Sansui eight trouble

[blhagstrom]

I have seen (or heard) some cases of random crackling, and they were caused by defective transistors in the preamp section.
In the case of the Eight, that would mean TR7001 through TR7008.

But the fact that both channels are similarly affected is disturbing, this leads to a power supply issue, or an epidemic failure among these transistors (all 2SC693).

Of course, if you can see the noise on your scope by probing in this area as DC explained above, this will point you directly to the culprit.

Yup, I had read (a while ago) that the 2sc693 are believed to get noisy and so when I was getting some other parts I ordered a bunch of replacements and changed them all. No joy. I recently read about "flying saucer" transistors having a bad rap and was looking there in the power supply. I have checked the PS (with a DMM) and the voltage is proper.

I'm hoping the scope works well enough to find it. I may need to look into re-capping the scope.

 

[DCinDC]

Yep, good call, Tichu, since troubleshooting is a little out of reach of the OP...

There are 10 each 2sc693 npn transistors on the preamp/tone board.

If the high filter reduces the noise, and you are not using the phono input for testing, then it is likely one of those.

Order ksc1845 replacements from Mouser.com and replace those transistors. They are very inexpensive. I order them a hundred at a time.

You can use these to replace the 6 phono preamp transistors (2sc632) also. It should make an improvement there as well, with reduced hiss.

The gurus here: Markthe Fixer, Echowars, etc turned me on to these.

The both channels thing is troubling, but these transistors could've gone noisy, although they are not one that's known to be. But time marches on...

 

[blhagstrom]

done

Yep, good call, Tichu, since troubleshooting is a little out of reach of the OP...

There are 10 each 2sc693 npn transistors on the preamp/tone board.

If the high filter reduces the noise, and you are not using the phono input for testing, then it is likely one of those.

Order ksc1845 replacements from Mouser.com and replace those transistors. They are very inexpensive. I order them a hundred at a time.

You can use these to replace the 6 phono preamp transistors (2sc632) also. It should make an improvement there as well, with reduced hiss.

The gurus here: Markthe Fixer, Echowars, etc turned me on to these.

The both channels thing is troubling, but these transistors could've gone noisy, although they are not one that's known to be. But time marches on...

Yup, our replys must have crossed. I have all the ksc1845s in already (in the pre-amp, thanks for the heads up on the phono section). I had found that info from the threads.

 

[DCinDC]

yep
OK
use the scope to tell us how much ripple is on the power supply outputs from board F-1201.

 

[Tichu]

As our dear Sherlock would have said:
"When you have eliminated the impossible, whatever remains, however improbable, must be the truth."

I notice the tone controls are in the preamp loop.
Can you check the pot housing are at ground potential ?
This area is very sensitive to external noise.

 

[blhagstrom]

As our dear Sherlock would have said:
"When you have eliminated the impossible, whatever remains, however improbable, must be the truth."

Sort of how I trouble shoot.

I notice the tone controls are in the preamp loop.
Can you check the pot housing are at ground potential ?
This area is very sensitive to external noise.

It doesn't have pots, it has stepped attenuators but they are at ground potential (verified). It is a "hot" area. Just poking fingers around there to wiggle wires tells me that, you can hear the effect through the amps.


The scope is tough to read. The wave looks nice and clean on the power supply. The output side of the volume control (to the pre-amp) looks clean but the input from the balance pot like there is garbage. The emitters on the first set of transistors in the pre-amp look clean too. The output line from the pre-amp also looks clean (!?!?!) Since I think I see garbage on the balance pot side of the volume, I unhooked the lead from the balance pot to the accessory board (F-1202) and there seems to be an improvement. The block diagram shows the F-1202 in the output path of the amp sooo...

Since the accessory board feeds into the middle of the pre-amp, I'm thinkiing its in there. The is a nasty pop from the low filter when switched and a fair pop in the high fitler switching (don't know if that's any help). The board has been re-capped and I verified that the right caps are there. I think I replaced the BP caps since they were not replaced previously. The only other things there are some resistors, ceramics and a couple pairs of diodes...sooo, I'm wondering about the diodes. There are a pair of variable resistors and I tweaked them to see if they were dirty, no help.

The SM lists the diodes as 1N34A and DS-410. The block diagram shows the F-1202 being mostly the diodes. I'll have to do some disassembly (again) to get at those diodes and see if a test on them shows anything.

Hmmm, the web shows IN34A as a germanium diode. Still around. I wonder if I need germanium in there. The DS410 seems to cross to a 1N4148 (got some of those).

 

[Tichu]

Scoping noise can be tricky, and more so with a low res scope like the Heathkit you got.
You need a very good and short connection to Ground.

The balance pot is directly connected to the input connectors, or to the phono preamp output.
It is "normal" that you should see some noise since it is a high impedance point in case of an unconnected input.

If the volume is set to minimum, the link to F1235 tone board must be quiet and clean.
You should probe the output of F1235 for noise: the "REC OUT" plugs for instance.

