990DB FM Tuner erratic

[vinylisfinyl]

Salutations, Sansui specialists!

I am in need of your input on a problem with my 990DB and its FM tuner. I got this unit from my aunt, who rescued it from her neighbor’s garbage (fifty lashes on the backside for throwing a Sansui out!!). After replacing the fusible resistors on the driver board and cleaning the switches/pots I get great audio from AM and AUX inputs, but the FM mode is giving erratic performance.

The tuner finds our local stations just fine but the audio constantly cuts between static and audio. It acts as if the FM signal suddenly goes from being strong to weak and then back, even with an FM dipole antenna connected. The signal and tune meters will swing back and forth from strong signal to weak and back. Once in a while it will lock onto the station, show FM STEREO and play good audio for a few seconds but then returns to the same behavior. The first time I ran the unit, it took two hours for this behavior to show up but now it happens anytime I turn to FM. The AM mode is rock solid and tunes/plays without issue.

I’ve ruled out the Dolby and selector switches since AM and AUX play just fine when the unit is in DOLBY OFF mode, and the selector switch showed no continuity issues. I recapped the tuner power supply, modded it with the CDFixer's X0X0DB Tuner Power Supply mod and confirmed it's supplying 12.9 V (same as before recapping). The HA1196 and filter transistor (TR01 and TR02) voltages measured within spec. I then scoped the signal input from the F-1519 board to the FM tuner board, and saw the signal going from low static to a strong signal in time with the audio cutting in and out, so looks like my issue is with the front-end pack (F-1519).

Here is where I’m stuck. The service manual shows no voltages or check points for F-1519, so I have no idea what signal I should see coming out of it or how to check the health of the components on board. I have seen two references here on AK for folks having a failed 3SK41 FET in their units (F01 in the attached schematic), which might explain the behavior but I’m not sure how to properly check a FET or how to confirm that this is actually my problem. When I scoped the legs of it I got nothing on my scope (might have been my error).

Can anyone recommend what my next troubleshooting step would be? Should I try replacing the FET with an NTE 222 and see what happens? Does anyone have experience with common trouble points or troubleshooting these Sansui front end packs? I've attached images of these boards' schematics for reference. Any and all input will be appreciated, thank you!

 

[LBPete]

Sounds like the tuning cap needs to be cleaned. Here's Dr*Audio's how to thread.

- Pete

 

[vinylisfinyl]

Oh that would be wonderful if it's something that easy! I have witnessed the tuner drifting slightly too so a cleaning I shall try. Pretty sure I have all the materials that that thread recommends for this job.

By the way, I posted my 990DB's serial number in the Sansui Database thread.

 

[Robisme]

Could be the oscillator is dirty or out of adjustment. I have an 8080DB here that was drifting and it was the oscillator.

Touching it is not recommended because it may need an alignment after, but I only moved it back and forth a tiny bit.
It could also be that it needs an alignment.

I think it's called the oscillator, I really don't do tuners. The adjustment on the tuning cap to the far left. Maybe a red screw.

Hopefully it just needs a cleaning as Pete suggests.


Rob

 

[vinylisfinyl]

Sigh, unfortunately I don't think cleaning was the complete answer. After pulling F-1519 out, desoldering the board from the cage, blowing out the fins, putting some Faderlube on the brass contacts with a toothpick, and resoldering the wires the symptoms are only a little better. I couldn't get much contact cleaner on the brass-housing connections (dried up too fast), but the inside of this tuner assembly was very clean so I don't think this was the issue. Am and FM share the same assembly.

FM sound doesn't cut out like before, but the TUNE needle swings slightly back and forth while locked onto an FM station. Whenever the needle swings, I hear some distortion in the audio and sometimes the STEREO indicator goes out and comes back. If I tune to an AM station, it plays rock solid with no static or distortion.

I checked out the BJTs on F-1519 and didn't see any dead shorts, though TR1 had 0.785 voltage drop from its collector to emitter when I was expecting to see OL...not sure if that is an in-circuit error or old-BJT thing. I put a meter across the source and drain of the FET and got .15 V, which I think is OK but I'm not 100% sure. I got OL when I tested gate to source and gate to drain.

