AU-9500 repair problems

gonzothegreat

Super Member
I ordered a SANKEN substitute 2SA909 and 2SC1586 for the blown outputs (2sa679/2sc1079) in my Au-9500 and installed them Monday night. I wasn't sure if I could get away with just replacing the two outputs that were shorted or if I had to replace all four devices. The idea of dropping another $25 on outputs was not a pleasant one. I'm just glad they weren't SX-1980 outputs - I'm fresh out of Unobtainium (atomic number 666 - look it up for yourself)

Anyways...

I threw in a pair of 7A fuses and fired it up. The left channel was OK but the right (with the new outputs) had 1.5 volts DC on the speaker terminals. Thank goodness I had the sense to use a DMM before toasting some speakers.

I thought about it for a moment and 1.5vdc seems too high to be something I could null out with a DC offset pot. Not even sure if the amp has one. My gut tells me this is due to the different transistors but I don't know for sure. The next time I look at it, I'm going to swap the outputs between channels and see if the problem moves.

Before you ask, no I don't have the schems. As for the other obvious question, the serial number sticker reads 215050. Not sure if that's the complete number as my cleaning job soaked the sticker and might have erased a trailing digit or two :(
 
Last edited:
That amp is an older Sansui. I think it followed the AU-999? Does this amp have a capacitor output, single + supply and class A input transistor? I am asking because I don't know this for sure.
 
Don't know where this beast fits on the Sansui timeline so I can't answer that one. I do know it was TOTL sometime in the early to mid 70s. Weighs a bloody ton (or about 50lbs worth of a ton).

I haven't done much with this amp except clean it (candle wax and white paint to highlight knobs) and toss in the new silicon. I do like the construction - shielding cages to protect each stage and the driver cards were on plug-in connectors so you could just swap the board and be on your way.

Its probably a dual supply DC coupled amp judging by the NPN-PNP paired outputs. Not sure about the driver stage though.

Anybody got a weblink to some schems?
 
Last edited:
I did some searching and came up with little to nothing. It looks like this amp has differential input, split power rails and capacitor less output. Maybe someone has the schematic for this.
 
Last edited:
I was pretty sure about the DC coupled when I heard the output relay click on.

I just checked the thread about the 8080 with low output on one channel and I had an idea or two. Since the driver cards on 8080s often lose some resistors and a Q or two when outputs short, I should check that before I start chasing ghosts (i.e. blaming the Sanken subs). I know I checked the transistors on the two driver cards after I bought it and they checked out OK. Nothing looked obviously burnt or broken but now I'm suspicious.

I haven't hooked it up to a scope either so I don't know if there is even any output. I wonder if the 8080 with the bad channel has any DC on the outputs?
 
This amp have Diodes in the driver bias area? Check those for operating condition and check the drivers carefully or it could be at the inputs like you suspect? Where is EW about now.
 
Never seen this one...but if there is a volt-and-a-half on the output, it can't be cap coupled.

Dude, you can flail about with this for weeks, or you can pop for $20 for the manual.

http://www.agtannenbaum.com/s_cat.html#SANSUI

The drivers and the fused resistors need to be checked (Sansui loved those damn resistors...) Also, a 1.5 volt offset is a lot for a mismatched input pair, but still quite possible that that is the root cause.

You can check for audio also. 1.5V offset is not enough to smoke a speaker (about 300mW into 8-ohms), but I'm finicky about my speaks, so I'd connect just long enough to se if I had output and determine if it sounded like a full-wave signal, or highly-distorted half-wave...
 
I am not absolute on this model - but I can say that Sansui was trying to be different and innovative in most of their circuit designs. The way I would handle this one without the manual is very carefull comparison of the good channel to the bad. Your looking for shorts, opens and/or weak components. Go right through the amplifier stages - output to input - draw a diagram and write down resistance and then voltage readings for each channel for comparison. DC amplifier circuits can be difficult to voltage troubleshoot as one component can throw off the whole stage and then there is the feedback circuit trying to compensate. Even without a manual - I would bet this circuit was similar to other makes/models of the era. Dig in there and pull out a loser :D
 
A little progress...

I swapped the driver boards between the channels and the problem didn't move. I've also checked a small pcb under the output heatsink with four power and four 1/4w carbon 47 ohm resistors and they measured OK.

If the problem was on the driver card I'd order the schems in a minute. I'm going to dig around a little more before I grab the schems.

