Installing Cathode Resistor - Questions

[quicksand]

I am about to install cathode resistors in a Fisher amp that uses 7868 OPTs.

Below is the schematic and a picture of the bottom. I put 4 red rectangles where I think the cathode resistor should be installed - ignore the highlighter.

The Cathode is pin 3

The 10 ohm resistors should go from pin 3 directly to ground from what I read.

There is a brown wire that runs from pin 3 to pin 4 on V7 - from there the brown wire continues to the other 3 pins on V8,9,10.

If I splice the brown wire before pin 3 on all tubes and install the resistor at the splice is this correct?

It seems that the only cathode resistor is needed but I am following what I have read.


Entire schematic: Click to enlarge - Ignore the highlighter.



---

A simple question about the bias pot adjustment. Currently V7 is red plating and I want to set the bias as low as possible at start.

Is adjusting/turning the pot all the way left the the lowest setting or right? I am guessing left but just want to make sure.

Thanks!

 

[amptramp]

Moving the pot to the left increases the negative grid bias. If V7 is redplating, check C21 for leakage. The bias point is common for all tubes, so the only thing that would cause such a large imbalance would be a bad capacitor and that would affect the other output tubes to a lesser degree.

 

[vfr800fiman]

The 10 ohm resistors should go from pin 3 directly to ground from what I read.

Not on this amp, look closely at the schematic. I have a Fisher 800b, and that is true for some of the other Fisher receivers, but not his amp.

There is a brown wire that runs from pin 3 to pin 4 on V7 - from there the brown wire continues to the other 3 pins on V8,9,10.

If I splice the brown wire before pin 3 on all tubes and install the resistor at the splice is this correct?

Yes, the resistor should be inserted between the existing wire to pin #3 and the actual tube pin.

It seems that the only cathode resistor is needed but I am following what I have read.

I assume you mean only one cathode resistor. Yes, you could use only one, but then you wouldn't be able to measure cathode current for each tube. You can do this already by measuring the voltage drop across R78 the 820 ohm resistor. voltage drop/resistance=current(in amps)

Amptramp is correct though, if only one tube is redplating, you may have other problems. Of course that one tube could be bad I suppose.
You should replace all those electrolytic caps while you're at it, like the one in the picture for example.

Glenn

 

[quicksand]

amptramp

I just replaced all the coupling caps - I guess it could have been bad out of the bag? Shame on me for trusting Illinois.

--

vfr800fiman

You are correct on both accounts.

I meant "it seems that only *one cathode resistor is needed..." good catch.

The picture is from before the bias electrolytic was replaced.


I have read about measuring the cathode current of each tube but not sure if there is any benefit since there is only one adjustment - currently.

The tubes test good on my emissions tester.

So I have an interesting day ahead of me if only one tube is red plating.


Thanks for the suggestions.

 

[WopOnTour]

This amp will not be easily converted to individual bias adjustments due to the way the bias supply circuit is shared with the preamp tube heaters. Not impossible, but difficult with a lot of trial and error and then there's no guarantee it will sound better.
You dont really need to add any cathode resistors as the load is basically present already back at the bias/heater network. (R76 in parallel with the series wired heaters and the bias adjustment string) and since your not able to change them individually anyways, all this will do is give you a method to identify a mismatch exists with no available work-around adjustment. (without affecting the heaters)
You might better off getting a carefully matched quad of output tubes and adjusting the cathode voltage to the recommended 38-40Volts.
JMO
WOT

 

[dsndblm]

The bias circuit for your Fisher is similar to my X-101-B, kind of a PITA. The bias voltage for the 7868s also powers filaments of V1, V2, V3, V4. It's a fine line to get the outputs biased correctly and have correct voltage for V1-V4. It's a must to have the 10ohm cathode resistors so you are able to calculate current for each 7868. I got some really good advice from a couple of the AK members and changed the values of the voltage divider resistors (R77 and R78 in your amp) to fine tune the bias and drive the 12AX7s correctly. I don't have the knowledge to give you advice on changing these resistor values. I bet someone else can if you are still having problems.

