 
Keith -- depending on what frequency you were using as a test signal, it is doubtful that those caps are the problem. I'm not questioning your reading, but may I give you an alternate analysis?

The caps you are questioning are not coupling caps, but called "compensation" caps, and are primarily effective at frequencies over say 5-8 kHz, acting to counteract Miller in V15 and also some of the shielded cable used throughout the receiver (due to the capacitance they add to the circuit). If you are intent on replacing them, its value is best determined using as high as a 10 kHz square wave, with your scope attached to the reverb jumper for the channel in question to monitor the signal at that point. While the tone controls are not in the circuit at this point, be sure they are centered none the less to ensure there is no abnormal load on V15. C31 and C33 are then adjusted for best square waveform without overshoot. If you do adjust them, don't be surprised if it takes a different value than specified, and a different value in each channel as well for best waveform.

Down in the real world however, where most gain measurements and troubleshooting use 1 kHz sine waves, these caps have a very limited effect. In this frequency range, more than likely what you're really fighting is an imbalance in value between (using left channel component designations) R36, R40, and R41, and their counterparts in the right channel. These components set (among other things) the basic gain for each section of V15, and is most likely where an imbalance between the stages is occurring.

All of this assumes of course that both sections of V15 are good, and reasonably matched. As a base then before you start, and using the reverb jumpers as a measuring point, when an 250 mV 1 kHz sinewave is applied to the Aux input of both channels, what is the difference in output level you are measuring at the reverb jumpers?

Dave

Ah, interesting. Having two "coupling" caps there did seem a little odd! I'll check my waveforms at the reverb jumpers and report back.
-k
 
Ken - Would you mind explaining how you used a scope to track down the problem? I've been wanting to learn how to use my scope to do exactly that. Thanks, Thorne

Hey Thorne,

First I plugged the Fisher into an isolation transformer and attached my L&R channel 8ohm dummy loads to the speaker jacks. I let everything warm up and took bias readings of my output tubes to verify that they were all balanced and set where I wanted. Then I fed a 1khz sine wave @ 250mV (per the Fisher specs) from my signal generator into the AUX input on the Fisher.
I set the volume at my normal listening level and I took my scope probes and attached them at the RCA jack inputs of the Fisher (scope channel 1 to the L channel input & scope channel 2 to the R channel input). I brought both channels up on the scope, verified that I was getting a clean wave, and superimposed them on top of each other to make sure that what was going into the Fisher was balanced. Everything looked hunky dory, so I took my scope probes and then attached them across the output jacks at the speaker terminal (still ch 1 to L & ch 2 to R). I looked at the scope and saw that the L channel sine waveform had a higher amplitude than the R. So I noted the amplitude difference and began to work my way backwards through the circuit- usually going from plate to grid of each tube. I thought maybe the balance pot was to blame, but I checked the signal coming off the switch and they were still equal up to that first cap I was talking about. It was after that cap that the signal jumped up, and the volume pot is further down the line, so that's how I pinpointed where the discrepancy was happening.
The volume difference wasn't really audible as far as I could tell, but I was checking the input & output waveform just for "fun" and discovered it.
Hope that helps!
-Keith
 
Thanks so much, Keith.

You put the scope probes on the INPUT jacks??? I would think you'd put them on an OUTPUT lug so that you get a picture of the signal being sent to the speakers. I would think the INPUT jacks would be where you want your 1khz signal. I don't know why my thinking is backwards.

When you worked your way backwards, did you continue to compare sine waves for differences until you got to the point where they were NOT different? And if so, then you knew that the next component in line was causing the imbalance, right?

Sorry if my "novice-ness" is showing. I am hungry to learn. Thanks again. Thorne
 
Keith -- depending on what frequency you were using as a test signal, it is doubtful that those caps are the problem. I'm not questioning your reading, but may I give you an alternate analysis?

The caps you are questioning are not coupling caps, but called "compensation" caps, and are primarily effective at frequencies over say 5-8 kHz, acting to counteract Miller in V15 and also some of the shielded cable used throughout the receiver (due to the capacitance they add to the circuit). If you are intent on replacing them, its value is best determined using as high as a 10 kHz square wave, with your scope attached to the reverb jumper for the channel in question to monitor the signal at that point. While the tone controls are not in the circuit at this point, be sure they are centered none the less to ensure there is no abnormal load on V15. C31 and C33 are then adjusted for best square waveform without overshoot. If you do adjust them, don't be surprised if it takes a different value than specified, and a different value in each channel as well for best waveform.

