Hi All,
I have an AM/FM here. It didn't take much troubleshooting to find out that the discriminator was 'ate up with' silver mica disease. I opened it up and dug out the old mica wafer, which under a magnifying glass showed the carbon track where the leakage was occurring.
A schematic of a similar radio can be found on this site, as an H723.
I had checked the capactance. It appeared that the primary T4-L16 had an 33pFcapacitor, and the two on the secondary, T4-L17 had two 47pF capacitors. The other capacitor between pins 4 and 5 seemed to be 15pF.
I replaced the wafer with a mica isulator designed for transistors and available from Digi-Key, then hung the capacitors outside the discriminator under the chassis on the radio.
The discriminator secondary didn't tune properly. Instead of getting a null point with positive and negative values on each side, I got a peak near the middle of the tuning range. I thought perhaps that the capacitors were too high, so I changed them out for 33pF, still the same result a peak. The peak is on pin 3 of the 19T8, the tuning point given in the directions. The peak is always a negative voltage.
I tried a little experiment. I clipped out the secondary capacitors entirely. I get a negative null. That is it goes up to about -1.2 volts, then drops to -0.45v then goes back more negative if I continue. I think this is the tuning point, but it isn't right according to the directions.
Any ideas?

Best Regards,

Bill Grimm

Two thoughts...
of the three caps involved..the first consideration is the one tuning the plate tank coil circuit between the plate of the limiter and its B+ / screen supply. It sounds like you are never approaching / or / are past 10.7 resonance, but never right on it.
Now, with the the secondary /discriminator winding, as long as those caps are matched , it would mainly be the detected amplitude that would grossly be affected. Can you start with two fixed and like value caps in their place and move over to the plate winding and temp sub in a variable air /ceramic / or compression mica cap for in circuit evaluation to zero in on its select fixed value. Then assure your self that the plate coils slug is at its max out of the core position and not deep with in the core, optimally within the (min) 1/4 into, up to the 1/2 (max) into core ....out of / to / into core positioning and then using the variable cap in incrementally larger values and testings, see if you can attain both voltage swings. Thereby confirming the capability to aproach...peak and pass 10.7 resonance.
In the interim I will dig and see I can pull one of my BIG BROTHER versions...has full AM RF stage... of that sets transformer and free the windings to get their separate L values as well as their C values.

BTW are you VTVM zero centering for the readings ?..or.... perchance tried scope and sweep generator alignment...where it would show up ever so apparent.

73's de Edd

:Two thoughts...
:of the three caps involved..the first consideration is the one tuning the plate tank coil circuit between the plate of the limiter and its B+ / screen supply. It sounds like you are never approaching / or / are past 10.7 resonance, but never right on it.
:Now, with the the secondary /discriminator winding, as long as those caps are matched , it would mainly be the detected amplitude that would grossly be affected. Can you start with two fixed and like value caps in their place and move over to the plate winding and temp sub in a variable air /ceramic / or compression mica cap for in circuit evaluation to zero in on its select fixed value. Then assure your self that the plate coils slug is at its max out of the core position and not deep with in the core, optimally within the (min) 1/4 into, up to the 1/2 (max) into core ....out of / to / into core positioning and then using the variable cap in incrementally larger values and testings, see if you can attain both voltage swings. Thereby confirming the capability to aproach...peak and pass 10.7 resonance.
:In the interim I will dig and see I can pull one of my BIG BROTHER versions...has full AM RF stage... of that sets transformer and free the windings to get their separate L values as well as their C values.
:
:BTW are you VTVM zero centering for the readings ?..or.... perchance tried scope and sweep generator alignment...where it would show up ever so apparent.
:
:73's de Edd
:
:
:

Hi Edd,
Thank you for the reply. Are you saying that I am likely off on the primary's resonance? Kind of makes sense. Also, I was wondering if an imbalance on those output caps is causeing the output circuit to favor the negative?
I will go and hunt up some variable caps. Your suggestions make sense.

Best Regards,

Bill Girmm

No waiting at all for you in getting all of your further adjunct tech info specs , as I still have that chassiss under my lab bench, waiting for the final curing of a clear lucite pointer that I had cast and molded a new one of, to be put back on its dial pointer assembly.