Changing the grid resistors from 330K to around 220K will help also. It's a good idea to change the coupling caps from .047uf to .1uf when swapping the grid resistors.

 

[quicksand]

WOT

It was just something I was considering - thanks for the heads up I probably will skip that thought.

--

dsndblm

I noticed that bias for the output was tied to the preamp stage but thought nothing of it. Was that biasing/ voltage divider resistor information posted in a thread on AK related to the x101b? Sounds interesting - I would like to read more.

Edit: is this the thread? http://www.audiokarma.org/forums/showthread.php?t=195556

I have made the grid resistor and coupling cap value changes already even thought the tubes I am using are old stock.

All good information.

 

[quicksand]

I am still not 100% clear where the cathode resistor goes on V7. Pins 3 and 4 are tied together on the socket - Pin 4 on V7 then connects to pin 3 on V8.

Option one: 10 ohm resistor between the brown wire and pin 3 on V7.

or

Option two: 10 ohm resistor between pin 4 on V7 to pin 3 on V8 - this option seems more correct since it is after the tube though it isn't the cathode.


I am can generate a sketch if the question isn't clear.

 

[dsndblm]

Quicksand

That is part of the thread on the 101-B bias circuit. If I remember right the rest of the info was sent to me in a pm by Jim McShane. I bought the output tubes and rebuild parts from him. Jim helped me out a lot with the rebuild. He's a great guy if you need parts and support. Sorry about the lack of info. That was almost a year and 3 projects ago......

On the cathode resistors, just connect one 10 ohm resistor in between pin 3 and the existing wire on pin 3. Just think of the 10 ohm resistor as an extension of the existing cathode connection wire.

On my fisher TA-600 I added the 10 ohm resistors up stream (not the correct term I'm sure) from the output socket pins. I traced the output cathode wires and soldered the resistors at a terminal strip. There was more room at the terminal strips. Same connection, just made it easier.

Hope this helps. Let us know how it turns out.

 

[FalconEddy]

I am still not 100% clear where the cathode resistor goes on V7. Pins 3 and 4 are tied together on the socket - Pin 4 on V7 then connects to pin 3 on V8.

Option one: 10 ohm resistor between the brown wire and pin 3 on V7.

or

Option two: 10 ohm resistor between pin 4 on V7 to pin 3 on V8 - this option seems more correct since it is after the tube though it isn't the cathode.


I am can generate a sketch if the question isn't clear.

Ok. . .first of all, DO NOT follow the information in the old thread you linked in one of your replies. That's a completely different amp, different pinouts on different tubes. Plus, (if I remember correctly) that thread was all messed with wrong information because mostly everyone was talking about 7591 output tubes. Wrong pin numbers were matched to wrong parts. . it got pretty messy.

Based on the partial schematic you've shown us, you have a Fisher X-100C, correct?

What, exactly, are you trying to accomplish by adding a 10 ohm resistor to the cathodes?? To my knowledge, lifting the cathodes from ground (technically, the common connection) to measure current draw is not a simple task on this amp. Is that what you're trying to do?? (Remember, all four cathodes are connected, correct??)

This amp is a bit difficult to properly bias since four of the 12AX7's are using the same feed voltage supply.

If V7 is the only tube redplating, have you checked the utility supply voltage feeding your amp? Is it a massive redplating, or just a bit at the plate seams?

. . Falcon

 

[quicksand]

Based on the partial schematic you've shown us, you have a Fisher X-100C, correct?

Hey Falcon, yes you are correct the amp is a Fisher X-100C and uses 7868s.

Ok. . .first of all, DO NOT follow the information in the old thread you linked in one of your replies...

This amp is a bit difficult to properly bias since four of the 12AX7's are using the same feed voltage supply.

Gotcha I wont follow that thread - however the reason I was looking for it was for the voltage divider information to help with biasing but the information was provided by PM and not posted on the boards so it is lost - bummer.

If V7 is the only tube redplating, have you checked the utility supply voltage feeding your amp? Is it a massive redplating, or just a bit at the plate seams?

