The trouble with having a predilection for older gear is that there is an ongoing need to keep them operating and an acceptable level. My Sansui AU555 is the latest component in need of some love and attention. The preamp has had a low level hiss since I adopted this lovely piece, but it now the small hiss has grown a lot louder in addition to a loud farting that increases in pitch when the unit is powered off. 

It is so serious that the AU555 has been taken out of the system and been substituted with a Rotel preamp and the FU32 valve jobby. Over the school holidays I hope to be able to address the issues with the Sansui - but where to start? I do not want to go and replace every component, but rather track down and remedy the issue. I have already replaced power supply caps, but I do have an inkling that the noise is somehow from the power supply - due to the farting sound that happens on power down.

We are what we pretend to be, so we must be careful about what we pretend to be.

Good luck.

BTW, does your doggie have a bung left eye, or is that just glare in the photo?

Hi papahemi - others may be better to comment but if the noises are common to bot ch's, yes i would focus first at the PSU side.
And yes, those sort of noises make you think about capacitors.

At a quick glance at the schematic, I see also a 2200uF 35V output blocking cap, ea ch.

DarkLantern blog - darklanternforowen.wordpress.com

Man they just dont make them like they used to LOL. 40 odd years old and giving you trouble ! Drop colinf a line, I know he has had a bit to do with old suis over the years.

Sept 27, 2020 17:27:16 GMT 12 Citroen said:

Good luck.

BTW, does your doggie have a bung left eye, or is that just glare in the photo?

My doggy's eyes see nothing, she suddenly and sadly passed away just over a month ago - it was devastating but her eyes were ok.

We are what we pretend to be, so we must be careful about what we pretend to be.

Sept 27, 2020 17:37:44 GMT 12 Owen Y said:

Hi papahemi - others may be better to comment but if the noises are common to bot ch's, yes i would focus first at the PSU side.
And yes, those sort of noises make you think about capacitors.

At a quick glance at the schematic, I see also a 2200uF 35V output blocking cap, ea ch.

The power supply caps have been replaced about five years ago. The output blocking caps are original. I will post some picks of the internals - the output transistors look new. As CGOL states the build quality is impressive. When I got to 50 years old I started to get hissy too.

We are what we pretend to be, so we must be careful about what we pretend to be.

Sept 28, 2020 5:51:49 GMT 12 papahemi said:

Sept 27, 2020 17:27:16 GMT 12 Citroen said:

Good luck.

BTW, does your doggie have a bung left eye, or is that just glare in the photo?

My doggy's eyes see nothing, she suddenly and sadly passed away just over a month ago - it was devastating but her eyes were ok.

Sorry for your loss. My 14 year old passed away about 4 months ago. Remember the good times.

Sept 28, 2020 7:41:51 GMT 12 Citroen said:

Sept 28, 2020 5:51:49 GMT 12 papahemi said:

My doggy's eyes see nothing, she suddenly and sadly passed away just over a month ago - it was devastating but her eyes were ok.

Sorry for your loss. My 14 year old passed away about 4 months ago. Remember the good times.

Yup - I still see her around, I have to look twice knowing that she is not there.

We are what we pretend to be, so we must be careful about what we pretend to be.

I know EXACTLY what you mean!

Hi Papahemi, I’m currently out walking long distance and only have a phone to view the schematic on.
Have you worked out if it’s the preamp or poweramp with problems? Try connecting the 555 preamp out to the valve poweramp to see if you still get the same noises. Are you still getting balanced sound?

AMR-iFi R&D

Hi Colin - Enjoy your walk. No I have not determined the exact source of the noise, tho' as stated earlier I feel that it originates from the power supply due to the increasingly high pitched fart on switch off. I have been doing a bit of reading and looking at the bits and i have my first suspect in the power supply diodes, I am not sure tho' how to proceed - that is identify other likely sources of trouble and replace those too. I will switch in the pre-amp to see if that makes the same noises - I suspect it will.

We are what we pretend to be, so we must be careful about what we pretend to be.

Well bugger me! I switched the valve amp into pre-amp as Colin suggested and no farting! It seems then that the issue is in the power amp. I do have some same spec output blocking caps on hand, but they are quite a bit smaller, the originals having 4 pins (two for support) 30mm dia, the new ones are only 20mm two pin. But I ask myself is this the issue? When hooked up on the table now the farting seems confined to LH channel. Hmmm....