Down in the real world however, where most gain measurements and troubleshooting use 1 kHz sine waves, these caps have a very limited effect. In this frequency range, more than likely what you're really fighting is an imbalance in value between (using left channel component designations) R36, R40, and R41, and their counterparts in the right channel. These components set (among other things) the basic gain for each section of V15, and is most likely where an imbalance between the stages is occurring.

All of this assumes of course that both sections of V15 are good, and reasonably matched. As a base then before you start, and using the reverb jumpers as a measuring point, when an 250 mV 1 kHz sinewave is applied to the Aux input of both channels, what is the difference in output level you are measuring at the reverb jumpers?

Dave

Ok, finally had time to take some more measurements and pics!

Here's a pic of my waveforms at the reverb jumpers... note that we're at .5 Volts/Div:

0yH22ch.jpg



Here we are equal coming into the AUX input jacks... still at .5 V/DIV:

wLEe5Ns.jpg



And here we are directly after C31 & 33... note that now we're at 20mV/Div:

QaIOpSz.jpg



My measured values for
R38/R40: 486K/2.8M
R39/R47: 483K/2.8M

Forgot to measure R41/R42 before I buttoned her back up.

Dave, you mentioned R36? How does that play into it? Assuming you meant R38?
I'm thinking that the discrepancy is so small that it's negligible?

-Keith

(edited for correct values)
 
Last edited:
Thanks so much, Keith.

You put the scope probes on the INPUT jacks??? I would think you'd put them on an OUTPUT lug so that you get a picture of the signal being sent to the speakers. I would think the INPUT jacks would be where you want your 1khz signal. I don't know why my thinking is backwards.

When you worked your way backwards, did you continue to compare sine waves for differences until you got to the point where they were NOT different? And if so, then you knew that the next component in line was causing the imbalance, right?

Sorry if my "novice-ness" is showing. I am hungry to learn. Thanks again. Thorne

Hey Thorne,

I look at the waveforms on the input jacks to get a baseline to compare against what I see along the signal path in the receiver. So if the waveform looks distorted or has some kind of noise on it, I know that it's actually happening inside the receiver and not coming from my signal generator. Both the scope probes AND the signal generator are "attached" to the input jacks to see the waveform entering the receiver. Then you move the scope probes together along the circuit attaching them to the same points in the L & R channel. Yes, I basically worked backwards until I found where the waveforms matched. The point I found was actually a little tricky, as Dave pointed out, because there's a few components there that are influencing the signal.
Hey, I'm learning as I go too!

-k
 
Keith -- The difference is rather small so ultimately it's nothing to really stay awake at night over. Regarding your data:

1. Yes, R38. My schematic is small enough that sometimes the numbers look similar.

2. I assume your R40 and R47 are 2.8 M (not K)?

3. I assume your using a frequency of 1 kHz?

4. Forgot to measure R41 and R42 (not 40/41)?

One other test to try is again measuring at the Reverb jumpers, but turn the tape monitor switch on -- that will remove most of the following stage load to remove it as a variable.

When trying to make channels identical (it's tedious to make all the stages cross-identical between the channels), there's nothing wrong with bridging (in this case) the appropriate R38 or R39 with a high value resistance to make the gain equal between the two channels for those stage. Differences in tube sections can cause slight differences as well, so a different tube may help. Ultimately, the reason that there is a difference in level after C31 and C33 is because that is the point at which the NFB around these stages is introduced. That's why the balance of the various resistors mentioned between the two channels will have a definite effect of the signal level observed at this point.

Dave
 
Ack! Yup, I meant 2.8M AND R41/R42.
Ugh, never post to AK when rushed and running out the door! And yes, with a 1k sinewave. She's sounding great and my bias is spot-on, so I'm just gonna leave her be and enjoy! Thanks for all the help, as always!
-k

(Edited previous post for correct values)
 
Last edited:
Thanks so much, Keith.

You put the scope probes on the INPUT jacks??? I would think you'd put them on an OUTPUT lug so that you get a picture of the signal being sent to the speakers. I would think the INPUT jacks would be where you want your 1khz signal. I don't know why my thinking is backwards.