Any other ideas or suggestions on what I can check? Is the oscillator a mechanical item, Rob, or something electronic? I know what an oscillator is in theory, just curious what you touched on your 8080DB.

 

[Robisme]

Far left adjustment. Turning it will tune stations, so if you choose to adjust it, only move it a tiny bit. Back and forth a couple times. If you tune to a station before
adjusting it, you can tune back to the station with the adjustment.

When I say a tiny bit, I mean barely move it.

 

[vinylisfinyl]

So when you "moved it only a tiny but", it cured your drifting issues, Rob?

If so I might take a crack at it, but seeing I don't have much tools besides a 50 Mhz Rigol scope and two Fluke meters I'm concerned that I might make this worse than it is. I'm wondering if it would be worth finding a good repair shop and have someone do it for me since everything else works so well. Know of any good ones around the Northwest PA area?

 

[kevzep]

So when you "moved it only a tiny but", it cured your drifting issues, Rob?

If so I might take a crack at it, but seeing I don't have much tools besides a 50 Mhz Rigol scope and two Fluke meters I'm concerned that I might make this worse than it is. I'm wondering if it would be worth finding a good repair shop and have someone do it for me since everything else works so well. Know of any good ones around the Northwest PA area?

Seriously, if you don't have the knowledge and understanding of what you are doing in there, get it done professionally.
Its very easy to get it wrong and then you'll need to take it to someone anyway.

There are a number of things which can cause this type of fault.
It could be a failing RF FET, partially shorted cap, a weak transistor in the IF stage there are also OP AMPS are used in the IF stage as well and one of those could possibly be causing this as well. But with the tuning meter swinging wildly I was guess there may be something amiss in the Discriminator as well. The LO (local oscillator) could cause such a fault. But to adjust it properly you need an RF generator which has Frequency Modulation and an Audio oscillator, they have a low range adjustment and a high range and should be adjusted correctly using the method laid out in the manual.
You don't want the LO to be off as it will not track the radio stations properly across the dial, they might be on at the low end of the dial, but then off at the top end....there is a method to adjust it, and its not moving it a "tiny bit", but thats from my book, everyone works differently though . I would not fiddle with it personally, but its up to you.
I work on tuners a lot and I can tell you a lot of successful diagnoses just comes from experience.

 

[vinylisfinyl]

I appreciate your honesty about this, and am feeling the same. I've read enough about FM tuner work that it's not easily picked up by newcomers to the hobby like myself. I am going to look for a shop that can do this for me, but first I may just replace the FET on the front end pack (F-1519 board) myself and see what happens. I don't have much to lose since it's not listenable as it is. The 3SK41 FET appears to cross over to a NTE222, so if no one thinks that shotgunning the FET is a bad idea I'm going to tackle that in the next few days here.

 

[kevzep]

I appreciate your honesty about this, and am feeling the same. I've read enough about FM tuner work that it's not easily picked up by newcomers to the hobby like myself. I am going to look for a shop that can do this for me, but first I may just replace the FET on the front end pack (F-1519 board) myself and see what happens. I don't have much to lose since it's not listenable as it is. The 3SK41 FET appears to cross over to a NTE222, so if no one thinks that shotgunning the FET is a bad idea I'm going to tackle that in the next few days here.

The FET is not a bad place to start, I have bought several 3SK41 dual gate fets off ebay before, just make sure they are NOS from America or Europe, if they are from China they'll surely be fakes...
Make sure you take the appropriate ESD precautions when handling the FET....
You might get lucky, its worth a shot..

 

[vinylisfinyl]

Sounds like a plan! I already found some from a place recommended here for other hard to find transistors, littlediode.com: https://www.littlediode.com/components/3SK41.html

Since I've not replaced FETs before, do you have any tips on ESD handling with these? Last thing I want to do is fry it as I install it...

 

[nosirrah]

ESD is easy, just ground your work area and avoid creating static, like no shag rug at your bench. Touch a ground before you touch a part

 

[rcs16]

All you need to do is short all the pins out, in the old days of non-diode protected MOSFETS, they came from the factory with a shorting ring, that was to be removed after installation. tightly wrap some 30AWG bare solid around the MOSFET to short all the pins out, then remove/install.