That is where I am at now - I have a few different value resistors to run in series with R65 to drop the voltage back down to 430vdc in the power stage - currently it is at 480vdc.

What, exactly, are you trying to accomplish by adding a 10 ohm resistor to the cathodes?? To my knowledge, lifting the cathodes from ground (technically, the common connection) to measure current draw is not a simple task on this amp. Is that what you're trying to do?? (Remember, all four cathodes are connected, correct??)

Yes all four cathodes are tied together - the main purpose of adding the 10 ohm resistors is to protect the output transformer. I like the idea of protecting the amp any way possible to ensure enjoyment for many years to come since it was my father's.

I wasn't sure of the benefit of being able to measure individual voltages on this amp since they are tied together and there isn't a way to adjust each socket individually. WOT eluded to this in post #5.


BTW the WMFs sound great thanks for recommendation and help Falcon!

 

[quicksand]

That is part of the thread on the 101-B bias circuit. If I remember right the rest of the info was sent to me in a pm by Jim McShane. I bought the output tubes and rebuild parts from him. Jim helped me out a lot with the rebuild. He's a great guy if you need parts and support. Sorry about the lack of info. That was almost a year and 3 projects ago...

You wouldn't have the emails or PMs around by chance?

I have all the major parts on hand so purchasing through Jim is out - I have read many good things about his service though.

On the cathode resistors, just connect one 10 ohm resistor in between pin 3 and the existing wire on pin 3. Just think of the 10 ohm resistor as an extension of the existing cathode connection wire.

Like this?

The resistor would connect between the pin and brown wire but it was easier to sketch it spliced to ask if the first/ last resistor went before or after V7 or V10.

After doing this sketch I am now asking myself if the 10 ohm cathode resistors will really protect the output transformer...

 

[dsndblm]

I didn't save any email or pm's but made notes on a copy of my schematic. The bias circuit has a total resistance made up of the 5K bias pot, R77, R78 and R76. I Changed the values of the resistors in 1K increments. For example R78 would be changed to 1.8K, R77 to 2.3K. However during all this R76 will have to be adjusted to make sure filament voltage stays correct.

If I was you I wouldn't try this until someone who knows more about this stuff chimes in and ok's it. This procedure worked excellent on my 101-B but I can't tell you 100% if this applies to you.

Your plate voltage is way too high just as my 101-B was. I run my amp off a variac set at 115V. My a/c line voltage here is over 124 volts. Even with the total recap, bias mods plugging the 101-B straight into the wall would cause problems.

 

[quicksand]

dsndblm

Thanks for the information.

I have the same problem with my line voltage - it is at 124vac. I was planning on adding a higher value resistor in the power supply stage (R65) to get the voltage back closer to spec - similar to 1k increments that you did for the bias circuit.

Is there a benefit to using a variac to correct the line voltage other than not having to get the iron out?

 

[FalconEddy]

dsndblm

I have the same problem with my line voltage - it is at 124vac.

There ya go, a utility supply issue. Get it down to 115V, make some adjustments to those resistors, and the plate voltages will come back into line.

That will most likely resolve the red-plating issue.

. . Falcon

 

[quicksand]

It seems to be a split decision whether cathode resistors will be any benefit if they are added to this particular amp due to all four cathodes being connected.

I would like to use this amp on a daily basis and have it around for years to come since it was my father's.

That said I would like to add them if it will protect the amp from "catastrophic failure" which I have read about.

 

[WopOnTour]

Protecting the amp from "catastophic failure" would be best served by a careful and thourough restoration of passive elements in the power supply and all pathways of high voltage, then replacing the power tubes with a carefully matched, qualified, and tested quad of NEW tubes. You can also make some changes to the resistors within the power supply (R65 thru 69) as suggested, to bring the plate voltages more in line with the amp's original specificatiion and compensate for your AC supply issues.

