• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Fisher 400 both channels distortion

dcgillespie said:
If the Negative terminal of the two voltage doubler caps were shorted together, the circuit would still rectify and double to some degree -- but it would have been very hard on the caps as they would have effectively been placed in series and then across the AC voltage supply, making them pass a significant amount of current -- no doubt more than they were rated to handle. This would absolutely explain the physical deterioration seen in these caps -- or more likely, one gave up the ghost, and then the other allowed the circuit to function as a doubler, but as a half-wave voltage doubler similar to the action I described earlier, where if one side of the doubler is compromised, the other side will still allow the circuit to effectively double -- just not as efficiently. That's why the unit still operated for years, with the one doubler cap likely blown out long, long ago. This also concurs with my original assessment that while this issue certainly needed to be fixed, it wasn't the heart of your problem, because it is based on the fact that too much negative voltage is being supplied to the output tubes. It is not surprising then that the original issue remained.

So now it would seem that we're into a two pipe problem. I was originally surprised that the transistors would have failed, but since they are the most obvious component whose failure would cause the problem at hand, that was the suggestion given. While I'm somewhat surprised that replacing the transistors didn't resolve the problem, I'm not entirely surprised that they are not the problem, either. As I said, all of the SS components used are way, way over-rated for the duties they handle. So now it's time for more troubleshooting: Using all necessary caution of course:

1.Remove all four output tubes (remembering location of course), and turn the unit upside down with the bottom cover removed.

2. Disconnect the 1M 1% resistor of the EFB™ Control Grid Regulator from the point where it connects to the screen grid buss. With the resistor disconnected from the screen buss, measure it's value. Next, with the resistor still disconnected:

3. Connect your voltmeter between the output of the EFB Control Grid Regulator (Emitter terminal of the Buffer Transistor -- the transistor that does not connect to the 1M resistor just mentioned), and ground.

4. Turn the unit on and note the reading on the voltmeter. Operating the unit with the output tubes removed will cause no harm to any circuits of the receiver.

5. Turn the unit off and unplug it. Leave the output tubes removed. The unit will self discharge as normal, but take much longer to do so.

Report back the measured value of the resistor and the voltage indication from the test.

Dave
Click to expand...

I removed the output tubes. I opened up the bottom though my efb is connected above the chassis behind the magic eye tube.
I disconnected the 1M 1% resistor from the grid regulator. the resistor is still reading 1M or 0.999M.
The emitter of the transistor not connected to the 1M resistor reads -2.5 to -2.6 volts
 
OK. So now we're into that vast gray troubleshooting area where something you think is true cannot be. Here are the fact as presented to be true:

1. The screen grid voltage supplied by the EFB™ Screen Grid Regulator is 285 vdc at pins 1&7, as given in post #4. This voltage is within a normal range, so the Screen Grid Regulator is operating correctly.

2. The output tube control grids are receiving way too much negative voltage from the EFB Control Grid Regulator at pins 2&6, implying that the Control Grid Regulator is malfunctioning. This is supported by the fact that the voltage at the output of the Buffer Transistor (Emitter terminal) is -46 vdc (given in post #6) -- which -- if you built the Control Grid Regulator according to the original schematic given -- is about 3X the normal amount appearing there.

It would be very helpful at this point to know if the output of the EFB Control Grid Regulator is directly supplying the four output tube control grids -- as was done in my original application and presentation -- or it has been modified to provide a somewhat greater output so as to be able to first supply an IBAM circuit before supplying its output to the output tube control grids. If this is the case, it becomes a new unknown element in the troubleshooting effort that must first be known so that the troubleshooting effort is based on actual circuit conditions.

3. You have now disconnected the 1M "sensing resistor" from the screen grid supply source -- where the output of the EFB Control Grid Regulator then dropped to (effectively) zero output -- exactly as it is supposed to do, which indicates that the Control Grid Regulator is operating correctly. Yet -- when the resistor is reconnected, the control grid bias voltage then becomes much too high?