A taking of a reading of the limiters bottom primary coil (A8) and its tuning cap revealed it to be a 24 pf unit for the resonating of that companion coil which measured out as 9 uh, with its tuning slug remaining in its optimally aligned positioning.

A reading of the disc top secondary coil (A9) and its tuning cap...a series pair... revealed each of then to be 70 Pf value units, with that coils inductive value being 6 uh with it also having its slug tuned to zero crossover alignment.

If visibly marked on the FM Disc T-former proper, your unit shows to be a Zenith 95-1153....whilst my newer ..chassis used a part number of 95-1865. Both of them interfacing with that same common IF circuitry and driven by the interstage capacitance values of that 12AU6 limiter tube.

And whilst in the final process of trimming in the initial A8 bottom coils hex slug adjustment and then the final
discriminator- zero-center of that top A9 coil...."Dont worry...be happy"..... you can always let your wrist laze and sag down and rest atop that selenium rectifier......LOL.

73's de Edd

:No waiting at all for you in getting all of your further adjunct tech info specs , as I still have that chassiss under my lab bench, waiting for the final curing of a clear lucite pointer that I had cast and molded a new one of, to be put back on its dial pointer assembly.
:
:A taking of a reading of the limiters bottom primary coil (A8) and its tuning cap revealed it to be a 24 pf unit for the resonating of that companion coil which measured out as 9 uh, with its tuning slug remaining in its optimally aligned positioning.
:
:A reading of the disc top secondary coil (A9) and its tuning cap...a series pair... revealed each of then to be 70 Pf value units, with that coils inductive value being 6 uh with it also having its slug tuned to zero crossover alignment.
:
:If visibly marked on the FM Disc T-former proper, your unit shows to be a Zenith 95-1153....whilst my newer ..chassis used a part number of 95-1865. Both of them interfacing with that same common IF circuitry and driven by the interstage capacitance values of that 12AU6 limiter tube.
:
:And whilst in the final process of trimming in the initial A8 bottom coils hex slug adjustment and then the final
:discriminator- zero-center of that top A9 coil...."Dont worry...be happy"..... you can always let your wrist laze and sag down and rest atop that selenium rectifier......LOL.
:
:73's de Edd

Hi Edd,
Thank you so much! I will try 70pF, 47pf+22pF on the secondary, and 22pF on the primary. I think my bridge isn't very accurate at such a low values.
The numbers match, too, 95-1153.
I will let you know how it does.

The selenium rectifier is disconnected. No problem there.

All the Best,

Bill Grimm

Please confirm the last of this second paragraph 2 to see if your mentioned capacitive values are in accordance.
It just might be a mere matter of semantics.

1...You are OK on the primary of the tranfo in using a 22 pf ...in not having that 24 value on hand as
it can readily be corrected in the inductive tuning of the slug in that companion coil winding.

2.....But on the secondary, as I mentioned, there are a pair of 70 pf units in series. Now what you might have mentioned was your ALSO not having that 70 pf value ...so what you were doing was working with what you had, and using a 47 and a 22 pf in series connection to get your as close as 69 pf to that specified 70pf value.
Then you would have to replicate it again to get another 69 pf value and then I am sure that there will be adequate room in that large can to get 4 caps nestled within its cavity and a tap in at the center of the cap pairs to get feed back to terminal 2 of the tranfo that is getting phase referencing from the plate of the limiter tube.

That was it

BTW...Initially, if a person should want a finite / initial cursory evaluation of the "health" of a sets discriminator circuit alignment, one would hook up an analog VTVM or FET-VM to the "B".... detector output...of a sets discriminator circuitry. Then you use a moderate strength, off the air signal and merely do an initial off station, gradual tuning into the capture range of that FM signal and watch the gradual rise of that voltage as you are tuning towards and acquiring the peak positive value of that voltage . Then as you tune ever closer to the peak response node of that transformers tuning there will be a point where the voltage reading will crest
to a peak value and then start a decline in ceclining voltage value as the tuning is shifting further down the response curve of that node, eventually diminishing to a zero voltage value and making a transition to a negative voltage value as you are tuning towards its climb onto the negative node of the transformers bandpass response.
In a like manner, that voltage will increase in the negative voltage direction until it peaks out and then starts its decline until it finally slips out of the tuning capture range of the tuner.

