Repair question for Type D High-Gain Differential plug-in


Daniel Koller
 

Howdy Folks,

  I'm getting around to repairing my defective plug ins, and I have a question about the type D, for anyone who may be familiar with the circuit.  

  In the output circuit, there is a pair of type 5879 tubes, V3704 and V3604, that are a matched pair.  How close in gain do they have to be to work in this circuit?

  I have three tubes, measured on my Hickok tester as follows:

     V3604 1210 u-Mhos
     V3704  560 u-Mhos
     Spare   975 u-Mhos

  The tube tester specs give 630 u-Mhos as the minimum transconductance for a good tube so V3704 was replaced with the spare.  But I still can't balance the 
amplifier at the most sensitive settings with the grids of the tubes shorted together, as described in the service instructions.  I presume that is because the gains are
so off.  Is this so, or could it be something else?  Is transconductance at a fixed bias enough info to get me a reasonably close pair of tubes without a curve tracer?

  Thanks for the advice.

  Dan


tubesnthings@aol.com <tubesnthings@...>
 

Dan,
I have not worked on my type D units, so without looking at the circuit but in general;
Differential amps do rely on closely matched sections. The matching criterium is not transconductance of the tubes as seen by a tube tester, but plate or cathode current in the plug-in's circuit(s). An easy way to determine the tubes' contribution to amp imbalance is to swap them with each other; if the imbalance "follows" the tubes you've isolated that problem. I suspect you'll need to shop for tubes…

Good luck.
Bernd Schroder

Sent from my Verizon Wireless Droid


-----Original message-----
From: Daniel Koller <kaboomdk@...>
To:
"TekScopes@..." <TekScopes@...>
Sent:
Sat, May 18, 2013 13:00:59 GMT+00:00
Subject:
[TekScopes] Repair question for Type D High-Gain Differential plug-in

 

Howdy Folks,

  I'm getting around to repairing my defective plug ins, and I have a question about the type D, for anyone who may be familiar with the circuit.  

  In the output circuit, there is a pair of type 5879 tubes, V3704 and V3604, that are a matched pair.  How close in gain do they have to be to work in this circuit?

  I have three tubes, measured on my Hickok tester as follows:

     V3604 1210 u-Mhos
     V3704  560 u-Mhos
     Spare   975 u-Mhos

  The tube tester specs give 630 u-Mhos as the minimum transconductance for a good tube so V3704 was replaced with the spare.  But I still can't balance the 
amplifier at the most sensitive settings with the grids of the tubes shorted together, as described in the service instructions.  I presume that is because the gains are
so off.  Is this so, or could it be something else?  Is transconductance at a fixed bias enough info to get me a reasonably close pair of tubes without a curve tracer?

  Thanks for the advice.

  Dan


Daniel Koller
 

Howdy Bernd,

   I've tried swapping tubes.  It's hard to do in this case since it requires both the plate balance and vertical controls to be moved to near opposite ends of their ranges.  I am too far out of balance.

   Unfortunately, to buy a pair of matching tubes, I can only go by what is offered - i.e. what is commonly measured.  Given that most people on e-bay seem to report only Gm (if that), can I use it as a rough guide to "match" the tubes?   If I buy a pair of tubes from identical manufacturer, presumably new, with the same Gm, are they likely to have the same plate resistance and gain characteristics?   Unfortunately, I think this is the only choice I have - buy a bunch of tubes, and match them myself, in-circuit.  

   My question then is one of manufacturing.   How far off were the tube parameters, say from GE or RCA, from tube to tube?  1%?  10%, 50%??

  Thanks,

  Dan



From: "tubesnthings@..."
To: TekScopes@...
Sent: Saturday, May 18, 2013 1:32 PM
Subject: Re: [TekScopes] Repair question for Type D High-Gain Differential plug-in

 
Dan,
I have not worked on my type D units, so without looking at the circuit but in general;
Differential amps do rely on closely matched sections. The matching criterium is not transconductance of the tubes as seen by a tube tester, but plate or cathode current in the plug-in's circuit(s). An easy way to determine the tubes' contribution to amp imbalance is to swap them with each other; if the imbalance "follows" the tubes you've isolated that problem. I suspect you'll need to shop for tubes…

Good luck.
Bernd Schroder

Sent from my Verizon Wireless Droid


-----Original message-----
From: Daniel Koller
To:
"TekScopes@..."
Sent:
Sat, May 18, 2013 13:00:59 GMT+00:00
Subject:
[TekScopes] Repair question for Type D High-Gain Differential plug-in

 
Howdy Folks,

  I'm getting around to repairing my defective plug ins, and I have a question about the type D, for anyone who may be familiar with the circuit.  

  In the output circuit, there is a pair of type 5879 tubes, V3704 and V3604, that are a matched pair.  How close in gain do they have to be to work in this circuit?

  I have three tubes, measured on my Hickok tester as follows:

     V3604 1210 u-Mhos
     V3704  560 u-Mhos
     Spare   975 u-Mhos

  The tube tester specs give 630 u-Mhos as the minimum transconductance for a good tube so V3704 was replaced with the spare.  But I still can't balance the 
amplifier at the most sensitive settings with the grids of the tubes shorted together, as described in the service instructions.  I presume that is because the gains are
so off.  Is this so, or could it be something else?  Is transconductance at a fixed bias enough info to get me a reasonably close pair of tubes without a curve tracer?