If that's the case, it implies then that there are only one or two potential problems that can cause the symptoms you are describing. Either:

A. The 39K/10K Bias pot connected around the inverter (first) transistor has become compromised, so as to effectively become much larger in resistance than it should be, OR:

B. If you are using some sort of IBAM on the output of the EFB Control Grid Regulator, the IBAM circuit has become compromised so that it is effectively operating "wide open". This possibility can be checked by seeing if the adjustment of any of those controls changes the current flow through the tube controlled by each adjustment. If the controls work (if present), then that tends to eliminate any problem with the IBAM circuit. If not, then the IBAM circuit is suspect.

Either way, more information is needed for effective troubleshooting to continue as at this point, with the information presented, it is in conflict with itself........

Dave
 
Incidentally, I just finished modification of EFB values to provide higher negative voltage for lower distortion readings in different build. My EFB bias adjust trimmer was maxed out so I bumped 39k resistor (47k in my build) by additional 5k to be able to run tubes a little cooler and have the trimmer close to center. If OP’s EFB trimmer was set at the end of the travel on the lowest resistance end for correct bias operation, and the trimmer wiper loses contact, 10k of the full trimmer resistance is added to the circuit which will result in high negative output out of the EFB. I don’t know if it would add -10v though.
Would this sound plausible?
 
It wouldn't add that much extra voltage to the output of the regulator, but along those lines, one of my concerns was that if the 10K trimmer pot went completely open, that would cause the output at the emitter of the buffer transistor to go to nearly the the maximum output of the supply powering the EFB Control Grid Regulator.

Dave
 
dcgillespie said:
OK. So now we're into that vast gray troubleshooting area where something you think is true cannot be. Here are the fact as presented to be true:

1. The screen grid voltage supplied by the EFB™ Screen Grid Regulator is 285 vdc at pins 1&7, as given in post #4. This voltage is within a normal range, so the Screen Grid Regulator is operating correctly.

2. The output tube control grids are receiving way too much negative voltage from the EFB Control Grid Regulator at pins 2&6, implying that the Control Grid Regulator is malfunctioning. This is supported by the fact that the voltage at the output of the Buffer Transistor (Emitter terminal) is -46 vdc (given in post #6) -- which -- if you built the Control Grid Regulator according to the original schematic given -- is about 3X the normal amount appearing there.

It would be very helpful at this point to know if the output of the EFB Control Grid Regulator is directly supplying the four output tube control grids -- as was done in my original application and presentation -- or it has been modified to provide a somewhat greater output so as to be able to first supply an IBAM circuit before supplying its output to the output tube control grids. If this is the case, it becomes a new unknown element in the troubleshooting effort that must first be known so that the troubleshooting effort is based on actual circuit conditions.

3. You have now disconnected the 1M "sensing resistor" from the screen grid supply source -- where the output of the EFB Control Grid Regulator then dropped to (effectively) zero output -- exactly as it is supposed to do, which indicates that the Control Grid Regulator is operating correctly. Yet -- when the resistor is reconnected, the control grid bias voltage then becomes much too high?

If that's the case, it implies then that there are only one or two potential problems that can cause the symptoms you are describing. Either:

A. The 39K/10K Bias pot connected around the inverter (first) transistor has become compromised, so as to effectively become much larger in resistance than it should be, OR:

B. If you are using some sort of IBAM on the output of the EFB Control Grid Regulator, the IBAM circuit has become compromised so that it is effectively operating "wide open". This possibility can be checked by seeing if the adjustment of any of those controls changes the current flow through the tube controlled by each adjustment. If the controls work (if present), then that tends to eliminate any problem with the IBAM circuit. If not, then the IBAM circuit is suspect.

Either way, more information is needed for effective troubleshooting to continue as at this point, with the information presented, it is in conflict with itself........

Dave
Click to expand...

I do have an IBBA board as well, the modifications to the EFB to suit this were

"To adapt the EFB circuit for use with the IBBA, on the EFB board:

1. The 22K resistor becomes 18K (.25 watt is fine)

2. The 39K resistor becomes 75K (.25 watt is fine as well)"
 
not sure if this useful information the EFB trimmer reads 9.17k The IBBA trimers read
Bias left 1.5k
Bias right 1.18k
DC balance left 4.1k 3.8k 7.1k
DC Balance right 3.6k 4.7k 7.6k
 
Analyzing Dave’s answer, any open in the circuit between rectifiers and driver transistor base will cause a full negative output from the buffer (output) transistor. Bad solder joint?
 
