• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

New to me AU-666

Never mind! I had one of the filter caps backwards. Got faked out by the black = negative and I'm too used to the single rail topology after working on all those 222s/555s/555As...


Got a crackle in the L power amp section.. is it liable to be those flying saucers? And these trimmers are jumpy. Guess I'll be doing another mouser order...
 
Last edited:
bhairavi said:
Never mind! I had one of the filter caps backwards. Got faked out by the black = negative and I'm too used to the single rail topology after working on all those 222s/555s/555As...


Got a crackle in the L power amp section.. is it liable to be those flying saucers? And these trimmers are jumpy. Guess I'll be doing another mouser order...
Click to expand...
Flying saucer transistors can go noisy and I found the trim pots to be very temperamental on this series
 
It looks like every one of these can work in place of the CDC 8002 saucers. I think I might have one or two at home...

KSC2690
KSC2383
BD139
ZTX694B
TTC004B


KSA992 replaces the XA495 at TR801/3.
 
Weirdly, crackle was gone today, even before I did anything else.

I have just enough 992s and 2383s to do both amp boards! So TR801/3 (inputs) and TR805/7 (flying saucers) have been replaced.

But after replacing them on the L side, still not able to dial in either the bias or the DC offset.

Bias only adjustable between 80 and 160 mV across one emitter resistor. (Prior to transistor substitution it was sitting higher at 90ish mV)

DC offset hovers around 500mV. I tested the trimmer resistance and it tracks smoothly for both VR801 and VR803.

Any ideas?



I checked a bunch of voltages on the preamp and they all seem fine (all these 871s have been replaced with 1845s):

1785441899113.png


Power supply:

1785442060919.png


The negative rail is not what is specced but every 666 thread I've seen people are getting the same integer value as the positive rail, so I think I'm fine there.
 
I forgot to mention that the amp SOUNDS fine. Even sound both channels.


Maybe the bias transistor? I could try replacing with an 1815...
 
Well I tried putting in an 1845 in place of the bias transistor TR809, it didn't change anything with regards to the bias.

Bias fluctuates wildly with any adjustment (which may be a trimmer issue), and won't go lower than 70 mV or so (I don't think that's a trimmer issue).

When bias fluctuates then the DBT also cycles. It's not going into oscillation but it'll go back and forth once or twice. It also lights up bright briefly when I first power on, unless I remove the L amp board.

I guess the next step is simply to replace the other transistors? There's really not much left, just the drivers and the outputs...

Current state of the board - green dot = replaced

Is it weird that Sansui does not provide any voltage checks for anywhere on the amp board?

1785461092291.png


Driver replacements: TTA004B/TTC004B or the Fairchild KSA1220/KSC2690
 
Last edited:
I’ve used ksa-1220/ksc2690 with success here or bd139/140

I’d keep the original bias transistor in place as it needs to be secured to the heatsink I believe if it must be replaced you can still find NoS replacements or ksc1815 may be a better sub - I’d change the trimmer first and recheck as I found it having stability issues in the past with the 3 x 666 I’ve refurbed .
 
Yeah so I was futzing around trying to figure out what exactly was the issue. I checked the output transistors and it seemed the original SD188 had low hfe, so I used a spare SD188(A) in its place - and R843 let out some magic smoke. Went back to the older output and tried sticking the R amp in the L socket and it played well, then poof. Checked again, outputs are fine, the bias transistor tests good, and when I desoldered those trimmers they fell apart. No wonder connection is intermittent. But also one of the drivers on both sides were blown.

So I'm hoping bad trimmers and bad drivers is the answer.

I have a local tech whos going to sell me some of the Toshibas tomorrow. I might be able to get away with not having to wait for another mouser order...
 
New drivers and trimmers in.

Bias barely budges, although dc offset could be set. However when dc offset is set to where it should be, then bias totally disappears (0mV).

Sound is distorted.

I'm going to go cry now.
 
I checked was once again that the output transistors are all functional.

The 2sd188 have hfe in the 20s, the 2sa627 hfe is in the 70s, both sides.

Also checked the diodes on both amp boards.




Another thing I observed was that I only register voltage across the emitter resistor going from the PNP to ground (green arrow), but not across the one going from NPN to ground(red arrow). I just put that together.. maybe my NPNs are bad? The listed min hfe is 30, and these were 22 and 24.

1785528797817.png




Could I try using my SD188(A) in place of these SD188s? The A version just has higher voltage handling, so it should be fine, right?

@Hyperion @mbz @ivandezande :D ?
 
Last edited:
Can you summarise what transistor subs you’ve used in which locations and did you double check transistor orientation
 
quadmatrix82 said:
Can you summarise what transistor subs you’ve used in which locations and did you double check transistor orientation
Click to expand...


Changed the bias transistor back, here's the current state of the boards:

1785537313385.png

Definitely did recheck transistor orientation when I put them in, tested all new ones on the tester before sticking them in.

Will check them again and trial using the SD188A in my possession.
 
Last edited:
UPDATE:

1. rechecked all transistors, they are all ECB and oriented correctly on the pcb
2. tried changing the SD188 to the SD188A, no change
3. tried changing out the bias transistor for an 1815, no change
4. tried changing both outputs to new MJs, no change

Both channels, rebuilt board gives very distorted sound. Both channels, DC offset can be set correctly. Voltage across one of the emitter resistors shows as 0-1mV, the other is 0mV.



I do want to mention that in order to mount/unmount the outputs from the heatsinks, I've had to screwing/unscrewing the heatsinks from the chassis, and twice when I've done this (with the amp off) I've had sparks on my screwdriver. Never had that happen on an amp before...

Some kind of grounding issue? But voltages all across the board are good. There is no hum or hiss.
 
Last edited:
I think you need to pause and check for safety check that the chassis isn’t live recheck the filter cap work you did check the power switch is ok and cable
 
quadmatrix82 said:
I think you need to pause and check for safety check that the chassis isn’t live recheck the filter cap work you did check the power switch is ok and cable
Click to expand...
Yup I've been looking at all the connections. I didn't see anything. Of course these amps don't have a ground wire to mains. I looked it up and it's normal to have AC voltage relative to earth ground on an ungrounded chassis. The current is negligible.

None of which solves my problem....
 
This amp is missing these diodes:

1785561255251.png

They are supposed to be located at the bottom of the pcb slots, but don't exist. Someone else had this variant, it appears, from searching AK.
 
Alright, figured it out. R843, the driver emitter resistor, was blown on both sides.

After sticking new ones in there, we have easy adjustment of offset and bias.


Hit a couple snags, first I didn't realize I was still on DBT and kept getting voltage sag when I turned it up. But then once I took it off DBT it still sounded distorted at high volume. So I stuck it on the scope and I was seeing some crazy waves... and after questioning my sanity realized it was because I had large harbor freight magnetic catch-alls sitting on top of the CRT!


SO, now I've checked the preamp and power amp on both channels. Preamp and R poweramp show nice clean sine waves, while L power amp shows this:


20260801_154551.jpg20260801_160557.jpg
 
You must log in or register to reply here.
Back
Top Bottom