Sansui AU-X11 Restoration

Visitors from New Zealand arriving today. So any input I can offer, like measuring the power board voltages on my X11, will be unlikely until the middle of next week.

Although... I will drag a hook through some Japanese restorer blogs and see if any data we can use pops up.
 
Visitors from New Zealand arriving today. So any input I can offer, like measuring the power board voltages on my X11, will be unlikely until the middle of next week.

Although... I will drag a hook through some Japanese restorer blogs and see if any data we can use pops up.

It is ok for me, thanks for the prompt reply.
The AC voltages i measured were across the speaker terminals, where i had a difference between left and right channels, with a 50hz sine wave passed as an input to the amplifier.

Thanks again for your help and support.

Have a good day!
 
An update to fellow forum members who are looking for 2ST2121 and 2ST5949 transistors.
I purchased the 2ST2121, from a reliable source, whereas the 2ST5949 were sourced from e-bay seller u-barn.
I some how got fooled by the seller's rating in e-bay and proceeded with the purchase.

Unfortunately the transistors shipped by u-barn are fake and i have requested a refund for the same in e-bay.

Attached are pics of 2ST5949 sourced from e-bay seller u-barn.
fake 2st5949_2.jpg

fake 2st5949.jpg
Please avoid this listing.
http://www.ebay.com/itm/ST-2st5949-TO-3-/361225500034?hash=item541ab7b582:g:wFEAAOSwGvhT55~a

These are the original 2ST212's.
2st2121.jpg

No care has been taken to make the clones, literally it is so easy to make out the clones from the original.
I just hope e-bay will help me with the refund processing.
 
Meanwhile, here are pics of sansui glue over the board just behind front panel.
Had do some post processing of the picture to increase its brightness so the glue is easily visible.

I was checking to see if the board could be removed, by pulling out the PCB, but looks like it is going to be tough, the board can be loosed by removing the screws securing it, but it not come out so easily, i believe the front panel has to be removed to work on this board easily, it is also tough to inspect the board being placed in a vertical manner.

It needs attention for sure, and since i am running the original transistors , i have to work on the black flags with priority.
Now that the transistor hunt being fruitless , it is better to keep the originals in tact.


sansui glue.jpg


Other points:

The power supply boards with sansui glue runs hotter than the heat sinks where the output transistors are mounted.
I have the same problem with the small mm and mc boards located on the right hand side of the amplifier, the small boards mounted onto eq and mc amp mounting board.
Is this some thing i have to look at or is it common with the other AU-X11's and X1's too?
 
I was checking to see if the board could be removed, by pulling out the PCB, but looks like it is going to be tough, the board can be loosed by removing the screws securing it, but it not come out so easily, i believe the front panel has to be removed to work on this board easily, it is also tough to inspect the board being placed in a vertical manner.
This is the worst affected board in my X1 when it came to the glue.

Compared to removing the power modules, getting this board out is straightforward by comparison. On the X1 anyway.....

You need to remove all the knobs, undo the 4 screws to the sides of the front panel, pull that off, you will need to remove one lug to one pot also and the led plug pulls out. There is another black steel panel under the front panel. Remove another 4 screws, that will "drop down" (as there are other smaller PCBS attached to this panel). Slightly unscrew the 2 screws at the top (ie: like the one in your photo) and the top of this board can then be pulled towards you. The bottom of this board is held by three PCB "tags" that sit in metal slots. Easy.
 
Though probably not optimal, Amp8 shows it's possible.



aux1-3-d.jpg


His usual approach
x1-8-2b.jpg


x1-6-9.jpg
 
I have the same problem with the small mm and mc boards located on the right hand side of the amplifier, the small boards mounted onto eq and mc amp mounting board.
I think these boards will run hot yes. not that I can say from personal experience as I'm still some time off getting my X1 running.

All these phono boards (in the X1 anyway) are setup to use the lid to the phono section as a localised heatsink. All my phono boards have heatsinking compound for where they meet, and screw into that phono section lid.
 
I think these boards will run hot yes. not that I can say from personal experience as I'm still some time off getting my X1 running.

All these phono boards (in the X1 anyway) are setup to use the lid to the phono section as a localised heatsink. All my phono boards have heatsinking compound for where they meet, and screw into that phono section lid.
The boards on the F-2772 do run hot on both my X1 and X11.

I own two X-1s and have viewed many X1 restorations online, and the use of thermal paste on those phono motherboard cards is never seen.