That is correct: For any given value of screen grid voltage, increasing the value of the resistance between the Base and Collector of the Inverter transistor in the Control Grid Regulator causes a corresponding increase in negative voltage to appear at the Emitter output of the Buffer transistor. With the values I quoted for adopting the original design of the EFB Control Grid Regulator for use with an IBBA, it should produce about -21 to -24 vdc at the emitter terminal output, depending on where the 10K Bias control is set.

Since you indicate that you are getting (basically) double the amount of voltage at that point than should be there (from post #6), it indicates that while the inverter transistor is indeed inverting (as proven in post #21), it is not properly regulating the output voltage to the target voltage level the circuit is designed to produce. That strongly suggests that there is a fault in the wiring of the revised 75K resistor/10K bias pot connections, or where they connect to the inverter transistor in the build.

Dave
 
dcgillespie said:
That is correct: For any given value of screen grid voltage, increasing the value of the resistance between the Base and Collector of the Inverter transistor in the Control Grid Regulator causes a corresponding increase in negative voltage to appear at the Emitter output of the Buffer transistor. With the values I quoted for adopting the original design of the EFB Control Grid Regulator for use with an IBBA, it should produce about -21 to -24 vdc at the emitter terminal output, depending on where the 10K Bias control is set.

Since you indicate that you are getting (basically) double the amount of voltage at that point than should be there (from post #6), it indicates that while the inverter transistor is indeed inverting (as proven in post #21), it is not properly regulating the output voltage to the target voltage level the circuit is designed to produce. That strongly suggests that there is a fault in the wiring of the revised 75K resistor/10K bias pot connections, or where they connect to the inverter transistor in the build.

Dave
Click to expand...
Sorry one correction I looked at my board today and looked back in the forum I put 2 75k resistors in series to get 150k for some reason the 75 k didn't bring my old tubes into the correct range. Still poking around looking for bad solder joints etc..
 
I tried resoldering everything around the trimmer and transistors, reconnected the 1m resistor still no luck. the trimmers on the IBBA as well as the trimmer on the EFB seem to change the ma to the tubes but only about .01 still need an additional 18ma to be happy. Thanks for all the trouble shooting so far Dave and grindfix
 
150k in place of 75k sound like overcompensation for issue somewhere else. Have you confirmed all resistor values? Testing in circuit is not always possible but decoding color bands can help.
Transistor basing?
It’s possible that all of previous mistakes are now corrected but that additional 75k will result in a lot of additional negative output.
 
grindfix said:
150k in place of 75k sound like overcompensation for issue somewhere else. Have you confirmed all resistor values? Testing in circuit is not always possible but decoding color bands can help.
Transistor basing?
It’s possible that all of previous mistakes are now corrected but that additional 75k will result in a lot of additional negative output.
Click to expand...

grindfix you are on the right track. I was wondering about this myself, but nobody had ever questioned it.
I removed one 75k resistor and tubes went to 0.89ma but I can hear voices clearly again. . Thanks unfortunately color bands aren't available because these are dale 1% resistors kind of a pain to read and decode the writing if its not installed writing facing up. I might just make an order of all the resistors on the board and refresh. I'm still interested in finding out what could've changed that would cause this to work for so long. We know I had the doubler caps messed up from the start but did this cause an issue down the line or is there something else that is also awry with my board. The change from 75k to 150k was done back in 2016 as well when I built the IBBA board and this worked fine for my original mismatched set of tubes and has worked for 2 years with my closer matched set as well. The 75k value didn't get me in range then and it doesn't now. I haven't experienced a 1% resistor drifting much or failing other than exploding. could this be another resistor being the wrong value in the circuit? If 75k is the correct value in this place why didn't it work in the first place and why doesn't it work now that the doubler caps are corrected?
 
Last edited:
I was going to suggest the same thing. Could it be that in your original build, one of the 75K resistors was somehow shorted out, so that addition of the second resistor then made things work properly? Then, the short across the original resistor came undone, which then placed both resistors in the circuit. A 150K+ resistor in that location would result in an output of about -44 vdc (depending on the setting of the EFB bias control), which is basically the output you are getting from the regulator. No doubt you are on the right track now.........

