• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Refurbishing and upgrading the AU-D11 II

P

phonomac

Active Member
I finally got round to working on my AU-D11 II (having redone the AU-D9 to use while the D11 II was on the bench). As a starting point I reviewed all of the threads that I could find on AK dealing with the D11 II, as a result of which one of the decisions made was to upgrade the whole amplifier to metal film resistors, as per LWBs posts in the DIY forum.

Reviewing the threads also brought up a number of issues, some of which I already knew about, and I decided to address these points and try out some solutions in going through the amplifier. These include the infamous 'Hot-Ground' offset drift, the use of constant current diodes or zeners on the driver boards, and the fact that some of the boards, especially the drivers, have piggy-back components strapped across others from the factory.

First off here are some of the dismantling pictures. In general the condition was not too bad, with really only the Power Supply board suffering from the glue problem. The amplifier is a European spec machine, and was working well without faults before I started, so no trouble-shooting required.

IMG_0274.jpg

IMG_0283.jpg

IMG_0281.jpg

IMG_0287.jpg

IMG_0295.jpg
 
Register to hide this ad
Apart from the use of metal film resistors mentioned above one of the things I really like about the AU-D11 II is that as a bridge amplifier the supply voltages for each half are quite low which allows use of the Panasonic FM range of electrolytics, which are only available up to 50 V dc working. The FM range are the best specified of the Panasonics by some margin, so it was good to be able to use them. Elsewhere the replacement components are the typical Nichicon ES bi-polars, and Wima polypropylenes. Due to my limited stock of values there were not many opportunities to use the Wima polycarbonates. I also used some Nichicon FG series in selected places (thanks Hyperion!).
 
The AU-D11 II is a very underestimated amplifier because the build quality is rather 'flimsy' - (although what this really means is 'not built like a battleship' as some Sansui's are :D ), and the number of trimmers is a bit offputting for some, :) but once you embrace this, you are onto a real winner. ;)

I'll definitely be following along with this one. :thumbsup:
 
One issue I had was that all of my searches for replacement bulbs for the input selector indication turned up dead links (links have since turned up!). So I tracked down some 6V 200mA units, physically very similar, and installed a dropper resistor, value selected by powering a bulb from a bench PSU and comparing its brightness to one of the good originals still in the amplifier. The dropper was then installed on the indicator board, and the coloured caps from the originals were transferred.

IMG_0299.jpg


IMG_0300.jpg
 
That looks like a good place to substitute LEDs. Nice write up. Please post the serial number for the database. Click on the link in my signature block.

- Pete
 
Hello Pete,
The issue with the AU-D11 II is that the indicators are driven from 8Vrms AC. I did think about using a common dropper, and a common reverse diode to half-wave rectify the supply, but decided that this way was simpler.

I'll post the serial number for the database.
 
The Equalizer and Control amp op-amps were replaced with Brown Dog adaptors. Initially I tried the LME49860s, but they produced an underlying oscillation around 15MHz at roughly 150 mV rms, so they were removed. I'm guessing that the compensation was not sufficient when operated with a preceding gain stage (the differential FET input). At this point I installed the OPA2604s which I already had lined up, and there were no further troubles with oscillation
 
Looking at the issues noted in the opening post the first one addressed was the driver board front end configuration. The purpose of the zener or constant current diode is to establish the operating current in the differential input FET pair. Either set-up will work since it is only a matter of setting a stable reference voltage for the base of Q1. My amplifier has the constant current diode configuration as shown in the SM. I assume that the change to the zener for later production was made for cost reasons, since a cc diode costs about 6 -8 times the price of a zener. However, in my view a zener version should have an electrolytic of at least 47uF across the zener - look at all the other places where zeners are used in AU amplifier input stage biasing and that is what you will find - including in the AU-D11 II itself. I am guessing that the design intent was that the cc diode version would give a lower noise level, but in that case why wouldn't you do it in the equalizer and control amps as well - more benefit to be had there.
 
