Help with FM Signal path circuit debug needed Technics SA-800

rocknroller

Active Member
Have this receiver with no FM audio at all (AM is fine). Not faint, no static - nothing. While I won't rule out the possibility it needs an FM alignment, my intuition tells me this is more of a part failure issue. See attached circuit diagram. My issue is that I don't understand (have the appropriate knowledge) on FM circuit design, so I'm not really clear where the signal is flowing and what is expected. With an audio tester, I hear tuned FM at Test point #1 (TP 1). I also hear an appropriately attenuated signal just past R102. However, I do not hear any audio at Test point 201 (not sure if I'm supposed to) nor do I hear any up to the right at R105. Again - not sure I'm supposed to but thought that may have been where the audio signal travels. If so, that might indicate that strange looking CF103 in between is bad.
I guess i hoping someone can explain where the audio is expected to flow so I can continue to try and debug where it's getting lost. Any thoughts on next steps of things to check would be greatly appreciated as well.

Technics sa-800 FM Path.jpg
 
I guess i hoping someone can explain where the audio is expected to flow so I can continue to try and debug where it's getting lost.

I'm not sure that schematic segment is enough for anyone to say definitely, but my distinct impression is that it shows mostly RF & IF paths, and not too much of the audio paths.

You might want to surf for datasheets for the chips used in the tuner section. I can't identify the ones shown in the segment. Datasheets often have very clear indications of what signals go in and out.

You might also want to tell us about a couple of observations: Is FM still silent even if the FM muting function is switched off? Do the tuning strength and center tuning indicators respond as you move the tuning (with an FM antenna connected)?

Cheers,

chazix
 
I'm not sure that schematic segment is enough for anyone to say definitely, but my distinct impression is that it shows mostly RF & IF paths, and not too much of the audio paths.

You might want to surf for datasheets for the chips used in the tuner section. I can't identify the ones shown in the segment. Datasheets often have very clear indications of what signals go in and out.

You might also want to tell us about a couple of observations: Is FM still silent even if the FM muting function is switched off? Do the tuning strength and center tuning indicators respond as you move the tuning (with an FM antenna connected)?

Cheers,

chazix

I was afraid if I grabbed a larger segment of the schematic it wouldn't be readable as a jpg. I uploaded the whole manual here https://drive.google.com/file/d/1xqsohR7ArcHw2_viqK2vlPPRxjtDPRB_/view?usp=sharing
I should have stated more clearly - The FM seems completely dead (Mute switch off) no meter indications of any kind (and meters work in Am mode, at least signal strength). That's how I figured it was further back from that second chip, since the tuning indicator seems to be feed from it.
 
Do you own an oscilloscope?
Yes - It's a dual trace 15mhz, though I'll be honest that I'm not very experienced with it. I bought it used at an auction with hopes of becoming more familiar with it's use. Up to this point, it has just seemed simpler to just use an audio tester to trace signals. Crude, but effective, at least when determining where it is and where it isn't.
 
Yikes- the pdf document isn't a whole lot better, but it looks like an AN377 chip does the last FM IF stages and the demodulation. So, the FM audio signal path starts at the audio output of the AN377 - which looks like pin 5 from the chip detail, but doesn't look like pin 5 from the schematic, so I still think surfing for a real AN377 datasheet might be helpful.

If there's no meter response, I think the theory that the problem is in the vicinity of the AN377 is a promising one.

I think the crystal filters are unlikely suspects - though not out of the running - and I think it's just a fluke if you detect audio in their vicinities.
 
you did not say what the tuning meters are doing.
you need a known good signal, try tune to a strong sig or of course use a RF SG ;-) if you have one.
read up on the use of your scope in its operating manual.
you should be able to scope the IF sig from the mixer o/p at Tr2 through the IF transformer, CF101,CF102, through the IC in the bot left which is a IF buffer to CF102, then the detector IC in the top right. The signal should have a small loss in the CF's and gain the the IF buffer. see where that gets you.
 