We are what we pretend to be, so we must be careful about what we pretend to be.

Ok good. If the noises are on the left channel poweramp in the 555, measure about 27v across C819 (2200uf output cap) for the left channel and C820 for the right channel. (Assuming B+ of 54v on C002, the 2000uf power supply cap.) If one of the channels differs a lot from 27v you most likely have a transistor problem in that channel. IME the small signal transistors Sansui used can go noisy after a long time and can make all sorts of noises, ranging from just increased noise to major popping and intermittent operation. The tiny wires inside the transistor going from the silicon die to the wires off the plastic case corrode and eventually lose contact altogether.
Otherwise the electrolytic caps will be getting pretty old by now. I’d replace them, especially C803, C805, C804 and C806.
BTW don’t worry about the small size of modern caps compared to older ones from the 70s and earlier. The aluminium foil etching techniques are far better now and you can get a same-spec cap with lower ESR for half the size or less!

AMR-iFi R&D

Yes- I measured 26v across the right hand side and 30+v across left hand side. When i adjusted vr802 the voltage changed and was steady on RHS, but adjusting vr801 gave fluctuating voltage on LHS. I was unable to get a steady setting of 26v, either rising to above 30v or dropping down to 20v and dropping still. I will order replacement electrolytic caps for entire board when I have worked out what transistors need replacing.

We are what we pretend to be, so we must be careful about what we pretend to be.

Nice. Now if you temporarily remove C807 and C811 see if you can adjust vr801 to get a stable 26v on the + of the output cap. Removing those caps disconnects the input transistor from the circuit.
Before you go changing the transistors measure the resistors to make sure they’re still in spec (you might need to lift one end of some of them to get a proper reading).

AMR-iFi R&D

Thanks Colin, removed c807 and c811 and still unable to get stable voltage on output cap. It will settle at about 20v but then with a micro-adjustment leap to over 30v. i am now checking the resistors for spec.

We are what we pretend to be, so we must be careful about what we pretend to be.

Checked all R on LH side for spec and all seem ok bar R837 & R839 wirewound which my crappy meter has trouble reading at 0.5Ω - I get a fluctuating reading down to the target area tho'.

We are what we pretend to be, so we must be careful about what we pretend to be.

Is vr801 itself ok? Measure it while turning it with the amp off. If in doubt just replace it. If you’re still getting unstable readings and the other resistors measure ok, measure the transistors. You can even swap the transistors from channel to channel to determine which one is noisy, starting with TR803. Resistor R829 should have around 0.86v across it at 26v on the output cap. (At 20v it has about 1.1v across it.) Also check D801 and D803.
You can remove the output transistors TR809 and TR811 to see if you get a stable reading as well, even TR807, and the circuit should still stabilise at 26v. (It can’t drive anything though!) Removing TR805 will cause the circuit not to work as the VAS (voltage amplifier stage) transistor TR803 won’t be receiving any bias.

AMR-iFi R&D

VR801 does not seem that well - I measured it as you said and at about the point where voltage leaps from 20v to 30 v so does the R of VR801 from 50kΩ to 100kΩ. While I still have many resistor legs lifted and c807 and c811 out of circuit I will test transistors on the board using forward diode test to see if I can find anything easily.

We are what we pretend to be, so we must be careful about what we pretend to be.

Testing transistors in situ proved frustrating - too hard to get a reliable reading so I have decided to remove them from the board - then the thought occurs that I will be replacing them anyway so the only real purpose of testing is to try and locate what the fault is/was. I am not certain as to the replacement TRs, I can find out the equivalent online for 2SC281, 2SC281 and 2SA561 but cannot seem to locate them at RS Components.

It seems that some transistors have been replaced in the past looking at TR 803 & 805, both supposedly 2SC734, but differing in appearance. I will investigate further when I have them off the board. My dilemma is that I really want to get all pats in one order, but this seems to be unlikely. So far I am looking at all electrolytics and transistors on the main board as well as VR801/2 (I deem it wise to replace the pair) and the 2 x 10D1 rectifier diodes in the power supply. I feel reasonably confident that all resistors and diodes on main board are within spec.
Any assistance in locating the TR and Diode substitutes would be appreciated. This electronic stuff is hard!