I have also own two X-11s and have also viewed many X11 restorations online, and the use of thermal paste on those phono motherboard cards is always done.

So I suspect another tech was in slimecity's X1 at some point.
 
Last edited:
It needs attention for sure, and since i am running the original transistors , i have to work on the black flags with priority. Now that the transistor hunt being fruitless , it is better to keep the originals in tact.

Remember the possibility of using a 'sacrificial set' of MJ21193 // MJ21194 in place of the original output transistors. I did this with my AU-X1 using just one pair per channel and it worked out well for me. You can set bias from the test points (if you put the sacrificial ones in the right socket locations on the heatsinks), set DC offset, and get audio output. Although I wouldn't recommend any listening tests - just sine wave signal generator tests for channel level comparisons and so on.
 
Remember the possibility of using a 'sacrificial set' of MJ21193 // MJ21194 in place of the original output transistors. I did this with my AU-X1 using just one pair per channel and it worked out well for me. You can set bias from the test points (if you put the sacrificial ones in the right socket locations on the heatsinks), set DC offset, and get audio output. Although I wouldn't recommend any listening tests - just sine wave signal generator tests for channel level comparisons and so on.

Will changing the output transistors help with the difference between left and right channel outputs ?
The difference with sine wave is around 1 volt, between left and right channels, measured with a 50hz sine.

I was thinking it could be something else,driver board components, but my knowledge is not as good as other techies in the forum.
 
Will changing the output transistors help with the difference between left and right channel outputs ? The difference with sine wave is around 1 volt, between left and right channels, measured with a 50hz sine.
Unlikely, much more likely to be something amiss on the respective driver board, also that's a very low test frequency, you ought to use 1Khz for these tests.
I was thinking it could be something else,driver board components, but my knowledge is not as good as other techies in the forum.
An AU-X11 restore, with no service manual available, :yikes: is a daunting prospect for any techie.
 
Will changing the output transistors help with the difference between left and right channel outputs ?
The difference with sine wave is around 1 volt, between left and right channels, measured with a 50hz sine.

I was thinking it could be something else,driver board components, but my knowledge is not as good as other techies in the forum.

Unlikely, much more likely to be something amiss on the respective driver board, also that's a very low test frequency, you ought to use 1Khz for these tests.

An AU-X11 restore, with no service manual available, :yikes: is a daunting prospect for any techie.

Yep.

I will be adding to my "Crowd sourced" X11 thread soon.

Without some kind of guide, on an amp this rare ( thus few techs can chime in )... it will be slow going. Doing my best to get that info out there.
 
I did note , that after playing music for 30 minutes or so. and then measure dc offset across speaker terminals, the values are around 30mv for each side, the left side and the right side.
Otherwise when i power on the amplifier, the offset is stable between 0-5mv.
Not sure if this is something to be worried off, i did stop the source from playing, turned down volume knob to zero ( the preamp volume knob with the big dial), but the dc off set was still at 30mv.
The only way i could get rid of it , is to turn off the amplifier wait for some 10 minutes and then turn it on again.

The service manual states for adjusting dc offset let the amp warm up for 10 minutes.

No matter what i do , the DC Off set increases from 0mv to around 30mv after playing music for like 30 to 40 minutes.

On a side note the protector ground in the driver board circuit, in my amp is wired to the copper chassis.
I am not able to look this up, in other photos, as its hard to trace this wire.
Want to know if its the same for the rest of you guys too.

Update :
This time i let the amplifier play for like 30 minutes, and then adjusted the DC Offset to settle at 0mv, and this works fine, but the only issue is that when i first power on the amp the DC offset is around -19mv and then slowly starts increasing to zero.
Not sure whether this is due to aging components, the same applies for DC Bias across test points, it takes a good long 15 to 20 minutes to reach 11mv.
 
Last edited:
I did note , that after playing music for 30 minutes or so. and then measure dc offset across speaker terminals, the values are around 30mv for each side, the left side and the right side.
Otherwise when i power on the amplifier, the offset is stable between 0-5mv.
Not sure if this is something to be worried off, i did stop the source from playing, turned down volume knob to zero ( the preamp volume knob with the big dial), but the dc off set was still at 30mv.
The only way i could get rid of it , is to turn off the amplifier wait for some 10 minutes and then turn it on again.

The service manual states for adjusting dc offset let the amp warm up for 10 minutes.

No matter what i do , the DC Off set increases from 0mv to around 30mv after playing music for like 30 to 40 minutes.

