Sansui AU-X711 upgrade

pete_mac

Super Member
How does one improve on the Sansui Alpha 607 (of which the X-711 is an 'international' derivative)?

You tweak the circuit to reflect the refinements found in the later 907s, copying Sansui's circuit enhancements and using Elna Silmic caps just about everywhere!!
default_smile.png


This approach has been adopted by a few European alpha fans doe a while now, after studying the schematics, identifying the improvements and applying them to the lesser models. In subjective listening comparisons they believe the heavily tweaked 607s can approach the sound of the later 907s and the B-2105 power amp (at sane listening levels, due to the power difference between the low and high end models).

Forum member csubzi (who has been very generous in responding to my questions, courtesy of Google Translator!) has provided a bucket-load of advice and recommendations. I haven't been able to follow his recipe 100% due to the difficulty in obtaining parts like Black Gate capacitors and the rare U-Con polypropylene caps, but if these parts turn up in future, I will apply them later.

Below is a big stash of Elna Silmic IIs, resistor upgrades (Riken, Amtrans, Takman and PRP), Polystyrene and polypropylene shunt capacitors, and some Nichicon Muse bipolars. New Bourns multi-turn trimpots will also be fitted.

I'll soon be indulging on some work on my own personal amps for the first time in many years. I can't wait to hear the results!! I have our AU-alpha 607MRX on-hand (my current daily-driver amp) to compare against. The 607MRX currently sounds better to my ears - greater clarity, resolution and transparency. Let's see if a reworked 711 can surpass it!

Alpha upgrade.jpg
 
Can you make a parts list for this upgrade? I have a AU-X711 that might be good for this project.
 
We are finally underway!

I decided to recap the regulated PSU and one channel first, so that I could hook up my meters to the hot and cold amps at the same time when setting bias and DC offset.

I mentioned earlier that czubsi helped me enormously with his formula for enhancing the 711/911 (thanks again Paul - greatly appreciated!!) Unfortunately my work isn't exactly like czubsi's, mainly as I can't find the rare Elna Silmic Alpha Super Gold caps, the Black Gates, the polypropylene U-Con caps, and the vast range of Riken RMG and Amtrans AMRG resistors really starts to add up in cost. This is why I went for the lesser Amtrans AMRT in many locations, along with Takman REY. I note that Amtrans use these exact resistors on their high end stepped-attenuators, so they can't be too bad ;)

I ended up using Panasonic polypropylene caps in place of the U-Cons too. Again, sourcing these for realistic prices is just too difficult.

In short, the upgrade makes use of empty spaces on the PCBs for things like extra capacitors, bypass caps etc which are found on the higher-end models but not on the 607s. There's nothing particularly revolutionary about the tweaks - they follow Sansui's own recipes.

All of the Nippon Chemicon caps have been replaced with Elna Silmic IIs, (apart from two in the power supply due to size issues). I've never done this before, as I always use long life, high temp, low ESR caps in the PSU. However, sticking true to Sansui's formula, I figured I'd give it a shot. Polypropylene and polystyrene caps are fitted where possible. The Silmics are much larger than the originals!





Due to a lack of space between the amp board assy and the chassis on one channel, two of the 470uf 50V Silmics are tilted over. In czubsi's mods, these are remotely mounted near the filter capacitors via flying leads. However, I prefer to have them installed on the PCB to reduce inductance due to the long wiring. This means that I've used the spare bypass capacitor location to accommodate a pair of 100uf 50V caps. As a result, I've fitted the polypropylene bypass caps on the underside of the PCB as there is ample room, along with all of the 4700pf Siemens polystyrene bypass caps.





Simple tweaks such as removing wire jumpers and fitting caps in a back-to-back pseudo-bipolar configuration in certain locations (as per the upper end models) is easily done, with polystyrene bypass caps installed.






A paralleled polystyrene and polypropylene bypass caps is also fitted across the wiring from the filter caps to the amp PCBs.

I was only able to replace the DC offset trimpots at this stage, as the bias trimpots have a reversed pin configuration vs the DC offset trimpots, thus the usual Bourns-style trimpots would not fit. Thankfully I was able to set bias easily with the originals. The Bourns-style multi-turns certainly come in handy with the DC offset on the Alpha X-balanced Sansuis, that's for sure!!