We are what we pretend to be, so we must be careful about what we pretend to be.

You're doing well so far...

You may have seen, there are a number of AU-555 restoration-repair threads, eg...

audiokarma.org/forums/index.php?threads/leestereos-restoration-upgrade-of-a-sansui-au-555a.816754/

www.stereo.net.au/forums/topic/286438-sansui-au555a-rebuild-restore-modifications-and-distortion-results-thread/

www.diyaudio.com/forums/solid-state/254023-replacment-transistor-hitachi-2sc2b1-sansui-au-555-a.html

Diodes - (PSU rectifiers D001/002) the popular upgrade is 'ultra-fast (UF) type or UF 'soft recovery' or 'Schottky' diodes - an easily audible sonic update usually.
Check the 'reverse voltage' rating of existing, usually low in a SS amp.

DarkLantern blog - darklanternforowen.wordpress.com

Try Little Diode (UK) for genuine transistors. Make sure you get the gain spec right, eg. 2SC734 in O or Y spec.
Best to replace vr801 and 802, I like 10 turn trimpots, but single turn is ok. Try to get some that fit the original footprint.

AMR-iFi R&D

finding replacement transistors has proved a real ache in the anus. I don't really see why it should be tho' as you would think that there are modern replacements to meet most parameters. I did try Little Diode Colin and they did have original parts, but at £28/pc that didn't appeal. I continue to study datasheets to find something to replace the following:

2SC281(B, C)                           currently (haha) BC237B looks a good bet from mouser or the KSC1815
2SC734(O, Y) or 2SC968         is this not the same as the one below?
2SC734(O, Y) or CDC8002-1   www.audio-high-store.com/product/cdc8002-1-replacement/ found this on a Bulgarian site
2SA561(O, Y) or CDC9002-1   www.audio-high-store.com/product/cdc9002-1-replacement/ same site

This is where I'm at at the moment does anyone have views as to the suitability of the replacements  - any to be aware of - it would be great to be able to source from just the one supplier but I can't seem to find one that has everything I want.

We are what we pretend to be, so we must be careful about what we pretend to be.

Those equivalents would work ok though the hfe isn’t listed for the CDC 8002 and 9002.
KSC1815-Y would be good for TR801 and 802. BC548A or B would work as well. For the drivers you could also use the Zetex / Diodes Inc transistors ZTX653 and ZTX753. The ZTX453 and ZTX553 pair have lower current ratings but are a touch more linear below 10mA. Mouser have all those.
By using modern transistors the sonics will change slightly as well, they all sound slightly different. If it were me I’d use transistor sockets so you can try different transistors to see which you like best. Make sure to insert them the right way around!

AMR-iFi R&D

Transistor sockets! There's an idea - possibly not a good one for this procrastinator - I will look them up. Thanks for your input Colin

We are what we pretend to be, so we must be careful about what we pretend to be.

I use transistor sockets in my discrete class A headphone amp, so you can use your own flavour of transistor for the input stage. Anything from germanium to modern silicon. It wasn’t designed to have vanishingly low distortion, just agreeable sonics.
That’s the beauty of a simple circuit like the AU-555, you can tweak the sound to your taste with different transistors on the input stage.

AMR-iFi R&D

I don't know that transistor sockets would be a go, the holes on the board do not seem to be a match. Colin could you please identify the driver transistors, are they TR807/808?

We are what we pretend to be, so we must be careful about what we pretend to be.

Drivers are TR805, 807 left channel and 806, 808 right channel. They form the first transistor of the quasi-complementary output stage.
The VAS is TR803 and 804 for which you could use ZTX453 or equivalent.
You could use snap-off IC socket pins for the transistor sockets, a single IC socket pin for each transistor leg, instead of real transistor sockets.

AMR-iFi R&D

Hika mā! I learn something new everyday, I am now looking at snap-off IC socket pins, these may be a goer as they also may address the issue of degenerating traces around a couple of holes in the circuit board. Cheers.

We are what we pretend to be, so we must be careful about what we pretend to be.

Just checking in, how are you getting on with your Sansui?

AMR-iFi R&D