  Thanks for the advice.

  Dan




tubesnthings@aol.com <tubesnthings@...>
 

Dan;
Tube tolerances vary greatly but in a few cases I have seen remarkable consistency.
Strangely, one example is a certain NOS 12AT7WA - 10-out-of-ten (20 triodes!) showed perfectly matched transconductance readings, exactly centered between min/max specs!!

I just picked up a stash of tubes which should contain at least a few 5879s but I haven't gotten around to sorting…

Tek matched tubes for current in test fixtures which resembled the circuit conditions the tube was designated for.

Bernd

Sent from my Verizon Wireless Droid


-----Original message-----
From: Daniel Koller <kaboomdk@...>
To:
"TekScopes@..." <TekScopes@...>
Sent:
Mon, May 20, 2013 14:39:47 GMT+00:00
Subject:
Re: [TekScopes] Repair question for Type D High-Gain Differential plug-in

 

Howdy Bernd,

   I've tried swapping tubes.  It's hard to do in this case since it requires both the plate balance and vertical controls to be moved to near opposite ends of their ranges.  I am too far out of balance.

   Unfortunately, to buy a pair of matching tubes, I can only go by what is offered - i.e. what is commonly measured.  Given that most people on e-bay seem to report only Gm (if that), can I use it as a rough guide to "match" the tubes?   If I buy a pair of tubes from identical manufacturer, presumably new, with the same Gm, are they likely to have the same plate resistance and gain characteristics?   Unfortunately, I think this is the only choice I have - buy a bunch of tubes, and match them myself, in-circuit.  

   My question then is one of manufacturing.   How far off were the tube parameters, say from GE or RCA, from tube to tube?  1%?  10%, 50%??

  Thanks,

  Dan



From: "tubesnthings@..." <tubesnthings@...>
To: TekScopes@...
Sent: Saturday, May 18, 2013 1:32 PM
Subject: Re: [TekScopes] Repair question for Type D High-Gain Differential plug-in

 
Dan,
I have not worked on my type D units, so without looking at the circuit but in general;
Differential amps do rely on closely matched sections. The matching criterium is not transconductance of the tubes as seen by a tube tester, but plate or cathode current in the plug-in's circuit(s). An easy way to determine the tubes' contribution to amp imbalance is to swap them with each other; if the imbalance "follows" the tubes you've isolated that problem. I suspect you'll need to shop for tubes…

Good luck.
Bernd Schroder

Sent from my Verizon Wireless Droid


-----Original message-----
From: Daniel Koller <kaboomdk@...>
To:
"TekScopes@..." <TekScopes@...>
Sent:
Sat, May 18, 2013 13:00:59 GMT+00:00
Subject:
[TekScopes] Repair question for Type D High-Gain Differential plug-in

 
Howdy Folks,

  I'm getting around to repairing my defective plug ins, and I have a question about the type D, for anyone who may be familiar with the circuit.  

  In the output circuit, there is a pair of type 5879 tubes, V3704 and V3604, that are a matched pair.  How close in gain do they have to be to work in this circuit?

  I have three tubes, measured on my Hickok tester as follows:

     V3604 1210 u-Mhos
     V3704  560 u-Mhos
     Spare   975 u-Mhos

  The tube tester specs give 630 u-Mhos as the minimum transconductance for a good tube so V3704 was replaced with the spare.  But I still can't balance the 
amplifier at the most sensitive settings with the grids of the tubes shorted together, as described in the service instructions.  I presume that is because the gains are
so off.  Is this so, or could it be something else?  Is transconductance at a fixed bias enough info to get me a reasonably close pair of tubes without a curve tracer?

  Thanks for the advice.

  Dan




Michael A. Terrell
 

I once had the joy of having to find 16 matched 6146 out of 150 tubes. This was for a video modulator, where they were run in parallel to get the required power and a low output impedance. I managed two almost matched sets, but they aged rapidly in use. Wasn't '50's brute force technology interesting? ;-)

-----Original Message-----
From: "tubesnthings@aol.com" <tubesnthings@aol.com>
Sent: May 20, 2013 3:26 PM
To: TekScopes@yahoogroups.com
Subject: Re: [TekScopes] Repair question for Type D High-Gain Differential plug-in

Dan;
Tube tolerances vary greatly but in a few cases I have seen remarkable consistency.
Strangely, one example is a certain NOS 12AT7WA - 10-out-of-ten (20 triodes!) showed perfectly matched transconductance readings, exactly centered between min/max specs!!

I just picked up a stash of tubes which should contain at least a few 5879s but I haven't gotten around to sorting…

Tek matched tubes for current in test fixtures which resembled the circuit conditions the tube was designated for.
Michael A. Terrell


Dave Wise
 

Daniel Koller, those tubes were selected and matched for grid bias at a given cathode current, and gain. The specs are in the "157" document in the Replaceable Parts Registry at w140.com . Just search for RPR.
You may have to build a test jig like Tek's.