As a pictoral referencing...see:

http://img58.imageshack.us/img58/5338/discriminatorresponsecuwj9.jpg

A brief synopsis of the above:

Tune into a station watching for the peak detected voltage level.
Log down that voltage level for further reference
Continue tuning thru O voltage level and expect another voltage peak of the opposite polarity of the initial peak reading previously experienced.
Compare the two peak reading to see if they are equal, if not, the discriminator secondary needs a slight trimming to acquire balanced + and - node voltage values.

Referring to the pictoral, one can see the potential degrees of imbalance with misaligned conditions.

A misaligned discriminator will result in lower as well as imbalanced, peak negative and positive node voltages coming from the discriminators detector output, along with a resultant output distortion.
If the tuning of the discriminators plate coil of the limiter circuit is proper and the zero centering of the detector output is "right on" there will be an optimally balanced output from both nodes, with them also being at their highest voltage levels, as well as optimal fidelity and minumal distortion from the discriminator circuit

Thassit

73's de Edd

:Please confirm the last of this second paragraph 2 to see if your mentioned capacitive values are in accordance.
:It just might be a mere matter of semantics.
:
:1...You are OK on the primary of the tranfo in using a 22 pf ...in not having that 24 value on hand as
:it can readily be corrected in the inductive tuning of the slug in that companion coil winding.
:
:2.....But on the secondary, as I mentioned, there are a pair of 70 pf units in series. Now what you might have mentioned was your ALSO not having that 70 pf value ...so what you were doing was working with what you had, and using a 47 and a 22 pf in series connection to get your as close as 69 pf to that specified 70pf value.
:Then you would have to replicate it again to get another 69 pf value and then I am sure that there will be adequate room in that large can to get 4 caps nestled within its cavity and a tap in at the center of the cap pairs to get feed back to terminal 2 of the tranfo that is getting phase referencing from the plate of the limiter tube.
:
:That was it
:
:BTW...Initially, if a person should want a finite / initial cursory evaluation of the "health" of a sets discriminator circuit alignment, one would hook up an analog VTVM or FET-VM to the "B".... detector output...of a sets discriminator circuitry. Then you use a moderate strength, off the air signal and merely do an initial off station, gradual tuning into the capture range of that FM signal and watch the gradual rise of that voltage as you are tuning towards and acquiring the peak positive value of that voltage . Then as you tune ever closer to the peak response node of that transformers tuning there will be a point where the voltage reading will crest
:to a peak value and then start a decline in ceclining voltage value as the tuning is shifting further down the response curve of that node, eventually diminishing to a zero voltage value and making a transition to a negative voltage value as you are tuning towards its climb onto the negative node of the transformers bandpass response.
:In a like manner, that voltage will increase in the negative voltage direction until it peaks out and then starts its decline until it finally slips out of the tuning capture range of the tuner.
:
:As a pictoral referencing...see:
:
:http://img58.imageshack.us/img58/5338/discriminatorresponsecuwj9.jpg
:
:
:A brief synopsis of the above:
:
:Tune into a station watching for the peak detected voltage level.
:Log down that voltage level for further reference
:Continue tuning thru O voltage level and expect another voltage peak of the opposite polarity of the initial peak reading previously experienced.
:Compare the two peak reading to see if they are equal, if not, the discriminator secondary needs a slight trimming to acquire balanced + and - node voltage values.
:
:Referring to the pictoral, one can see the potential degrees of imbalance with misaligned conditions.
:
:
:
:A misaligned discriminator will result in lower as well as imbalanced, peak negative and positive node voltages coming from the discriminators detector output, along with a resultant output distortion.
:If the tuning of the discriminators plate coil of the limiter circuit is proper and the zero centering of the detector output is "right on" there will be an optimally balanced output from both nodes, with them also being at their highest voltage levels, as well as optimal fidelity and minumal distortion from the discriminator circuit
:
:Thassit
:
:73's de Edd

Hi Edd,
Got it!