The PCBs have quite a lot of flux residue on the top side. I'm progressively cleaning it off as I go, but at the end of the day, it's only cosmetic. The zener diodes were placed very close to the PCB and caused localised heating and darkening of the flux. I've since replaced them with new units mounted higher off the PCB. You can see the discolouration under the original zener here:



Next step is the other channel (currently partly stripped), the protection circuit, and the EQ/phono boards.




I'm very much looking to forward to the first listening test, and further pics will come!
 
Last edited:
Both amp boards are now finished!



You can see the polystyrene and polypropylene caps on the rear of the amp board PCB here:



More Amtrans and Riken resistors:





I'll definitely look into alternative Bourns trimpots with the correct leg configuration for the bias pots, or will modify some to suit.

The phono board has been fully reworked, with Silmics and a lone WIMA cap, and the three key signal path resistors per channel replaced with Takman REY. Those 1000uf 10V caps are FAT and end up rubbing shoulders.







I took no photos of the EQ and protection circuits, but they are also fully Silmic'd/recapped/tweaked/bypassed/unicorn-dusted as appropriate.

Taking a leaf of out Sansui's book re: dealing with resonances, I've applied foil-backed butyl deadener on the larger panels throughout the amp. This is a tiny snapshot of what I did. The amp is now noticeably heavier and the chassis sounds less 'clang' and more 'thud' when tapped.






She's all ready to clock up some mileage before serious listening. However, even straight off the bench after some DC offset and bias adjustments, she sounds absolutely fantastic! Noticeably better than before. The question is... should the 607MRX be worried?? Let's wait and find out!





 
Very nice Pete - great work.

Do my eyes deceive me or did you only 'lay down' the large filter capacitors on one of the driver assemblies? (looks like there might be more clearance on one side?).

Can't wait for listening impressions...
 
Very nice Pete - great work.

Do my eyes deceive me or did you only 'lay down' the large filter capacitors on one of the driver assemblies? (looks like there might be more clearance on one side?).

Can't wait for listening impressions...

You are spot-on - one channel has less clearance to the chassis, so the 470uf caps need to be laid down on one PCB.

I've made full-use of the available PCB space, fitting 220uf 50v and 470uf 50v caps in the original electrolytic cap positions, and 100uf 50v caps in the factory bypass cap location. The polyprop bypass caps is then mounted on the rear of the PCB instead. I preferred to mount the caps directly on the board, rather than on flying leads off the board.

There are some little aspects that frustrate me (OCD) like the red PRP resistors being mounted askew. They look better in the flesh, but they have been twisted so as to provide thermal clearance to the capacitors.

Same goes for some capacitors - physical constraints mean that they aren't absolutely perfect. Still, it's not too bad.

I'm very much looking forward to a big listening comparison soon - restored AU-X701, original 607MRX, and restored/modded AU-X711.

I do have some left-over Silmics, as well as a few resistors, polypropylene and polystyrene caps... maybe the 607MRX needs a freshen-up too... ;)
 
Beautiful work on a beautiful amp. Can you post the serial number for the database?

- Pete
 
Well, early days, but this amp sounds just superb. It already sounded good in the first place, but in original unrestored condition it was a bit smoother than the AU-X701 and less clear/resolving than the 607MRX. The recap and mods have well and truly been worth the effort and expense.

I should note that I still have some 100R resistors to add to the amp board, and once these arrive in the next week or so, I'll put a few more hours on the amp, reset the bias and offset, and let her rip in an old-fashioned shoot-out.

Might have to borrow Skippy124's ear's methinks! :)

Beautiful work on a beautiful amp. Can you post the serial number for the database?

- Pete

Will do Pete - I have a few bits of gear to add. :)
 
Last edited:
:trebon: Thanks Pete for sharing your good work and photos. I have a 707 DR in a box waiting to be repaired (I still do not have experience to do it, but I'm learning in AK with post like yours). I understand that 711 is similar, is it correct? (Impossible to find the 707DR service manual).
 
Back
Top Bottom