Sansui AU-919 Troubleshooting

P

Plimpington2

Super Member
Hi everybody,

I have an AU-919 with a blown fuse. I have taught myself to rehab and even build tube amplifiers, but I am brand spanking new for solid state. Tube amps seem very straight-forward as I feel like it is easy to separate the different stages. I guess that can be done for solid state as well - but they certainly seem more interconnected. So, I am looking for some sort of a foot-hold here so I can just start to get into this thing. We have a unit that instantly blows the power supply fuse.

1) First issue is this: The blown fuse in the unit arrived to me with a 3A fuse at 120V. The service manual calls for 10A at 120V. How EASY would that be if this damned thing normally draws 5amps, and the previous owner just has a hopelessly underrated fuse in it? That would be TOO easy. And so I am skeptical. But I will start there.

2) Assuming the EASY thing above is not the issue, what is the first step to troubleshooting something like this? It seems to me the first step is to isolate the PSU from the load (pre-amp boards, output boards) and see if it's still drawing too much current. Is that a reasonable first step?

I know I am in over my head.
Judd
 
Register to hide this ad
Good advice given, but the first imperative is to get the amplifier powering up (on DBT preferably), without blowing a fuse. (or overheating and sending out smoke signals ;) )
 
Last edited:
kwayde said:
Put the correct fuse in then hook it up to a dim bulb tester and power it up to see if the bulb dims.
Click to expand...
Very bright bulb. So will need to move onto the first step - which I THINK is to isolate the power supply from its load. Is that the next step??
 
Ok, this unit has two transformers - one for the preamp stage, one for the output stage. Each stage also has its own PSU board. I disconnected the preamp PSU board and the light was bright. So I connected that back up and disconnected the output stage PSU board the light is now dim.

So I THINK this means the fault is either on the output stage PSU (makes sense since this has the giant power supply caps on it) or the output stage itself.

For the record, the output stage PSU is dual mono. So then I decided to test the L versus R (actually I don’t know which is left or right - one is forward and one is aft). Turns out the AFT channel is the bad one.

So, I THINK I’ve got this isolated to the AFT output stage PSU or the AFT output stage itself. In the collective experience of people here, is it liked to be the PSU or the output stage itself that is the problem?
 
I never do initial troubleshooting under power. We know it blows fuses so don't make things worse. First, I do an in-depth visual inspection under bright light and with a magnifier, looking for anything burned, discolored or otherwise amiss. I also look for previous work that might have been botched. I look for bulging caps, though it's not something I ever see. Second, I look at anything I can get access to with my DVM, checking resistor values and, with the diode setting, the voltage drop across diodes and transistors. You should know what readings you're looking for. If something is odd, check both channels to see if they match. As mentioned above, the 919 has the black flag caps and should never be powered until those are replaced. The outputs are unobtanium and everything needs to be done to preserve them, assuming they aren't already toast. Usually, if the access is good enough to measure well, I can be fairly confident that when I finally power the thing up, odds are it will work, or at least not smoke!
 
ConradH said:
I never do initial troubleshooting under power. We know it blows fuses so don't make things worse. First, I do an in-depth visual inspection under bright light and with a magnifier, looking for anything burned, discolored or otherwise amiss. I also look for previous work that might have been botched. I look for bulging caps, though it's not something I ever see. Second, I look at anything I can get access to with my DVM, checking resistor values and, with the diode setting, the voltage drop across diodes and transistors. You should know what readings you're looking for. If something is odd, check both channels to see if they match. As mentioned above, the 919 has the black flag caps and should never be powered until those are replaced. The outputs are unobtanium and everything needs to be done to preserve them, assuming they aren't already toast. Usually, if the access is good enough to measure well, I can be fairly confident that when I finally power the thing up, odds are it will work, or at least not smoke!
Click to expand...

This thing has never been touched and it is covered in 40 years of dirt and dust. I have def isolated the fault to one channel’s output stage PSU or output stage itself. However, to do a visual inspection of everything as you suggest I will have to remove the output driver stage and output block en mass. It could be done, but I hate to do it if this is a PSU cap issue. It will be so much easier to test everything if it’s in one piece, I think.