Try to correlate what you see and what you hear (headphone is nice for this).
And play with filters and tone controls.


As for F-1201, this board holds lots of switches.
I suppose that you have cleaned them with deoxit or similar booze ?

The diodes you mention are connected to the amp outputs and their job is to detect DC (not the DC-in-DC) and shut down the relay.
Since you already proved that the problem is in the preamp, I would no bother with them.

 

[dr*audio]

It is possible you have a bad ground connection somewhere, or the power switch could be arcing. Do the panel lamps flicker?

 

[DCinDC]

The output side of the volume control (to the pre-amp) looks clean but the input from the balance pot like there is garbage.

This is a red herring. The volume pot is after the balance pot. If it's clean at the output of the volume pot, it MUST be clean at the output of the balance pot. I suspect your oscope use here. So, twist the volume pot up with no signal in, and these two places should match. Remember you're dealing with level changes here and need to adjust the scope (and volume) to get similar traces, while looking for noise.

Since the accessory board feeds into the middle of the pre-amp, I'm thinkiing its in there. The is a nasty pop from the low filter when switched and a fair pop in the high fitler switching (don't know if that's any help).

Remove the face, remove the escutcheon, remove the switch buttons and squirt cleaner along the lower axis of the switch levers to thoroughly clean the push switches. Those switches can cause lots of odd things. They must be clean.

The diodes are as tichu says, but I can't find where 'P' goes from the 1202 board

 

[Tichu]

Oops, I mixed up the DC detection board (F1401) and F1202.
So, the diodes on F1202 are really there for the "balance check" feature.

The "P" other connection is on Switch S10, just on the top right of the FM Tuner schematic.

When selecting "Balance check" mode, the FM center indicator should stay center if balance is in the middle position. (you need to get some sound out of the amp to do this)

 

[blhagstrom]

Scoping noise can be tricky, and more so with a low res scope like the Heathkit you got.
You need a very good and short connection to Ground.

The scope may not be up to it or working well enough. With the probe grounded I can't see anything on the trace. Ungrounded I can see the trace get "fuzzy" at the volume output. I have V-gain at 100x (highest setting).

Another issue is that the noise is transient. It's not a steady thing or constant. At times its very noisy and at times, nice and quite. It almost seems to get better with time, like warm is better, but then it comes back in.

The balance pot is directly connected to the input connectors, or to the phono preamp output.
It is "normal" that you should see some noise since it is a high impedance point in case of an unconnected input.

I have shorting plugs on the phono inputs.

If the volume is set to minimum, the link to F1235 tone board must be quiet and clean.
You should probe the output of F1235 for noise: the "REC OUT" plugs for instance.

I didn't think to probe there. I'll try that. I was looking at the output to the speaker switch.

Try to correlate what you see and what you hear (headphone is nice for this).
And play with filters and tone controls.

Yup, been doing that a LOT but it doesn't indicate anything. No real coorelation to the noise and doesn't change the behavior.

As for F-1201, this board holds lots of switches.
I suppose that you have cleaned them with deoxit or similar booze ?

Yup, about 99% of the things I find with issues are dirty contacts. I clean everything first when I get things.

It is possible you have a bad ground connection somewhere, or the power switch could be arcing. Do the panel lamps flicker?

I've been over the ground connections a dozen times. I've run test jumpers from ground to ground points (bypassing to the boards, etc...)

This is a red herring. The volume pot is after the balance pot. If it's clean at the output of the volume pot, it MUST be clean at the output of the balance pot. I suspect your oscope use here. So, twist the volume pot up with no signal in, and these two places should match. Remember you're dealing with level changes here and need to adjust the scope (and volume) to get similar traces, while looking for noise.

Hmmm, good point. I suppose the volume pot at zero would "clean" the trace.

Remove the face, remove the escutcheon, remove the switch buttons and squirt cleaner along the lower axis of the switch levers to thoroughly clean the push switches. Those switches can cause lots of odd things. They must be clean.

I have it stripped to get to the F-1202. Been over those switched a few times. Can't hurt to look again.

Oops, I mixed up the DC detection board (F1401) and F1202.
So, the diodes on F1202 are really there for the "balance check" feature.

Hmmm, well then they aren't the issue. It doesn't matter where the balance check switch is set.

Having cleaned all the controls and tried every which way setting them and the switches and wiggling all the wires and checking all the traces and solder joints, the only thing that seems to make any difference is with the tone filters ON. That tones the noise down....So somewhere upstream of those???

 

[DCinDC]

Thanks Tichu. I was looking for gain protection ala other Sansuis, I forgot about the Balance Check feature.

ok let's sum up.

- It's not in the power amp, because using the main ins the sound is clear.
- Switching the low and high filters affects the noise, so yes it is upstream of those.
- You replaced the caps and transistors in the preamp.
- You can't see much with the scope. Doesn't sound like it is working at it's most sensitive settings. The Probe MUST be grounded, and you should see something. Try running music (FM) so you can see what a normal signal looks like. The noise should look like about a tenth of the music.

