Can I fix a Sansui au X701?

So, I just recieved the components, and I'll be putting them in now.

What do I do about the trimmers, should I measure anything and adjust them to what the old ones are set to, to have that as a starting point when I begin adjusting the DC offset/bias?
 
So, I just recieved the components, and I'll be putting them in now.

What do I do about the trimmers, should I measure anything and adjust them to what the old ones are set to, to have that as a starting point when I begin adjusting the DC offset/bias?

You could do, but unless you do this right you could give yourself a problem. If you have remembered how to adjust DC offset on these when protection is operating - I say leave them alone - if they are Bourns type they will be centered as supplied.

Just bear in mind that after you have changed the trimmers, both DC offset AND bias will be off until you set them again. Be quick about setting the DC offset and do not delay setting the bias :zoom:

If you cant set the DC offset then get the bias set first.

Afterwards monitor the heatsink temperature :yes:
 
Okay, I have adjusted nothing yet. I have changed the 8 largest caps now.

Can I adjust both DC and bias correctly, while the amp is in protection?
 
Okay, I have adjusted nothing yet. I have changed the 8 largest caps now.

Can I adjust both DC and bias correctly, while the amp is in protection?

For the DC Offset you will need to be measuring the amplifier output BEFORE the protection relay.

The Bias adjustment will be fine.....I would get the bias right before you adjust the DC Offset.....
 
You could do, but unless you do this right you could give yourself a problem. If you have remembered how to adjust DC offset on these when protection is operating - I say leave them alone - if they are Bourns type they will be centered as supplied.

Have you understood how to adjust DC offset when you do not have the convenience of the speaker terminals to measure it at ?

Okay, I have adjusted nothing yet. I have changed the 8 largest caps now.

Can I adjust both DC and bias correctly, while the amp is in protection?

Yes you can try - see above.

Also Kev is correct to say that adjusting the bias FIRST might be a better way. - this will be the first of a few times that you will need to set/check the bias - the last time (as described in detail before) when the amp is fully warmed up.
 
Last edited:
I actually am not sure about either adjustment, neither before or after coming out of protection.

Now the first thing would be bias, while in protection. How do I do that? A visual help would be great.
 
Do you have the service manual ?

Can you understand this procedure ?
 

Attachments

  • Bias_AUX701_901.jpg
    Bias_AUX701_901.jpg
    67.6 KB · Views: 77
Last edited:
I think so. I measure between + and - for each pair? Being the outer leads of each of the big white resistors?
And this will work while in protection?

Edit: Done with the replacements, here's a pic from before everything was soldered on, so everything is kinda crooked.

DWf2nO.jpg
 
Last edited:
Very nice, that's clean and clear.
Now, start with the procedure for bias:
- connect the voltmeter on the rectangulars, adjust do 17mV. Keep in mind that you need to made several turns of the bourns to make first reading on the voltmeter.
- after obtaining readings for all four test points of roughly 17mV, proceed to DC adjustments
- connect the voltmeter on the TP1 and TP2...first between out and out, on the 5-pin connector (pins 2 and 5 as seen on the picture), and adjust at minimum. This is DC hot-cold, adjust with 100 Ohms trimmers, right and left trimmers on the most upper corners of the picture
- reconnect now to "G" and one at the time "outs" (pins 2 and 3 and 3 and 5), also adjust at minimum obtainable, but now using the 500 Ohms trimmer, the other one(s) on the upper corner.
- repeat for the other channel...

When done, at this time, the protection circuit should be released (an "click" should be hear).
Repeat ll the procedures as you want - either as described, or on the speaker terminals, but now after preheating as described previously
 
Yep, correct - you should easily be able to get it below 10mV - but 0V is better :thmbsp: -(note this is DC offset we are talking about)

John, I think this is the right place for a small update:

Although on the older Alpha models (such are X901 and X701), where the service manual describes DC at <3mV on speaker terminal (easily obtainable for hot-cold, but very difficult for hot-gnd and/or cold-gnd), the more late models of Alpha suggests any values for >30mV is OK.

I checked my Alpha907MRX, and for hot-cold it can be adjust do <3mV, but as in older Alpha's, hot-gnd works well if less than 30mV, but it is impossible to reach <10mV - there is always some leakage from plus to minus voltage.
But, it is obviously just sufficient.

That's the reason why a classic VU voltmeter i is better for this kind of adjustment, otherwise it is very frustrating (especially when Sanusi fans want to do everything as without compromise :yes:)
 
So, I was adjusting on the first of the 4 Bias adjustments. I never saw more than 0.1 mV. And now stuff blew up.

I was measuring between the 2 blue arrows. As far as I saw, at least the two resistors with red cross lit up and died. :( The big heatsink as well as the white resistor did get rather hot. I was only at it for like 30 seconds before this happened.

x8Yt4V.jpg
 
Wow:sigh::tears:

There is some shortened transistor, almost for sure :-(
Probably the pair on the line with the marked parts (emitter resistor of the driver, and emitter resistor of the output transistor, and probably something else)
Uhhh...now is the harder part of this job...finding the culprit, and this include desoldering of the npn-pnp OT, at least only one pair of the respected channel, and for sure the drivers

Fortunately, the transistors are easily available, Sanken C3284/A1303, but the drivers are a bit tricky - those Toshibas C3298/A1306, are these days obsolent and too much counterfeit around; the nearest (or even exact) replacements are Sanyo's A1606/C4159.
 
Last edited:
There's no way it can have been me, turning the trimpot too much either way? Since it just happened to be parts that had something to do with what I was adjusting.

I'm also using a way too cheap Multimeter, that I just found out, was not even working (probes not properly connecting, have fixed it now)
 
There's no way it can have been me, turning the trimpot too much either way? Since it just happened to be parts that had something to do with what I was adjusting.

I'm also using a way too cheap Multimeter, that I just found out, was not even working (probes not properly connecting, have fixed it now)

No, don't worry...and don't blame yourself
The probable scenario was that something initially was bad, or - this is important - you've made some less than perfect soldering during changing the caps and trimmers.
Carefully check and recheck every thing you've done, especially the polarity of the caps (I know, this is trivial, but not rare mistake);
Check the output, driver, and pre-driver transistor, and surrounding resistors. Desolder everything in line and then check again...this is now more CSI job and you need to be patient
 
Less than perfect soldering sounds probable, what do you mean, shorting between pads or something else?

What I meant with the multimeter is that because it wasn't showing any changes, I might have gone all the way to minimum or maximum on the trimpot.

When I noticed that the big white (what is this actually called) was getting hot, I checked all the other 3. None of them were even warm. Only the one I was adjusting for.
 
Last edited:
That's a shame - a duff meter !

You might be lucky and that resistor R67, 120Ω ½ watt NIR, might have opened before more damage was done.

Do you have a working meter now ?.... check the 'big white' resistor - it should read a very low resistance 0.44Ω on the outside pins, compare with one of the other 3. (let's hope it isn't open).

You will need to know how to test transistors with a meter, you might have lost a pair of output transistors by this mishap :sigh:

Now is the time to decide if it should go to a repair shop, or whether you want to continue to try and fix it.
 
The big white one is dead, just measured. I even saw it light up as well.

I don't know about transistors, not even sure the meter is capable.

Would it be totally silly to try and replace the for-sure-dead resistors (r67, r69), and then try again, with the actually working meter? Would this be too risky, if other parts died?

Would it be anything close to a conclusive test, if I change the two, and I am then able to adjust the bias, and so on, without anything else dying?
 
Back
Top Bottom