If it makes you feel better you can still place a 1/4 or 1/2 watt resistor of almost any lower value (1 or 10 ohm?) in series between the 4 cathodes and the parallel ground network formed by the preamp tube heaters and R77 thru R79. However I suspect some of these resistors are likely already 1/2 watt and are essentially accomplishing the same thing by creating an electrical " shear point" should cathode current "runaway" due to a loss of bias or a bad tube -so there' really nothing to be gained IMHO.Plus any voltage measurement across this resistor would be reflecting the combined current from all 4 tubes rendering it essentially useless from an adjustment standpoint (current/4 would be an unreliable methodology here IMO but others still use it, see calculator link below)

The primary purpose for adding 1 or 10-ohm resistors is to simplify the conversion of cathode current from a voltage drop measurement via Ohms law. This is especially useful in amplifiers using cathodes that are directly grounded and/or have the facility to adjust the bias individually by observing the resulting changs in cathode current flow. In this amp it would probably be easier to check and adjust the the bias using a direct to cathode current measuring device such as the Weber Bias-Rite, Groove-Tube Bias Tool, or Tube Depot's Bias King ,combined with the aformentioned "matched" quad of power tubes. (I personally use a DIY "dual" tool similar to the GrooveTube and Weber devices used with a regular DVOM) If you had a supply of new or NOS tubes, you could also use these tools to create your own "best match" pairs.
Otherwise, get thee to Jim McShane!
JMO
WopOnTour

EDIT: Here's a handy "bias calculator" page from the Weber web-site
For instance using the LAST calculator on tha page- measuring 1.75 Volts across a single 10 ohm resistor, on your amp with 4 tubes and 430V on the plates yields approx. 42mA per tube or 18 watts. Remember guitar amps generally run a lot colder than hi-fi as they are not as concerned with sound-quality, distortion and output power.

 

[quicksand]

That is a great bit of information WopOnTour!

I have been reading here on a regular basis and you summed up a lot of information in one post and then some!

I will reread the entire thread a few more times to fully digest it.

Thanks for all those who have contributed - I will report back after the work is completed.

Cheers!

 

[dsndblm]

WopOnTour: Great information! I knew that someone here could set this right.

In my limited experience I found the 10 ohm resistors to be very useful in this type of circuit. I'm still able to monitor current draw of each tube and match them in the amp. My X-101-B service manual calls for +44v as bias for the outputs. My Jim McShane matched quad of EH 7591's all measure within a milliamp or so of each other with this bias. My vintage 7591's are another story. I have 11 7591's and it took all 11 to find 4 that were within 5 milliamps of each other.

I also had a vintage 7591 fail in the 101-B. I had the amp running for a few minutes and heard a fairly loud pop, then lower volume in the left channel. The vintage 7591 shorted and the little 1/4 watt resistor gave up quickly even with this cathode circuit arrangement.

Quicksand. This is a tough amp as a first rebuild. I was exactly in your shoes, first rebuild, tough circuit but had some great help. Jim McShane must have sent to 10 emails until I understood well enough to make it right.

As suggested, replace multisection electrolytic and coupling caps. Its a must for a reliable daily driver. I cut open the old electrolytics out of my 101-B and they were shot. They had both swelled on one side as well. After the rebuild my Fisher plays for hours at a time without concern of something frying and letting out all the magic smoke

 

[WopOnTour]

In my limited experience I found the 10 ohm resistors to be very useful in this type of circuit. I'm still able to monitor current draw of each tube and match them in the amp.

Strictly for matching purposes, sure!
(providing your grid currents are pretty consistant- doubtful with a mish-mash of NOS tubes though)
That's really the advantage of the "direct to cathode" tools as you're not also measuring grid current (unlike the cathode Vdrop method)

So four 10-ohm resistors in parallel would only represent adding 2.5 ohms of resistance to the circuit. But if you're going to do that you might as well use 1 ohm and directly read your voltage results in milliamps. (Did you also add 4 top-side test points to facilitate matching?)
But for adjusting bias with a pre-matched quad- and since it's not individually adjustable you might as well use just R76 and do the math.(Vdrop/500) as it's already there. If you get a matched quad from a reliable seller, you'll not really need individual current numbers.
Pretty much agree with everything else you said though...
:thmbsp:
WOT