1. Make sure you get around 12V on IC102 pin 11. Then with a FM antenna connected, tune around while monitoring IC102 pin 1and pin 8 with your scope. You should see a 10MHz signal at pin 1 and pin 8, and it should be much larger at pin 8. If you don't see a signal at pin 1 no matter what position you are on the dial the local oscillator may not be working. This is why you really need proper test equipment to troubleshoot rf. You need an rf FM generator and a scope that has at least 100MHz bandwidth. The 15MHz is only good enough for looking at the IF. If you have an oscillator that can put out 10MHz you could inject that into the IF and see if IC102 is amplifying.
 
you did not say what the tuning meters are doing.
you need a known good signal, try tune to a strong sig or of course use a RF SG ;-) if you have one.
read up on the use of your scope in its operating manual.
you should be able to scope the IF sig from the mixer o/p at Tr2 through the IF transformer, CF101,CF102, through the IC in the bot left which is a IF buffer to CF102, then the detector IC in the top right. The signal should have a small loss in the CF's and gain the the IF buffer. see where that gets you.
The signal strength and FM meter does NOTHING in FM mode. No movement at all, regardless of tuner position on the dial. In AM mode, the signal strength works fine. Just a Blip movement on the FM meter, though wouldn't expect the FM meter to do much in AM mode...
 
Ok - To address both rcs16 and Dr*audio's posts - I was able to get the oscilloscope up and going on this. I confirmed the scope sees a signal at TP1 (same place my audio tester hears the station - pretty clear and tuned in). When I jump to the other side of CF103 (Pin 1 on IC 102), I lose the audio on my audio tester, but still see a signal on my scope, though slightly attenuated.

As an education point - what does CF103 do and why does it strip the audio out? Is it somehow modulating that audio to a different frequency that I can't hear?

Following the scope to Pin 8 on IC102, there is still a signal, but it is not boosted. It has actually decreased. buy maybe 1/3rd or even 1/2. Voltage at pin 11 is reading 10.4v.

I thought I had a couple of AF/RF oscillators from my Dad when he passed, but neither seem to be working, (even at the lowest frequencies of audio) so I don't have a generator source to test with. Since apparently my scope can't handle it anyway, it's probably a mute point.
FYI I posted a pic of that part of the circuit below so everyone has a quick reference to it.

Technics sa-800 IC102.jpg
 
It looks like IC102 is bad. The audio tester is for audio signals and not meant for rf. It is surprising you can hear a station on it at all, but it is probably loading down CF103 and that's why you can't hear anything on the output side. CF103 is a 10.7 MHz ceramic filter for the 10.7MHz IF signal. The filters get rid of out of band noise.
 
As an education point - what does CF103 do and why does it strip the audio out?

All three crystal filters (CF101, 102, 103) operate on the 10.7MHz IF (the "intermediate frequency" that is output by the FM front end). I know it is hard to discard hard-won diagnostic evidence, but please believe that any audio signal you detect anywhere near those filters is a red herring. The filters' function is to strip out any signals outside of a narrow range of frequencies centered around 10.7MHz.

Note that, in between CF102 and CF103, the IF signal goes through IC101 for the first stages of IF amplification. So, IC101 is a possible suspect. (On the other hand, IC101 also serves several functions for the AM tuner, which is working.)

I continue to think that the CF parts themselves are unlikely to be the culprits. It is not unheard of that the CFs would go bad, but it's rare.

The output of CF103, which is the input of IC102, is still the 10.7MHz IF signal.

The main function of IC102 is to demodulate the IF signal - that is, to extract the audio signal and output it at pin 6. IC102 also has other jobs, including driving the tuning meters and implementing the FM muting function. So, IC102 is a possible suspect.

If this unit was on my bench, one quick check I would make is to touch a test probe to pin 1 of IC102 (the IF input). The other end of the test probe would be connected to myself, acting as a broad-band antenna. If I still got no audio - no hiss or anything, with FM muting switched off - then I would be very, very suspicious of IC102.
 
If you can take a close up photo of the tuner board that might be interesting to see, it seems kind of rare for these IC's to go bad. Does anything look suspicious nearby? You can also check the soldering underneath for cold solder joints. probably nothing wrong there, but it's an easy thing to check.

I have an SA-700, the tuner layout is probably similar, although yours has a couple extra features. My experience has been that it's a very nice tuner section, it does't give up much to a standalone tuner.
 