When you worked your way backwards, did you continue to compare sine waves for differences until you got to the point where they were NOT different? And if so, then you knew that the next component in line was causing the imbalance, right?

Sorry if my "novice-ness" is showing. I am hungry to learn. Thanks again. Thorne

Thorne,
Mark at Blueglow did a great video on using the scope for troubleshooting that I bet would be helpful to watch:

 
Thanks, bacon. I just finished watching the entire video and I learned quite a bit about WHERE to measure and WHAT to measure, how to use the scope for this one problem and some basic troubleshooting questions to ask. Very educational for me. I appreciate it. I'm going to look for other videos by him. Thorne

PS - One thing I don't yet understand... How can current (and signal?) travel in both directions? In the video he says that something (current or signal) travels IN to and OUT of pins 3 (cathode) and 6 (plate) on the preamp tubes. Am I misunderstanding?
 
Last edited:
Thanks, bacon. I just finished watching the entire video and I learned quite a bit about WHERE to measure and WHAT to measure, how to use the scope for this one problem and some basic troubleshooting questions to ask. Very educational for me. I appreciate it. I'm going to look for other videos by him. Thorne

PS - One thing I don't yet understand... How can current (and signal?) travel in both directions? In the video he says that something (current or signal) travels IN to and OUT of pins 3 (cathode) and 6 (plate) on the preamp tubes. Am I misunderstanding?

Hey Thorne, I haven't watched that video for a while so not exactly sure what he's referring to. Maybe he's talking about AC & DC both being present in the same wire? Dunno!
-k
 
bacon - He says, I thought, that something travels OUT the plate and to the power transformer. Does ANYTHING go in that direction from ANY tube? I'm under the impression that everything from the PT goes OUT through the secondary, not IN back to it. Maybe I did misunderstand. It's possible.

While I have your attention, if I may... I want to add 10 ohm resistors to the cathode pins of the power tubes so I can read voltage there when I'm biasing. The cathode pin on a 7189 is pin 3. All the tubes' pin 3's are tied together. Do I leave the ties alone and just insert the resistor from pin 3 to ground? Is there something else I need to do to compensate for the fact that I'm adding resistance so somewhere else needs to get a reduced resistance? I seem to remember that with my 500-C tubes, I had to change 2 resistors from 330 to 220 ohms when I added the cathode resistors. Maybe that was for an IBAM board. In my notes I left out that part.

Thorne
 
bacon - He says, I thought, that something travels OUT the plate and to the power transformer. Does ANYTHING go in that direction from ANY tube? I'm under the impression that everything from the PT goes OUT through the secondary, not IN back to it. Maybe I did misunderstand. It's possible.

While I have your attention, if I may... I want to add 10 ohm resistors to the cathode pins of the power tubes so I can read voltage there when I'm biasing. The cathode pin on a 7189 is pin 3. All the tubes' pin 3's are tied together. Do I leave the ties alone and just insert the resistor from pin 3 to ground? Is there something else I need to do to compensate for the fact that I'm adding resistance so somewhere else needs to get a reduced resistance? I seem to remember that with my 500-C tubes, I had to change 2 resistors from 330 to 220 ohms when I added the cathode resistors. Maybe that was for an IBAM board. In my notes I left out that part.

Thorne

Which Fisher model are you referring to? The 500c uses 7591 output tubes with pin 5 bringing the cathode to ground.
 
I'm talking about 7189 tubes, not my 500-C 7591 tubes. I was referencing my 7591 tubes as a job I did a while ago. I don't remember that the pin 5's were all connected like the pin 3's are on my 7189's.
 
I'm talking about 7189 tubes, not my 500-C 7591 tubes. I was referencing my 7591 tubes as a job I did a while ago. I don't remember that the pin 5's were all connected like the pin 3's are on my 7189's.

I'm probably not the best person to ask. I think it all depends on the circuit.
 
Cut the lead from pin 3 and insert the 10ohm resistor in SERIES btwn the tube and the lead. DO NOT GROUND THE Circuit. The 10ohm resistors will NOT affect the circuit. Testing is accomplished by metering ACROSS THE RESISTOR, not to ground like on a 7591/7868 fixed bias.
 
Thanks, Larry. You saved a potential short circuit. But yeah, wiring to ground felt wrong to me because the cathode pin wasn't wired to ground.
 
Back
Top Bottom