When I went to change out the capacitors, I was surprised to find that the primary capacitor was already 22pF! The primary core, however, was bottomed out. My capacitor misadventures on the secondary had gotten me to mistune the primary by quite a bit.
It seems that with the primary way off, the phasing goes to the secondary wrong. Refereencing the picture, I was at the left example of mistuned, but even worse! I couldn't get a positive voltage.
I found out that the primary and secondary have an effect on each other, so I had to tune back and forth a few times, but got peak on the primary, and null on the seconary. The tome quality is noticible improved.

I have the two 69pf (47pf and 22 pF in parallel) capacitors on the outside of the can. I have done inside and it can be a bit rough, especially when I am experimenting.
Now it seems experimenting is over thanks to you. 70pf secondary capacitors are it!

I have a scope and a signal generator that will FM modulate. I have been wanting to put this on a scope, but the modulation signal comes out of the signal generator at 400v P-P, too high for my scope. I think I will make time to do that resistive divider I have been meaning to get to, to see the displays. That will tell how well I am really doing.

Many thanks and 73's

Bill Grimm

I have the two 69pf (47pf and 22 pF in parallel) capacitors on the outside of the can. I have done inside and it can be a bit rough, especially when I am experimenting.
Now it seems experimenting is over thanks to you. 70pf secondary capacitors are it!

heh.. heh... heh I knew what you were doing and confirmed it above, but my fingers were distracted and typed them as being in series !

As well on the transformers core positions.... if in doubt and having to to a confirmation, I always take the respective slugs to their outermost
positions well out of the winding area proper and walk in the disc plate coil 'til it hits 10.7 resonance then start working with the disc
secondary coil slug and in the end you will find its moving down into the proximity of the initially peaked coil will have it being shifted also
so the two have to be alternatively trimmed in again until the plate coil is at peaak again at the same time that the disc secondary is at zero center

I have a scope and a signal generator that will FM modulate. I have been wanting to put this on a scope, but the modulation signal comes out of the signal generator at 400v P-P, too high for my scope. I think I will make time to do that resistive divider I have been meaning to get to, to see the displays. That will tell how well I am really doing.

NOT REALLY...what you have there is not a signal generator...but instead..... a radio stations transmitter exciter ...LOL.

