fisher MPX decoder troubleshoot

The apparently only feasible mistake could have been at the ceramics from modulator outputs to grids of matrix mixer (.02 uf). However, when comparing with a spare adapter they are ok. Martin
 
I've gone through all 18 pages of this thread trying to digest some of the infos: got maybe 10% understood...

Anyway, interesting to notice that I crudely tweaked my FM-200-B for 32-33 dB but phase reversed, meaning left is right and right is left. I simply reverse the output RCA leads to correct this...

My crude way: feeding L only and R only (looping for 30 seconds each) with an fm transmitter battery powered and looking for best separation by connecting the output to my computer running "Audiotester" software.

This way, I can play with the different cans (19 kc, 38 kc) until I reach the best numbers. I've got so far 32-33 for a Scott 350, 38 dB for an LT-110 and presently get 30 L to R and 44 R to L on a Sherwood S7700-III. This one puzzles me as I achieved it just by swapping two known good 6GH8A's in the ratio detector position and 38 kc doubler position.

Please note that I didn't aligned the IF strips prior to this so I guess I could have better separation on the 350 and the 200-B.

Dave, if you have the time, do you have an idea why I get this difference of separation on the Sherwood? I didn't messed with the 19 kc and 38 kc coils and xformers BTW.

Thanks,

Patrice
 
Patrice -- I'd have to see the schematic of the Sherwood to offer any reasoned thought on your observation. If you can post it, I'll be happy to look at it.

On your alignment process, a couple of points I'd offer: I too have one of those little battery powered FM Stereo transmitters designed to let your personal players etc., play through (typically) your car radio, but really any nearby FM Stereo radio. Do note this however, the little FM transmitter like that that I have? It's channels are reversed compared the standard. Left signals come out of the right channel, and visa versa. With your transmitter then, you might play some music through it of known stereo orientation to you, and make sure that when it is broadcast, it is decoded with the correct channel identity through a receiver of known correct channel identity. Of course, it is also very possible to adjust the 38 kHz transformer for good separation and audio performance -- albeit with the channels reversed, too.

One other thing to be aware of, the little FM transmitters often will not transmit a pilot signal of sufficient accuracy to meet FCC specifications (19 kHz +/- 2 Hz), but something close enough to at least work. Therefore, you could be adjusting 19 kHz transformers to a less accurate adjustment -- again, just something to be aware of.

I hardly want you to think my comments are meant as criticism, but rather, informative. For a MPX detector that is so screwed up as to produce no stereo at all, your approach can be a valuable tool indeed!

Dave
 
Thanks Dave for your invaluable knowledge. For the phase of the fm transmitter, knowing my Beatles recordings, it's easy to me to say that it is ok.

Interesting about the 19kc thing, meaning all the mpx I aligned this way could be finally off...! At least, I hear there is good separation, for tube mpx that is.

Schematic for the Sherwood, and maybe clues as why the separation is so differing in each channels, thanks.

https://1drv.ms/b/s!AsCftN_jdRMFr1UZajPUwXdP9SPv
 
Hi Patrice -- I would imagine that the 6GH8A in the 38 kHz position produced the greatest change in performance between different examples of the tube.

In the IF strip, both halves of the 6GH8A there are serving to support the muting function. The pentode section is an additional limiter stage, which certainly can't hurt performance. But its primary purpose is to to be a controlling element in the IF signal chain, which is controlled by the triode section in the other half of the tube. Since the IF strip at this point is operating in a saturated (and therefore, limiting) mode, as long as the ratio detector transformer is properly adjusted, its effect on the performance of the stereo decoder should be minimal if anything -- other than maybe at the point where muting is immanent.

On the other hand, the phase of the 38 kHz sub-carrier is everything in terms of reproducing proper stereo sound of the correct channel identity. The sub-carrier is generated in the pentode section of the 6GH8A, which operates at a high transconductance level due to the reduced screen voltage employs. Basically, this stage is just an amplifier, with a resonant circuit connected to the plate, which is tuned to the first harmonic of the pilot frequency -- 38 kHz. But as a high transconductance amplifier, this stage is subject to all the variables these types of tubes are famous for when used as a classic direct coupled pentode to triode AF Amplifier/Phase Inverter combo as used by Dynaco, Scott, and others. Additionally, with the screen voltage relatively fixed in this design (rather than floating as so typically used in audio designs to help stabilize operation), it would only act to accentuate these concerns. Therefore, when the precise accuracy of the 38 kHz phase is considered, relative to the high degree of variability that different 6GH8A tubes could display in this location, I would think it would have the greatest effect on the separation performance of the decoder.