In any event, I am trying to test the main PSU caps first to look for something obvious. None are bulging (of course - I don’t ever see that in tube amps either).

In any event, none seem grossly shorted. After I discharged all of them, they all read out short for a few moments, but then resistance begins to go up (reads out on my multimeter), and the continuity tone shuts off. My multimeter has capacitance ratings too, but I don’t know what to make of them. The caps on the good channel read out at something called 12.5 mF. The caps on the “bad” channel read out at twice that (about 24.5 mF). The thing is, that is NOT microfarads. My meter starts out in the nF, then switches to uF. And then ends up at mF. I don’t know what an mF is.
 

Attachments

  • IMG_2627.jpeg
    IMG_2627.jpeg
    98 KB · Views: 3
CrSens said:
More likely it is the output stage.
Click to expand...
Really? That sucks. So that would be the driver stage or the actual output transistors? I don’t know how to test transistors. I guess I will isolate the driver stage and output stage from the PSU and see what is what. Ugh.
 
You will have to learn to test transistors, slim chances of getting it done without it. There would be lots of quessing.It's not hard at all, 6 way diode test. 919 is very elaborate amplifier, made to be super fast (special output transistors), and is not a good beginner's project. So, take a deep breath, this will not be an easy fix. You can read several threads on 919 repair and restore here on AK, lots of valuable info(and much than I can provide).
 
Hyperion said:
'mF' is milli-Farad - or 1000's of micro (µ) Farad. The old notation for capacitors was often written as 'MF' when they meant µF, and they would write Pico-Farad
Click to expand...
As a milifarad, the capacitors on the good channel are measuring about right (12,500uf), which are supposed to be 15,000 uf. But the other ones are coming in at 24,500 uf, which seems odd. Capacitors don’t gain capacitance as they age, do they?

Anyway, I have to test the output transistors next, I guess. I have to figure out a way to do that with just a multimeter.
 
Testing transistors with a DVM is a fundamental thing everybody needs to know. You're going nowhere without it! Lots of descriptions on the 'net, though they all seem to be awful. This one is so-so: https://vetco.net/blogs/news/test-a...21NKObJyrx1lcLNm57_ioUFrknXsaKjrgUYXjb0-Rr4A9
There are videos that are probably better but I don't have the patience for videos. If you have a few loose transistors, practice until it's second nature.

A DVM on "diode check" (see manual) will usually let you test small signal transistors in-circuit. Power transistors are harder as they may have low value resistors and connections that make the measurements confusing. They often have to be removed from the circuit to be sure. Or, they may be lumps of 3-way shorted out rubbish that's easy to detect.
 
Yes, that capacitance rise is common with age and it won't always be associated with a rise in dissipation factor or anything else. Still, it tells you that changes are going on inside the capacitor and it should be replaced. The common wisdom that "capacitors dry out, the value goes down and the ESR goes up" is not guaranteed!
 
Ok, two of my output transistors are bad. I think. Tell me if I am on the right track. This unit has four output transistors per channel - two complimentary pairs (each with a NMA1012 and NMC1012). Now as far as I know, if I orient the transistor with the pins facing up and slightly offset to the left, the bottom pin will ALWAYS be the Base, the top pin will ALWAYS be the Emitter and the case with be the Collector. Is that right? In any event, one complimentary pair (ie., the NPN and PNP pair) gives me .538V drop between Base and Emitter and also between Base and Collector for the NPN and the same for the PNP (though, obviously, with the leads reversed. On the other complimentary pair, I get something like 0.006V or 0 no matter how I test them, on BOTH the NPN and the PNP. So, two of my output transistors are dead and gone.

So my next question is this - would these two dead output transistors be enough to explain the short in that channel, which is blowing my fuse?

The next miserable issue, of course, is that these are no longer made. Which means I need to find 4 suitable substitutes for each (8 total).
 
Alas, zero is bad and will be the reason for blowing the fuse. The shorted transistors connect the positive and negative supply rails together, which is never a good thing. I know there are some transistors that will replace the originals. The amp probably won't have the same bandwidth, but it should still sound very good. Search for AU-919 replacement transistors.
 
You must log in or register to reply here.
Back
Top Bottom