Rare or not, it could be IC102, but you can do some more tests first.
You can verify that you can null the quad ckt. pin 10 is the ref bias and pin 7 is the AFC, when the quad ckt is nulled, the AFC is the same DCV as the ref bias. As you de-tune about the center, the AFC V should swing +/- from the ref bias V.
Pin 13 is the level detector, if you do not get any DCV response it is either a bad IC or the IF signal itself is very weak, basically noise.
pin 12 functions as the "OR" of the IF signal level and the deviation mute (DCV deviation of the ref bias and the AFC V)
As a troubleshooting technique, I have taken signals out of one circuit and coupled them into another to verify them.
As an example you could take the signal at IC102-1 (or anywhere else in the IF path), AC couple (use a 10nF ceramic cap) into another tuners IF signal path or discriminator input. Best to use a piece of coax. You will have to open the IF path from the other tuner, so that you can inject the signal from the suspect tuner into known good. It is easy to do by removing a IF filter.
You can always short out CF's (bypass with a 10nF cap) to determine if they are open/lossy.
Compare units DCV for IC102 vs the AN377 data sheet.
Lots of stuff to test out. read the data sheet, lots of information to understand how the IC functions.
I see lots of sources for a AN377, from a US supplier, I would try this over a ECG/NTE part, to start with. get a good socket in there so it is easy to switch out.
https://www.ebay.ca/itm/AN377-FM-IF...hash=item5b2b8b15bc:m:mYKLJTTmFvgiz3VO1mIDqQw
 
Last edited:
Note that, in between CF102 and CF103, the IF signal goes through IC101 for the first stages of IF amplification. So, IC101 is a possible suspect. (On the other hand, IC101 also serves several functions for the AM tuner, which is working.)
one quick check I would make is to touch a test probe to pin 1 of IC102 (the IF input). The other end of the test probe would be connected to myself, acting as a broad-band antenna. If I still got no audio - no hiss or anything, with FM muting switched off - then I would be very, very suspicious of IC102.

Since I do see a scope signal at Pin 1 of IC102, and it does vary as I tune around the dial, can I assume IC101 is doing something properly? As you said, it is working on am as well.

I tried the probe test to pin 1 - nothing.
I also confirmed the mute switch is physically working (and of course is off on on these tests)
 
If you can take a close up photo of the tuner board that might be interesting to see, it seems kind of rare for these IC's to go bad. Does anything look suspicious nearby? You can also check the soldering underneath for cold solder joints. probably nothing wrong there, but it's an easy thing to check.
.

Shown here. Nothing unusual top or bottom. I also did a cursory cap ESR/uf check for the 1+uf caps around there. All seems good.Technics sa-800 baord.JPG
 
Rare or not, it could be IC102, but you can do some more tests first.
You can verify that you can null the quad ckt. pin 10 is the ref bias and pin 7 is the AFC, when the quad ckt is nulled, the AFC is the same DCV as the ref bias. As you de-tune about the center, the AFC V should swing +/- from the ref bias V.
Pin 13 is the level detector, if you do not get any DCV response it is either a bad IC or the IF signal itself is very weak, basically noise.
pin 12 functions as the "OR" of the IF signal level and the deviation mute (DCV deviation of the ref bias and the AFC V)
As a troubleshooting technique, I have taken signals out of one circuit and coupled them into another to verify them.
As an example you could take the signal at IC102-1 (or anywhere else in the IF path), AC couple (use a 10nF ceramic cap) into another tuners IF signal path or discriminator input. Best to use a piece of coax. You will have to open the IF path from the other tuner, so that you can inject the signal from the suspect tuner into known good. It is easy to do by removing a IF filter.
You can always short out CF's (bypass with a 10nF cap) to determine if they are open/lossy.
Compare units DCV for IC102 vs the AN377 data sheet.
Lots of stuff to test out. read the data sheet, lots of information to understand how the IC functions.
I see lots of sources for a AN377, from a US supplier, I would try this over a ECG/NTE part, to start with. get a good socket in there so it is easy to switch out.
https://www.ebay.ca/itm/AN377-FM-IF...hash=item5b2b8b15bc:m:mYKLJTTmFvgiz3VO1mIDqQw

I captured all the voltages around IC102. See pic. I get NO variation on pin 7 as I tune around. Pin 13 is dead. As mentioned early on I only get 10.4 on the input voltage. Someone mentioned 12V expected. The data sheet I found is not in english, though it looks like 13 is expected at that pin (or maybe 13 is the max?) The rest of the sheet is pretty undecipherable to me. (And probably would be still even if it were in english Lol) Would a low input voltage cause this? Or is it more likely the chip is simply bad?

Technics sa-800 IC102 Voltages.jpg
 
Back
Top Bottom