Dave
 
dcgillespie said:
I was going to suggest the same thing. Could it be that in your original build, one of the 75K resistors was somehow shorted out, so that addition of the second resistor then made things work properly? Then, the short across the original resistor came undone, which then placed both resistors in the circuit. A 150K+ resistor in that location would result in an output of about -44 vdc (depending on the setting of the EFB bias control), which is basically the output you are getting from the regulator. No doubt you are on the right track now.........

Dave
Click to expand...
Not sure what you mean Dave how would a resistor be shorted and adding another in series correct this? They were tied together above the board so the only way they could be shorted would've shorted both. I tested both resistor and they are both reading 75k on the board and off. Also by removing the one now the value ma at the tubes is far too high again? Im just trying to wrap my head around this but its still not making sense
 
Last edited:
It doesn't feel right to play around with the value of the 75k resistor like I did originally to get things to work. If the guy who designs the circuit tells you the values they should be then that's what they should be who am I to question it. I don't want to go that route again it seems like more trouble down the road.... Like being told to clean your room and just putting all your mess in a closet.
 
OK. So we're back into the realm of something we think is true, cannot be. Here's some pointers to work from:

1. The actual negative voltage at the output tube control grids should be in the neighborhood of -15 to -16 vdc to produce the desired quiescent current, depending on the condition and brand of output tubes installed.

2. For an IBBA modification to work correctly, it would need about -22 to -24 vdc applied to it from the output of the EFB Control Grid regulator. However, you are getting about twice that amount applied to the IBBA circuit.

3. With the original EFB Control Grid Regulator modified to replace the 39K resistor with a 75K resistor, it would provide the required -22 to -24 vdc for the IBBA to operate properly.

4. But you instead had effectively 150K of resistance in place of the 75K -- which would effectively double the negative voltage output over that required produced by the 75K -- which is (basically) exactly the voltage you were getting at the output of the regulator -- yet you indicate that in this condition, the receiver operated properly. By all analysis of circuit theory, this simply cannot be.

5. Now, by removing one of the 75K resistors, leaving 75K in the circuit, the tubes are drawing far too much current.

If this is all correct, this again makes me wonder about the IBBA build itself.

I would do this:

A. Again, remove the output tubes.

B. With only the one 75K resistor and bias trim pot in place across the inverter Base and Collector terminals, verify the voltage at the Emitter of the Buffer transistor. It should be in the ballpark of #2 above.

C. If it is, then verify what the maximum voltage is that can be sent to the control grid terminals of the output tubes. It should be in the ball park of #1 above.

This will help give a better idea of where things are going south.

Let us know!

Dave
 
streetwise said:
Not sure what you mean Dave how would a resistor be shorted and adding another in series correct this? They were tied together above the board so the only way they could be shorted would've shorted both. I tested both resistor and they are both reading 75k on the board and off. Also by removing the one now the value ma at the tubes is far too high again? Im just trying to wrap my head around this but its still not making sense
Click to expand...
Shorted resistor in series with 75k will produce 75k total across the pair. In parallel it would be short circuit altogether.
 
Last edited:
I'm not sure i've ever seen a shorted resistor, though that doesn't really mean much. That seems like it would be an uncommon fail mode though. I suppose it could be a construction error, somehow they missed cutting the spiral in the carbon film?
 
streetwise said:
I removed one 75k resistor and tubes went to 0.89ma
Click to expand...
Was that per each individual tube? If so, it is very high but somewhat supports your -15.5v at the grid. What tubes are you using? Modern EH replacements do require higher negative voltage to bias properly. Since you have EFB, after the tubes warm up and readings stabilize, your bias target per tube should be 21mA each, measured across cathode current sensing resistors, most commonly 10ohm.
But first, I'd take Dave's suggestion to run the unit without output tubes until you get the bias voltage in the right ballpark. What adjustment range do you get on grids with trimmer on the EFB?

What is the negative voltage out of the rectifiers now? Your numbers in your first measurements were lower than what Dave specified in his drawing. -52v vs. -62v specified. This may be why your 75k value resistor position requires an increase to get higher voltage out of the buffer output to IBBA.
 
You must log in or register to reply here.
Back
Top Bottom