fisher MPX decoder troubleshoot

CLUE? D.C measurements at output of left ring= 2.9V , compared with Rt ring out= 0.4V. Compared to good unit out rt&left= 0.0v .
Could this be biasing the left ring to behave bad? The coupling electrolitic is the same in both units. Martin
 
Any chance that one of the .001 uF caps that couples the 38 Khz signal to the rings is leaking?
 
Improvement: Flipped left ring , output to input solved the abnormal dc in output terminal. Separation 10bB low. Will replace left ring when I get the Fairchild diodes.
Lesson in this one:If the dc v at output of ring is high, diodes may be flipped. Martin
 
Vintelectra 's axiom: it is more difficult to detect our own mistakes than spontaneous circuit faults,
because ours do not follow logic, while the other does. Martin
 
New here! First post. I've read with great interest this entire thread as I am working on a HH Scott 350D. This is my first venture into FM stereo multiplexing so am learning a tremendous amount from Dave. Thank you! I have been unable to find any alignment procedures for the Scott - only have a schematic. Dave, a couple of questions, but first a bit of info. Utilizing a spectrum analyzer with a TG, I find the IF slightly below 10.7 mhz (10.5) and not necessarily peaked. I read with interest your instructions for peaking the IF without instruments. Sometimes, I understand, IF needs to be stagger tuned for a wide enough band pass for proper frequency response. How wide should the response be for, say, a 3db drop on both sides of center? How flat the top? And finally, so that this stays in the domain of de-multiplexing, how do these adjustment affect the detector/de-mux? The Scott is working with reasonable stereo on stronger stations and sounds ok, but I think it is capable of much better performance. Unfortunately I don't have the likes of the fisher 300 - wish I did! Can a reasonable job be done with a spectrum analyzer, sweep generator, frequency counter, and a rebuilt Tektonix 465? Finally, do you do custom alignment work!?
 
RF -- Welcome to AK! By the time the units were developed (among all manufacturers) that employed on board MPX decoders, all of the IF circuitry was changed to "wide band" operation -- which by that time was primarily concerned with the discriminator/ ratio detector stage. By the beginning of the 60's most manufacturers had abandoned the stagger tuned type of IF design in favor of a more traditional type that could simply be peaked. The older stagger tuned transformers had a much narrower bandwidth, requiring the stagger alignment to get the flattest results. The newer transformer (from about 58-59 on) were generally all of the peaking type, so that the need for stagger alignment went by the wayside. Most of the early "stereo ready" tuners included transformers that would pass about 260 kHz between 6 db down points, which you would certainly need all of for good MPX stereo. You couldn't really expand this any further since the FM stations are separated by just 200 kHz. Any greater bandwidth in the IF strip would certainly impair good selectivity. Max deviation for FM is 75 kHz on either side of the IF frequency, so a flat response is needed over at least 150 kHz, centered on the IF frequency. The extra bandwidth beyond that was simply the roll off on either side of the center frequency.

The alignment of the IF strip is crucial to the proper operation of MPX decoder, because as you can see, there's not a lot of wiggle room for the strip to be off if the full stereo channel is to be faithfully reproduced. Of particular importance is the discriminator stage transformer's alignment, so that zero output corresponds directly with 10.7 mHz, with linear deviation on either side of this frequency with modulation.

The Scott ratio detector is an unusual animal, producing a very low output. This then requires a pentode pilot signal amplification stage in the decoder, as well as the additional audio output stage that most examples of their decoder contains. For alignment of the MPX section without an FM Stereo generator, once tuned, the pilot signal of most any strong station can be used to peak the 19 kHz and 38 kHz sub-carrier transformers, watching that you don't accidentally reverse the channel identity in the process. The difference between the L and R stereo channels is simply matter of the 38 kHz sub-carrier's relationship to the 19 kHz pilot signal: it can be peaked for great operation -- with the channels perfectly reversed.

For the tuner section, the FC is your friend if using a sweep generator, so that a very good job can in fact be done with that equipment. Without the counter, it's tough to know if the frequency you're aligning the strip to is accurate or not if the generator does not include an accurate digital readout as to frequency.

I have been winding down the work I do considerably, but if you'd like to discuss that it would be best to do so via PM.

I hope this helps!

Dave
 
Wow, thank you! I may venture into the black art of FM alignment with this unit. I've hesitated as I have read several times to leave it alone unless one has the "really serious" equipment. I can see how the IF stages respond and what frequency they are peaked at on the spectrum analyzer. If it's a matter of peaking each stage with the width of the peak determined by the design of the transformer, I can do that. My sweep generator is a cheap digital unit but it's frequencies are spot on according to the counter and the analyzer. Unfortunately, it doesn't go up into the FM broadcast band (60mhz) so I have to use the harmonic, which does work. I'll probably do the IF's with the analyzer and tracking generator. Nice to see what your are doing.

The 19k pilot and the output of the doubler are clear, but not clean. The peaks on the input to the 38k transformer are are quite modulated. The crossing points are very distinct. I would assume the MPX signal from the secondary is coupling into the primary (looking at a tuned in station as I don't have a MPX generator.) The output of the diode matrix is a bit less distinct than I think it should be and its symmetry is highly dependent on minute tuning changes. The flat part of the rectified signal is not flat but has a definite curve - flat on one end then curves up to the signal.

One question that concerns me is the IF transformers. Is it possible that these use the built in mica caps that are noted for silver mica disease? Is there a source for service info on the Scott's? I can't find any, even on the HH Scott web page. It's not listed in the Sam's index.

I have learned a tremendous amount from this thread and I thank you for your patience! I will try not to bother you again.
 
Last edited:
500B or 800B manual. WX is 2 12at7's(V100, V102) and 1 12ax7 (V101). The KM-60 also uses the WX.
 
Hello Boss and Fans: What would cause this distortion on the Lissajous 2:1 in this MPX 65 driven out of chassis?? everything else apparently OK Thanks Martin
 
That was my 1st thought, but resist-wise, is ok. I found a 1.7 v ripple in my external PS that I will correct and then recap the doubler, unless you suggest otherwise.
thanks Martin
 
This one turned easy! corrected the ripple and viola" I dont understand why thou
 

Attachments

  • DSCN0254.JPG
    DSCN0254.JPG
    47.3 KB · Views: 2
If you are operating the sub-chassis from an external power supply for some reason, then the 165 volt supply to the oscillator stage must be well decoupled. If not, the high resulting AC impedance will not allow the proper coupling to take place between the two halves of the primary winding in the transformer coil, and prevent the oscillator from operating.

Dave
 
Back
Top Bottom