I found piggy-back components on the driver boards, specifically the lower resistor R33 in the 'Hot' driver output chain which works in conjunction with VR2 to set the 'Hot-Ground' offset. While the schematic calls for 120R, the resistor installed on board assembly (ie on the component side of the board) was 1k. This was paralleled by a 120R on the track side of the board, giving a net R33 value of 107R. Since I was replacing resistors with metal films anyway I place a correct value 120R component in R33.

Another piggy-back was an additional film cap in parallel with the on-board one in the equalizer.

IMG_0296.jpg



This along with the output resistor from the op-amp gave a filter cut-off frequency of ~ 48kHz. Since I wanted to use a Wima polypropylene here, but didn't have an exact value to substitute I used a single 4.7nF with no piggy-back, and changed the output resistor from 560R to 680R, giving a similar cut-off frequency.
 
I did one of these years ago, I also have two sitting at home at the moment. One thing I wondered was where the "daughter boards" were soldered to the "motherboard". I thought about drilling tiny holes in the connections, and putting pieces of wire in there to help with the conductivity and also mechanical location. Of course, it would be more of a problem next time the amp needed recapping in another 30 years, but I will be approaching 80 years old by then, and probably not interested.

Watching this thread with interest, keep up the good work, and keep the pictures coming.

Lee.
 
For those who would like more pictures, this is one of the few worth-while ones that I haven't posted yet:

IMG_0293.jpg


Most of the others are just my reminders to me of which wire goes where!

The observant will note that I didn't change the resistors in the tone control circuits. I never use them, and just wanted to press on and get the job finished!
 
Last edited:
Finally I looked at the 'Hot-Ground' offset issue overall. One thing that struck me is that since the transformer secondary is not centre-tapped there is no true low-impedance ground as a reference. The 'ground' is established by two 6.8K resistors across the large smoothing caps in the PSU. So the effective ground path is through a 3.4K resistance (two 6.8Ks in parallel). I decided that this could be improved on so I replaced the 6.8K 1W devices with 1k 2W metal film resistors. The standing dissipation is a bit higher than I would have liked at ~ 1.3W but so far no ill effects have been noted. With a 500R resistance to ground the 'Hot-Ground' offset can be set to 20 - 30 mV , and it apparently drifts up to around 250 mV max, which I think is a bit better than normal.
 
Here is the completed unit back in my equipment rack, complete with brand new front panel courtesy of Nico at Sansui Parts Shop, and a re-painted top cover. At some point I will probably redo the veneer on the side panels but I'm too busy listening to it at the moment!

IMG_0323.jpg



So the final question - how does it sound after all this work? The unrestored AU-D11 II was still a good-sounding amplifier, somewhat better than the restored AU-D9. Now it is by far the cleanest sounding amplifier I have heard (list includes AU-719, AU-919, Audiolab 8000, Onix 20A, Allnic tube-something-or-other). The clarity and timing of music is superb, vocals carry every nuance of the performer's intent, and its ability to handle complex passages with frequency and dynamic extremes (often simultaneously) is unsurpassed by anything I have listened to. And as others have noted this amp really shines at low volume levels. So overall I am well satisfied, and in my book it was definitely worth the effort.
 
phonomac said:
The clarity and timing of music is superb, vocals carry every nuance of the performer's intent, and its ability to handle complex passages with frequency and dynamic extremes (often simultaneously) is unsurpassed by anything I have listened to.
Click to expand...
+1
I couldn't have put it better - an excellent description IMO.
 
Last edited:
I think the AU-D11II is a real star, and it's great to see another one refurbished with components that allow Sansui's AU-D11II circuitry to sound its best. My MF-equipped unit is my best sounding amp, I love it. That said, when I did it I was unaware of the potential benefits of WIMA stacked film caps and that will be my next mod.
 
The WIMA stacked film polypropylene caps are excellent, and well worth using as replacements for polyester (Mylar) caps. If you can find the values you need the WIMA polycarbonates are even better - I used several in the MC and MM stages of both the AU-919 and AU-D9 refurbishments. With polycarbonate film discontinued by Bayer some time ago these caps are getting harder to find. It is worth emphasising for other readers that the first choice should be the stacked film and foil caps, not the metallized film variants.
 
You must log in or register to reply here.
Back
Top Bottom