We've covered all the likely suspects. I say pull the mullards out of the scope, sell em and buy a cheap scope that works. You need one to find this.

 

[blhagstrom]

Duh

OK, I spent some time fiddling with the scope and learned how to use it.

I spent some time chasing my tail and learning a little bit about how things would show up on a scope based on what they do.

SO, now that the OP is a bit less dumb, he can get somewhere.

It could be a little more sensitive but maybe it will serve.
It does appear that things calm down the longer I run the unit.
I can see the noise on the emitters on the final transitors in the pre-amp.
I'll probably have to work from cool starts when its most noisy.

BUT, HEY! I'M USING A SCOPE!

 

[blhagstrom]

update

Hi guys,

Well I fished around with the scope and backed from the final output of the preamp (TR-7009, TR-7010) all the way to about the middle of the board (TR-7007, TR-7007). I checked about every point along the way that I could get to. The "static" seemed to be all the way back that far.

Went through all the resistors and checked them in circuit and most read correct. a few didn't but they are in circuit.

I thought I had it cornered at cap C7028 and C7027. Originally those were solid aluminum and they replaced them with electrolytics so I was wondering if that was it. Did some reading on ESR. Solids seem to have very low ESR.

Went through some other mental flip-flops and finally just pulled the connection on the accessory board.

With the low filter unhooked at 7NL and 7NR, the noise is gone. Of course, the low filter doesn't work. I tried other dis-connections but the one listed solves the issue.

I checked the resistors on the accessory board. They are proper at 470K. The switchs can't get any cleaner.

So, I looked over the pre-amp, trying to figure out if there is some sort of feedback type issue there with the low filter circuit connectd. All I got is resistors and mylar caps. R7059 and R7060 on the feed side (to 7ML and 7MR) are right, 3.3K. But that also feeds the high filter side and there is no noise with that in place. SO looking at the output side of the low filter, I checked R7063 and R7064, yup, OK, 1K. I checked R7061 and R7062 and they are good at 22K. The low filter output path goes through C7037 and C7038 but those are original solid aluminum so I discount those as the issue. The transistors were replaced as mentioned before.

SO, I'm left staring at C7035 and C7036, mylar caps on the low filter path side. I'm leaning away from C7033 and C7034 since they are towards the circuit still in use by the high filter.

Haven't read anyting about mylars going bad in pre-amps. Is that a possible problem here? Or is something I'm not understanding about the circuit in tha area (a REAL possibility, I think I understand components, but I'm a bit noobie on whole circuits).

 

[DCinDC]

You're not understanding the circuit correctly. with 7NL and R disconnected the low filter is engaged, even more than normal, since the R7405 is also bypassed. I believe it is effectively just filtering out your noise.

Play music through it like this and you will hear the drastic change in frequency response with those wires disconnected.

This is a red herring. The problem is before this stage.

 

[DCinDC]

If you've got noise on the balance pot, and you're not using the phono input, then the noise is coming from your source.

 

[blhagstrom]

You're not understanding the circuit correctly. with 7NL and R disconnected the low filter is engaged, even more than normal, since the R7405 is also bypassed. I believe it is effectively just filtering out your noise.

Oh duh. Open = infinite resistance.

I don't see noise on the balance pot. I even unhooked at the pot. I also pulled the pre-amp inputs after the volume pot. No change.

The tone controls do not effect the noise. So now I'm thinking its between the tone controls and the filters.....but there isn't much there.

 

[dr*audio]

Sorry, I just haven't had time to look at the schematic till now. On board number F-1235, short point 7A to ground and see if the noise stops on that channel. If so, it's coming from upstream of that point.

 

[blhagstrom]

Sorry, I just haven't had time to look at the schematic till now. On board number F-1235, short point 7A to ground and see if the noise stops on that channel. If so, it's coming from upstream of that point.

Shorted 7A L and R. No change.

Just to clarify the thread. There is noise on both channels but its not the same. So "stereo" noise. It gets better or worse on each channel independently.

With this scope, I have not been able to "see" the noise ahead of about C7028 and C7027 or TR7007 and TR7008. Up that far, I'm not sure I'm "seeing" it. The scope may not be up to it.

As a long shot I tried to remove the tone controls from the equation but disconnecting at 7B,C,D,E,F,G and jumpering them together (B-E,C-F,D-G).
No change.

I'm sure its in the pre-amp but with new electrolytics and transistors, there isn't anything left but the chokes, resistors and mylars. All the grounds are good and I've been over the solder joints and traces a dozen times. I see no noise on the power line feeding the pre-amp.

 

[dr*audio]

As I recall, you said the noise is not there if you remove the preamp - poweramp jumpers, so the noise has to be coming from F-1235 board. I will look at the schematic again shortly. I think you are right that the scope isn't sensitive enough to see the noise. There is another way.