 

[Blue Shadow]

Maybe not associated with the tuner problems but the tuner power supply of the 9090DB I saw was burned up, but still working. The dropping resistors just didn't drop enough and all the caps on that small board front top left were looking bad. Maybe check some voltages just to be sure.

 

[vinylisfinyl]

@Blue Shadow: Thanks for the suggestion, I will check again though last time I probed the board it was putting out a steady 13V with no fluctuation or noise to be seen. Unless perhaps 13V is too high? I also replaced all the caps a few weeks ago and did CDFixer's resistor mod to lower the incoming voltage more.

@rcs16, nosirrah: Thanks for the tip! I actually just installed the FET this evening before I saw your post, and took precaution by grounding myself out to my power supply's chassis ground. I also stood on a rubber mat in our cellar so I think I had a safe installation of FET01.

Unfortunately that does not appear to have fixed much. I was all set to yell YABBA DABBA DOO this evening because I had crystal clear FM reception after the FET replacement! But after the 30 minute mark, the static-noise began appearing in the audio again and is now fairly constant. The best I can describe it is a soft popping kind of static as if someone were blowing inconsistently into a microphone. The Signal meter no longer swings as wildly as it did but still moves slightly while the Tune meter moves in-time with the static. Occasionally the static goes away for 10 seconds or so and then comes back.

I retraced the signal path from the Tune meter back through all the transistors and ICs to the front end pack and could see the signal change with the static all the way up to the output of the front end pack. (See image titled Output of Front End Pack). I then started scoping more on that board (F-1519). When I scoped TR01 I could see the signal at the base changing in time with the static. See attached images TR01-1 through TR01-3 to visual this (overlay the three on top of each other in your mind to see what I see, since my scope takes instantaneous pictures). When I tried scoping between C20 and C11 (trying to check the output of TR02), the FM signal died and only came back when I removed the probe.

I recall from my earlier testing that TR02 or TR01 (can't remember) showed 0.6 V from collector to emitter, when I was expecting OL. I dismissed that as an in-circuit test anomaly at the time, but now I am wondering if one of these BJTs is leaky and needs replaced. I really wish I knew what each transistor's function was so I could better understand how their leakiness might be affecting the signal. I think I am going to go ahead and replace these too since nothing else appears amiss on this board.

 

[rcs16]

Sounds to me like a transistor or IC is going bad, so it comes down to identifying which one it is in the chain. I suggest tune to the strongest signal you can find and monitor the signal as it traverses through the IF chain to the detector, that may detect something on the scope.

 

[vinylisfinyl]

Yeup, that is what I have been doing and I know for sure it is on the front end pack, F-1915. Tonight I replaced the other two transistors, TR01 and TR02, with NOS from bdent.com and it played beautifully for about 35 minutes. Then right after that it started the popping and statics sounds again while losing and gaining stereo constantly. I also noticed with the new transistors in place that it acted like FM muting was on all the time even when I had the switch in the OFF position.

I'm beginning to wonder if its a bad solder joint or intermittent coil. Twice now its taken over 30 minutes to see the issue, and until then it plays perfectly..almost like something has to heat up before the issue comes about. I scoped the base of TR01 and could see the signal jump around a bit with the static, I will see about getting a video posted here for you to see what it does.

 

[vinylisfinyl]

Been delayed in looking into this due to work and life in general but this weekend had some time to check out the Sansui again. The other night it started popping and spluttering on FM after only a minute of playback, tonight I tried for five minutes before I had to quit and didn't hear a thing. I tried scoping some more, and found similar reactions at both TR1 and TR2's bases. When I tried scoping before L03 the signal drops completely with the scope attached, so I think the capacitor in the scope lead is affecting the circuit. The hardest part is I'm not sure what I should be seeing around TR1 and TR2, which is making it difficult to decipher what I see, such as the signal grows about 40 mV each the time the signal starts cutting out.

I've uploaded a video to Youtube for anyone to see what the actual symptoms are. I took the video while attached to TP1 on the board: