DC Offset/Bias adjustment on Sansui AU-555a

canuckaudiog

On a quest for high fidelity
Someone shared this with me and I wanted to share it with the group here :thmbsp:

Being a cap coupled unit you can't measure DC offset at the speaker terminals. But there is another way.

To adjust DC offset, put your positive lead on this leg of R847/R848, and black to ground, and use the trim pot marked to adjust the voltage to 28 volts:

28vbm10.jpg


To adjust bias, put your positive lead on one leg of R843/R844 and the negative lead on the other leg (best to have alligator clips for this). Let the unit warm up for at least 10 minutes before doing this. Put all settings to OFF, balance to the middle, no input selected, etc. Adjust with the trim pot closest to these resistors and adjust it to 10mV.

I did mine and it sounds quite a bit better and it was a simple adjustment. Of course, always exercise caution working inside electronics that are hot (live, turned on). You don't wanna fry something or electrocute yourself.
 
its not exactly dc offset you are setting it is the centre voltage ..
its best to set centre voltage with a scope for equal clipping ..setting half of rail voltage does get you in the ballpark though .
 
its not exactly dc offset you are setting it is the centre voltage ..
its best to set centre voltage with a scope for equal clipping ..setting half of rail voltage does get you in the ballpark though .

A scope eh? I have one - so do I just set it to the same test point I did on the DMM and monitor for a clipped signal?
 
clip onto speaker terminals ..
the full instructions are on page 8 of the service manual .."main amp-section output adjustment "
 
Hi,

I think that you will find that the centre voltage adjustment shown above was done for people that don't have access to an oscilloscope, it wasn't meant to replace the service manual procedure as such.

Cheers

John
 
One question, for the dummy load, it says to use 8 ohms of at least 50 watts rating. So, if I buy an 8ohm resistor rated for 50 watts, will that do the trick? Or am I not understanding what a dummy load is correctly?

I've got an oscillator that'll do 1kHz so I have that covered.
 
Bumping this old post to update this info...
After setting my bias to 20ma as per service manual, I decided to check this with the OP 's info of adjusting bias from resistors as opposed to the fuse and found that 20 ma equates to 5.7 mv . So 10 mv would actually be raising the bias to somewhere in the 35ma range. At least on my unit this is the result....
Charles
 
This is typical that you will find a discrepancy. They don't convert so you must choose. I choose that if Sansui wanted 20ma across the fuse holders, then that's the right setting, even if that measured and converted at the emitter resistor comes out to a different number.

Having said that, keep in mind that the variance on those resistors are 20% - so a 10mV setting could via the 0.5ohm resistor at +/- 20% yield a bias from 16-25 mA. For that reason Iam not too religious about bias as one exact number, at least on the older gear.
 
Last edited:
Bumping this old post to update this info...
After setting my bias to 20ma as per service manual, I decided to check this with the OP 's info of adjusting bias from resistors as opposed to the fuse and found that 20 ma equates to 5.7 mv . So 10 mv would actually be raising the bias to somewhere in the 35ma range. At least on my unit this is the result....
Charles
Following the service manual I can't set my bias to 20ma. I only have range from 7-8ma. However adjusting bias from the emitter resistors (R843, R844) I can set to 10mv. How can this be?
 
Following the service manual I can't set my bias to 20ma. I only have range from 7-8ma. However adjusting bias from the resistors (R843, R844) I can set to 10mv. How can this be?
The emitter resistors may have a discrepancy? If they are outta spec your measures are not accurate. Seems as if some components are not in spec somewhere?
 
Following the service manual I can't set my bias to 20ma. I only have range from 7-8ma. However adjusting bias from the emitter resistors (R843, R844) I can set to 10mv. How can this be?
Quite possibly your 'current range' fuse is blown, preventing you from measuring current properly. When measuring bias across the emitter resistors you are using a voltage range on your meter, most (but not all) meters when used on a voltage range, are unaffected by a blown current range fuse.
 
Quite possibly your 'current range' fuse is blown, preventing you from measuring current properly. When measuring bias across the emitter resistors you are using a voltage range on your meter, most (but not all) meters when used on a voltage range, are unaffected by a blown current range fuse.

Nice work Hyperion. After checking the the current range fuse, on my Klein MM400, I can conclude that it is indeed blown. I love it when it turns out to be something simple like a blown fuse. Off to the local hardware store to get the replacement fuse 500mA 600VAC
 
Oh no!! I was just executing the above procedures, adjusted the DC offset just fine, but when attempting to adjust bias I was successful only in creating a small spark on the board & blowing the left channel fast-acting fuse (F002 or F003).
Not surprisingly, the associated resistor now measures zero.
Possibly unwisely, I decided to test with audio at low level since the unit was powered on anyway. Right channel sounds fine. Left sounds completely distorted/granulated.
Is it likely that damage is limited to the blown fuse or is there a chance that all sorts of stuff is now fried? Possibly D001/2?
I'm planning to recap/restore the unit, but don't want to fail to replace something faulty that will just cause problems- maybe fry newly installed components.
 
Pop in a new fast acting fuse at same amp rating and hopefully the fuse did its job. A blown fuse will typically give you distorted sound, a blown output is dead silent...no pun.

Use mine grabbers next time ;-)
 
Pop in a new fast acting fuse at same amp rating and hopefully the fuse did its job. A blown fuse will typically give you distorted sound, a blown output is dead silent...no pun.

Use mine grabbers next time ;-)
Ok good to know there's hope! For some reason I was testing on the solder side- there's a chance I touched something I shouldn't have which got me worried I'd fried all sorts of stuff. Fingers crossed it's just the fuse. Note taken- I'll be getting some grabbers soon. Thanks!
 
I'm now thinking I may have been measuring the wrong legs of R847/R848!
Without that crucial pic in @canuckaudiog's op I'm a bit lost....
If anyone can clarify I'd really appreciate it!
 
Similarly, going off the pdf version of the manual for main-amp section current adjustment, it says to test with leads on B1/B2 positions. Are these the ends of the fuse holders? Which is which?

For some context, I'm a newbie.
Immeasurable thanks to all the knowledgeable people posting here!
I just finished recapping the unit, also replaced all the obvious transistors (paying careful attention to pinouts) and trim pots. I essentially followed @Leestereo's restoration/upgrade but opted to keep values as close to original as possible. Several e caps went in before I learned I should be double checking e-cap polarity. I did find one mis-printed Nichicon about to go the wrong way on the eq board! Now I'm worried I might have missed some earlier....

I decided to restore because one channel sounded thin; no bass.
Sounds much better now (not perfect), but the heat sink seems to be getting fairly warm pretty fast even at very low volume.
I should be receiving new output transistors, thermal washers & compound tonight.

Also, can anyone confirm the procedure for testing center voltage? I know I've seen a thread on it but can't find it now.
 
Begin by checking and setting the offset / center voltage in each channel to establish some stability. Connect one lead of your voltmeter to a good chassis ground and the other meter lead to the 56VDC bus/rail voltage from the power supply. Make a note the voltage, and calculate what half that voltage is, exactly. With the meter negative lead connected to a good chassis ground, connect the positive meter lead to the positive lead of the output electrolytic capacitor for the channel being tested. C819 is for the left channel, C820 for the right. If those are difficult to access, those are also connected in parallel to the collectors of TR813 (left) and TR814 (right). Adjust VR811 and VR812 (left/right channels) to obtain a stable voltage of exactly half of the DC power supply voltage, or as close as possible. Then connect the voltmeter between the output capacitor positive lead and the 56VDC power supply bus. If that voltage is the same as the previously measured voltage, the center point should be acceptable.

It would be easier if two voltmeters are connected simultaneously, one between the output capacitor positive lead and chassis ground, and the other between the output capacitor positive lead and the 56VDC power supply bus. Adjust the trimmer potentiometer to obtain exactly equal voltages between those two points.

After setting the center point voltage / offset, move on to checking and adjusting the bias / idle current with zero signal and zero volume. There are two 2A output protection fuses in the unit (F002/F003). The fuses can be removed and a multimeter set to a milliamp scale can be connected in the place of one of the fuses at a time. With the unit at full normal operating temperature, measure the current and adjust VR813 / VR814 (Left channel / Right channel) to obtain a 20mA (0.020A) measurement.

The better method employs an AF wave generator, dummy loads, and an oscilloscope to monitor peak clipping at both ends of a sine wave for dynamic testing of the output stage balance, but for safe operation the current measurement is sufficient to prevent overheating.

Once that is set, the offset / center voltage can be measured again and readjusted as necessary.

The procedure is outlined in the factory service manual, in the "Adjustment" section.
 
@Wathour thanks so much for this. Very helpful!
If I understand correctly, the bus/rail voltage for each channel is taken from the collector (the case) of one of the output transistors. Does it matter which one?
When testing fuse holders, does it matter which end the positive lead connects to?
 
Back
Top Bottom