I definitely won't order them from AG Tannenbaum. I bought my Altec 9440a schems from them and they were CRAP. :dammit:

They scanned them from poor xeroxes and while the text is readable, the circuit diagrams are nearly useless. What were supposed to be straight lines look like morse code. Has anyone else had trouble with crappy schems from AG Tannenbaum?

So, what's next? I'm going to swap the new transistors between channels to rule them out. After that, I'll probably order the schems.
 
Have you disconnected the pre-amp by pulling the jumper or disconnecting the wire on the bad channel? Then take a reading - Is the problem in the pre-amp or voltage supply to amp or outputs?
 
Definitely in the power amp

I considered that as I wrote my last post but I didn't bother including it. Yanked the jumpers - no change. Its definitely in the power amp stage.

Now I need to swap all the outputs to the opposite channel. Not only will this rule out the new transistors but it will also check if the other two outputs are really OK. It may be possible that the other half of the bad channel didn't escape unscathed. After that, it's time to fire up the old noggin and dig in deep.
 
Last edited:
They scanned them from poor xeroxes and while the text is readable, the circuit diagrams are nearly useless. What were supposed to be straight lines look like morse code. Has anyone else had trouble with crappy schems from AG Tannenbaum?
Yep....I complained, and they sent me a beautiful copy.
 
I must be #@$*&% blind!

I've missed some obvious stuff in my time but boy this one takes the cake. I swapped the transistors tonight and the problem didn't move. I pulled the bottom cover and started looking around. I had opened the bottom before and didn't see anything unusual. Pretty clean. A little dust, some sawdust looking stuff and a little gray strip of something. I figured it was just some weird booger that crept inside and ignored it.

Waitaminute - there's a lead over by the point-to-point wiring going to the driver board. Its just hanging off a terminal. I smell blood in the water! Sunnuvabitch - A popped 4.7uf 100v cap. It puked its guts when the transistors shorted. Look in the center of the pic - its standing up with an aluminum foil dingleberry pointing skyward.

In hindsight, I see the fault in my troubleshooting. I was raised fixing cheap receivers and they didn't use any point to point wiring. Strictly PCBs. The wiring of this section looks like an old tube guitar amp. I troubleshot the driver and output boards and ignored the possibility that there were other components hiding underneath.

I couldn't find a replacement in my parts box so I need to order some from MCM or Digikey. Maybe I should order one of those boxed assortments of caps. I ordered a 5lb bag of caps from someplace once (Jameco) and ended up with 5lbs of obsolete, low voltage, high ESR, 85 degree CRAPacitors.

A shoutout & thanks to everyone for their two cents and a big hand to Stene for posting the alignment instructions.

EW - I just emailed Tannenbaum about the crap schems. Hopefully they'll straighten it out.

OK folks, show's over nothing to see here. Move along...
 
Last edited:
Not quite fixed...

I ran down to the local electronics shoppe and grabbed a pair of 4.7uF 250V axials to replace the two caps in the pic. Powered it up and checked the DC offset. About 250mV - yay! I hooked up a cheap tuner and speakers and took a listen. Ugh, tons of distortion in the bad channel while the other sounds great.

I have swapped the output trannies so I know that isn't it. Same for the driver cards. The DC offset is rather unstable. While the left channel is rock steady, the right drifts up about 5mV/minute. I need to check the small transistor mounted on the output transistor heatsink. Is this the bias transistor?
 
Looks like the BA-3000 amplifier is from the same era. It has a one transistor 2SC1364 per channel mounted by the output transistors - This is the bias transistor on this one. I think you are getting close to solving this puzzle.
 
FIXED!

When we last left our hero, he was distorting badly on one side after he replaced a blown cap and a matched output pair.

I was about to order parts from MCM so I checked every component on the output boards. Nothing measured funny and everything was within tolerance compared to the other channel. I was stumped. I sent off the order to MCM and stuck the 9500 back in the corner.

Then about a week later, I dug it up and on a whim I unscrewed the fuseholders on the back. WTF? One of the right channel fuses had popped. At first I thought I had forgotten to replace the fuse when I put in the new outputs but the markings on the blown fuse matched the ones in my parts bin. I tossed in another one and powered it up w/o speakers. Checked the fuses - OK. Checked the output offset - OK. Hooked up a recently fixed Apex DVD player and tossed in some Clapton.

ITS ALIVE!

ITS ALIVE!

ITS ALIVE!

Near as I can figure, when I powered up the amp before I replaced the blown cap, it fried the fuse. After I recapped it, the blown 7A fast acting fuse kept half the amp from working.

So remember to recheck what you thought you checked.
 
Back
Top Bottom