73's de Edd

::Please confirm the last of this second paragraph 2 to see if your mentioned capacitive values are in accordance.
::It just might be a mere matter of semantics.
::
::1...You are OK on the primary of the tranfo in using a 22 pf ...in not having that 24 value on hand as
::it can readily be corrected in the inductive tuning of the slug in that companion coil winding.
::
::2.....But on the secondary, as I mentioned, there are a pair of 70 pf units in series. Now what you might have mentioned was your ALSO not having that 70 pf value ...so what you were doing was working with what you had, and using a 47 and a 22 pf in series connection to get your as close as 69 pf to that specified 70pf value.
::Then you would have to replicate it again to get another 69 pf value and then I am sure that there will be adequate room in that large can to get 4 caps nestled within its cavity and a tap in at the center of the cap pairs to get feed back to terminal 2 of the tranfo that is getting phase referencing from the plate of the limiter tube.
::
::That was it
::
::BTW...Initially, if a person should want a finite / initial cursory evaluation of the "health" of a sets discriminator circuit alignment, one would hook up an analog VTVM or FET-VM to the "B".... detector output...of a sets discriminator circuitry. Then you use a moderate strength, off the air signal and merely do an initial off station, gradual tuning into the capture range of that FM signal and watch the gradual rise of that voltage as you are tuning towards and acquiring the peak positive value of that voltage . Then as you tune ever closer to the peak response node of that transformers tuning there will be a point where the voltage reading will crest
::to a peak value and then start a decline in ceclining voltage value as the tuning is shifting further down the response curve of that node, eventually diminishing to a zero voltage value and making a transition to a negative voltage value as you are tuning towards its climb onto the negative node of the transformers bandpass response.
::In a like manner, that voltage will increase in the negative voltage direction until it peaks out and then starts its decline until it finally slips out of the tuning capture range of the tuner.
::
::As a pictoral referencing...see:
::
::http://img58.imageshack.us/img58/5338/discriminatorresponsecuwj9.jpg
::
::
::A brief synopsis of the above:
::
::Tune into a station watching for the peak detected voltage level.
::Log down that voltage level for further reference
::Continue tuning thru O voltage level and expect another voltage peak of the opposite polarity of the initial peak reading previously experienced.
::Compare the two peak reading to see if they are equal, if not, the discriminator secondary needs a slight trimming to acquire balanced + and - node voltage values.
::
::Referring to the pictoral, one can see the potential degrees of imbalance with misaligned conditions.
::
::
::
::A misaligned discriminator will result in lower as well as imbalanced, peak negative and positive node voltages coming from the discriminators detector output, along with a resultant output distortion.
::If the tuning of the discriminators plate coil of the limiter circuit is proper and the zero centering of the detector output is "right on" there will be an optimally balanced output from both nodes, with them also being at their highest voltage levels, as well as optimal fidelity and minumal distortion from the discriminator circuit
::
::Thassit
::
::73's de Edd
:
:Hi Edd,
: Got it!
:
: When I went to change out the capacitors, I was surprised to find that the primary capacitor was already 22pF! The primary core, however, was bottomed out. My capacitor misadventures on the secondary had gotten me to mistune the primary by quite a bit.
: It seems that with the primary way off, the phasing goes to the secondary wrong. Refereencing the picture, I was at the left example of mistuned, but even worse! I couldn't get a positive voltage.
: I found out that the primary and secondary have an effect on each other, so I had to tune back and forth a few times, but got peak on the primary, and null on the seconary. The tome quality is noticible improved.
:
: I have the two 69pf (47pf and 22 pF in parallel) capacitors on the outside of the can. I have done inside and it can be a bit rough, especially when I am experimenting.
: Now it seems experimenting is over thanks to you. 70pf secondary capacitors are it!
:
: I have a scope and a signal generator that will FM modulate. I have been wanting to put this on a scope, but the modulation signal comes out of the signal generator at 400v P-P, too high for my scope. I think I will make time to do that resistive divider I have been meaning to get to, to see the displays. That will tell how well I am really doing.
:
:Many thanks and 73's
:
:Bill Grimm