On a side note the protector ground in the driver board circuit, in my amp is wired to the copper chassis.
I am not able to look this up, in other photos, as its hard to trace this wire.
Want to know if its the same for the rest of you guys too.

Update :
This time i let the amplifier play for like 30 minutes, and then adjusted the DC Offset to settle at 0mv, and this works fine, but the only issue is that when i first power on the amp the DC offset is around -19mv and then slowly starts increasing to zero.
Not sure whether this is due to aging components, the same applies for DC Bias across test points, it takes a good long 15 to 20 minutes to reach 11mv.
Sounds like aging to me.

I will be adding that info I promised in about 12 hours.

I will refer the above in my measurements.
 
Sounds like aging to me.

I will be adding that info I promised in about 12 hours.

I will refer the above in my measurements.
Another Question regarding the driver boards, did you desolder all the wires in the driver boards, say the preamp wires and the power supply wires that go to the driver board for an example.

I was thinking, just wondering if i should bullet use connectors for the joints, where de-soldering the boards is done, this ensures easy plug and play operation and the board can be easily removed for any work that needs to be done on it.

To include a bullet connector in the existing wires, this requires splicing the existing wires and a adding a female and male connector to both sides of the wire.

For instances where we need to remove the boards, often or put them in to check this would be an easier approach.

Would like inputs or suggestions on other similar techniques that could be applied.

Another Question again, sorry i just think that i have too much going over my head, have you also measured the resistance over the pots, used , all of them that are there in the front panel.

In this case did you de-solder the wires, or you measured it in the circuit, want to if there's any circuit induced resistance say when we measure directly with it plugged on.
I haven't removed the front panel yet, i am just waiting for couple of de-oxit solution which i have ordered from the U.S.A, a relative of mine will be bring it for me when she arrives here .

Meanwhile going through other restoration posts, and other documentations in audiokarma.
 
Last edited:
This time i let the amplifier play for like 30 minutes, and then adjusted the DC Offset to settle at 0mv, and this works fine, but the only issue is that when i first power on the amp the DC offset is around -19mv and then slowly starts increasing to zero.
Not sure whether this is due to aging components, the same applies for DC Bias across test points, it takes a good long 15 to 20 minutes to reach 11mv
Sounds pretty normal to me.
To include a bullet connector in the existing wires, this requires splicing the existing wires and a adding a female and male connector to both sides of the wire.
On the face of it, it sounds sensible and with many amplifiers I would think this was just about acceptable if there was no other way. However, I wouldn't do this with an AU-X1 (or AU-X11 if I had one). Connectors should be reduced to a minimum, and adding more to these models should be considered an absolute last resort in my humble opinion.
 
Last edited:
Sounds pretty normal to me.

On the face of it, it sounds sensible and with many amplifiers I would think this was just about acceptable if there was no other way. However, I wouldn't do this with an AU-X1 (or AU-X11 if I had one). Connectors should be reduced to a minimum, and adding more to these models should be considered an absolute last resort in my humble opinion.

Hyperion's suggestion is the wisest course.

I have not worked on my X11s.

I bought one recently that turned out to be nuked. So I am using it for amp forensic research before I even begin to attempt recapping my working X11. This second (working unit) is the one from which I got my voltage measurements.

Smurfer77 got elbow deep into his before he temporarily set aside his X11 project.

He actually got a third unit since he took a pause in his X11 adventure.
giphy.gif





He and Hyperion would be the guys to ask. I am a mere hobbyist.
 
Ok, i was going through another X11 restoration post, and @Matbonny has actually plotted a graph of the bias and the DC Offset values.
If you notice the bias remains constant at set value for some period of time and then starts to drop.

In AU-X1 the schematics show a dc bias voltage of 20mv, but whereas the adjustment procedure stays 25mv.

After setting bias to 11mv in the X11, and using the amp for like an hour or more, i noticed the thermal or protection relay engages and the bias voltage drops from set 11mv at idle state to like around 8mv.
Could this be the reason the X1 manual says, adjust it to 25mv, probably or could be after this relay engages, the bias will drop down to 20mv as mentioned in the schematics.

bias.jpg

If you look at the graph it idles at 11mv after 45 minutes.
Now i am thinking whether to wait for the relay to engage and then set the bias at 11mv.
So at start it will actually be higher than 11mv, and after this relay clicks in it will idle at 11mv.

Credits to @Matbonny for the image.
 
Last edited:
Back
Top Bottom