I've never worked on one of these units to test these thoughts out, so that's all they are at this point. But it's a starting point maybe for others who have worked on these units.

Dave
 
Yes, this particular 6GH8 exhibits maximum separation without touching 38 kc and 19 kc coils/cans. The original one gave a reading of around 24 dB left-right and 28 dB right-left. Swapping 12AT7 also modify dramatically the results. By putting another one than the original (stronger), separation decreases under 20 dB.

Now, I have to verify if the fm transmitter really outputs 19 kc. Also, you have no advice as to what I can tweak to obtain a more equal reading as for stereo separation? 38 kc primary or secondary? Anything else. Please also note that even if stereo is easily heard, the neon pilot doesn't light at all ( yes, it's a brand new one before you ask), except sometimes between stations where there's a lot of hiss and distortion...

Anyway, it would be a temporary fix as one of the AK member here in Montréal will thoroughly align it in a month or two (once its backlog will be cleared).

Regards,

Patrice
 
Hi Dave! restoring this MPX, find spec good, replaced audio caps, adjusted deemhasis and replaced electrolitic. All good until I observe a severe distortion in left channel only, with good sine in Right ch. This happens with two different F 300 gens driving directly, all components apparently good. I am inclined to check the correspond modulator diodes, but these are encased in the covering.
Any of your thoughts will be greatly appreciated. Martin
 
Then the issue lies with the demodulated sub-carrier signal. The ring demodulator would in fact be the next stop.

Dave
 
Yes, this particular 6GH8 exhibits maximum separation without touching 38 kc and 19 kc coils/cans. The original one gave a reading of around 24 dB left-right and 28 dB right-left. Swapping 12AT7 also modify dramatically the results. By putting another one than the original (stronger), separation decreases under 20 dB.

Now, I have to verify if the fm transmitter really outputs 19 kc. Also, you have no advice as to what I can tweak to obtain a more equal reading as for stereo separation? 38 kc primary or secondary? Anything else. Please also note that even if stereo is easily heard, the neon pilot doesn't light at all ( yes, it's a brand new one before you ask), except sometimes between stations where there's a lot of hiss and distortion...

Anyway, it would be a temporary fix as one of the AK member here in Montréal will thoroughly align it in a month or two (once its backlog will be cleared).

Regards,

Patrice
Patrice;

Neon lamps are notorious for not firing at the same voltage every time and within a group of neon lamps (of the same type) each one will likely fire at a different voltage. I had a H-K F-50-XK tuner with such a stereo indicator. It was a kit tuner I put together myself way back in the early 1960s. It never worked consistently even after I modified the circuit to allow a ten-turn variable resistor to help find a more accurate firing point of the lamp. It never worked correctly and consistently from day one. I still have the unit and replaced the entire multiplex decoder with a LM4500 IC based decoder which works like a champ and is easy to adjust and drives an LED indicator now. McIntosh had some early tube type multiplex decoders that drove an incandescent lamp, but had a transistor to turn the lamp on and off. That was a better system than any of the neon lamp types.

Joe
 
Hi Dave : The output of both rings have a normal sine with 30K filler, may I assume the problem lies in the final net work coil- resistance -coupling ?
 
Patrice;

Neon lamps are notorious for not firing at the same voltage every time and within a group of neon lamps (of the same type) each one will likely fire at a different voltage. I had a H-K F-50-XK tuner with such a stereo indicator. It was a kit tuner I put together myself way back in the early 1960s. It never worked consistently even after I modified the circuit to allow a ten-turn variable resistor to help find a more accurate firing point of the lamp. It never worked correctly and consistently from day one. I still have the unit and replaced the entire multiplex decoder with a LM4500 IC based decoder which works like a champ and is easy to adjust and drives an LED indicator now. McIntosh had some early tube type multiplex decoders that drove an incandescent lamp, but had a transistor to turn the lamp on and off. That was a better system than any of the neon lamp types.

Joe

Good hint but I suspect that a good alignment is needed anyway. It stayed lit with another 12AT7 but separation was reduced under 20 dB...
 
Have you replaced the output coupling caps? I've seen them go flaky and cause weird problems when they do......

Dave
 
Did you check the ring demodulator for the right channel? To produce a L only or R only output requires both demodulators to be operating correctly.

Dave
 
Back
Top Bottom