:I have the two 69pf (47pf and 22 pF in parallel) capacitors on the outside of the can. I have done inside and it can be a bit rough, especially when I am experimenting.
:Now it seems experimenting is over thanks to you. 70pf secondary capacitors are it!
:
:heh.. heh... heh I knew what you were doing and confirmed it above, but my fingers were distracted and typed them as being in series !
:
:As well on the transformers core positions.... if in doubt and having to to a confirmation, I always take the respective slugs to their outermost
:positions well out of the winding area proper and walk in the disc plate coil 'til it hits 10.7 resonance then start working with the disc
:secondary coil slug and in the end you will find its moving down into the proximity of the initially peaked coil will have it being shifted also
:so the two have to be alternatively trimmed in again until the plate coil is at peaak again at the same time that the disc secondary is at zero center
:
:I have a scope and a signal generator that will FM modulate. I have been wanting to put this on a scope, but the modulation signal comes out of the signal generator at 400v P-P, too high for my scope. I think I will make time to do that resistive divider I have been meaning to get to, to see the displays. That will tell how well I am really doing.
:
:NOT REALLY...what you have there is not a signal generator...but instead..... a radio stations transmitter exciter ...LOL.
:
:73's de Edd
:
:::Please confirm the last of this second paragraph 2 to see if your mentioned capacitive values are in accordance.
:::It just might be a mere matter of semantics.
:::
:::1...You are OK on the primary of the tranfo in using a 22 pf ...in not having that 24 value on hand as
:::it can readily be corrected in the inductive tuning of the slug in that companion coil winding.
:::
:::2.....But on the secondary, as I mentioned, there are a pair of 70 pf units in series. Now what you might have mentioned was your ALSO not having that 70 pf value ...so what you were doing was working with what you had, and using a 47 and a 22 pf in series connection to get your as close as 69 pf to that specified 70pf value.
:::Then you would have to replicate it again to get another 69 pf value and then I am sure that there will be adequate room in that large can to get 4 caps nestled within its cavity and a tap in at the center of the cap pairs to get feed back to terminal 2 of the tranfo that is getting phase referencing from the plate of the limiter tube.
:::
:::That was it
:::
:::BTW...Initially, if a person should want a finite / initial cursory evaluation of the "health" of a sets discriminator circuit alignment, one would hook up an analog VTVM or FET-VM to the "B".... detector output...of a sets discriminator circuitry. Then you use a moderate strength, off the air signal and merely do an initial off station, gradual tuning into the capture range of that FM signal and watch the gradual rise of that voltage as you are tuning towards and acquiring the peak positive value of that voltage . Then as you tune ever closer to the peak response node of that transformers tuning there will be a point where the voltage reading will crest
:::to a peak value and then start a decline in ceclining voltage value as the tuning is shifting further down the response curve of that node, eventually diminishing to a zero voltage value and making a transition to a negative voltage value as you are tuning towards its climb onto the negative node of the transformers bandpass response.
:::In a like manner, that voltage will increase in the negative voltage direction until it peaks out and then starts its decline until it finally slips out of the tuning capture range of the tuner.
:::
:::As a pictoral referencing...see:
:::
:::http://img58.imageshack.us/img58/5338/discriminatorresponsecuwj9.jpg
:::
:::
:::A brief synopsis of the above:
:::
:::Tune into a station watching for the peak detected voltage level.
:::Log down that voltage level for further reference
:::Continue tuning thru O voltage level and expect another voltage peak of the opposite polarity of the initial peak reading previously experienced.
:::Compare the two peak reading to see if they are equal, if not, the discriminator secondary needs a slight trimming to acquire balanced + and - node voltage values.
:::
:::Referring to the pictoral, one can see the potential degrees of imbalance with misaligned conditions.
:::
:::
:::
:::A misaligned discriminator will result in lower as well as imbalanced, peak negative and positive node voltages coming from the discriminators detector output, along with a resultant output distortion.
:::If the tuning of the discriminators plate coil of the limiter circuit is proper and the zero centering of the detector output is "right on" there will be an optimally balanced output from both nodes, with them also being at their highest voltage levels, as well as optimal fidelity and minumal distortion from the discriminator circuit
:::
:::Thassit
:::
:::73's de Edd
::
::Hi Edd,
:: Got it!
::
:: When I went to change out the capacitors, I was surprised to find that the primary capacitor was already 22pF! The primary core, however, was bottomed out. My capacitor misadventures on the secondary had gotten me to mistune the primary by quite a bit.
:: It seems that with the primary way off, the phasing goes to the secondary wrong. Refereencing the picture, I was at the left example of mistuned, but even worse! I couldn't get a positive voltage.
:: I found out that the primary and secondary have an effect on each other, so I had to tune back and forth a few times, but got peak on the primary, and null on the seconary. The tome quality is noticible improved.
::
:: I have the two 69pf (47pf and 22 pF in parallel) capacitors on the outside of the can. I have done inside and it can be a bit rough, especially when I am experimenting.
:: Now it seems experimenting is over thanks to you. 70pf secondary capacitors are it!
::
:: I have a scope and a signal generator that will FM modulate. I have been wanting to put this on a scope, but the modulation signal comes out of the signal generator at 400v P-P, too high for my scope. I think I will make time to do that resistive divider I have been meaning to get to, to see the displays. That will tell how well I am really doing.
::
::Many thanks and 73's
::
::Bill Grimm

Hi Edd,
Ah yes, your description of how core positions affect the tuning makes excellent sense. I think that from now on the tuning will be easier since the new capacitors will be similar in size to the original.
I have done a few of these radios and plan to do more. The discriminator is commonly infected with silver mica disease, since the secondary capacitors have around 50 volts across them.
I understand that the null balance can be checked with a radio station now. I will try that before doing the signal generator thing. The trick with the signal generatore and scope will create a display similar to your picture.

Best Regards, and 73's

Bill