• 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

Installing a new 2772 into an X1

Ronito6

Ronito6

Super Member
This posting is for the benefit of several AKers who recently acquired my F-2772 for their respective AU-X1's. I thought a single reference point would be the easiest way to answer questions and to share ideas.

Instructions for installing the new 2772 board into your X1.

PARTS LIST
Needed to complete the board.

6 x Omron G5V-2-DC24 signal relays.
1 x 33Ω ¼ watt resistor.
1 x 5.6Ω ¼ watt resistor.
3 x 1N4003 rectifier diode. (Replacing 10D1)
4 x 470µF 35V electrolytic capacitors. (I used 63V here)
2 x 0.047µF 100V film capacitors.
1 x 0.27µF 100V film capacitor.

Also, collect all wiring and hardware from your old broad too. You will use all of it on your new board.

(Parts list courtesy of Hyperion. http://audiokarma.org/forums/index.php?threads/replacing-au-x1-f-2772-phono-mother-board.833557/)

PREPARING THE NEW BOARD
First of all make sure that the corner mounting holes have been turned into ovals by either filing them down, or using a drill bit to grind them down, or by whatever means is easiest for you.

Doing this will guarantee that the board will lay flat against the chassis.
upload_2018-9-4_12-51-45.png
The distance required to file down can be seen by the white silk screen that marks where the holes are. Removing material to the point where the silkscreen indicates should be enough to allow the board to properly seat.



NOTATIONS
The board has notations for all wiring.
So where the wire originates and its destination point on the new board should be easily sussed.
upload_2018-9-4_12-53-30.png
The board also has notations for all jumper wires.

NOTATIONS: CLEAR VS WHITE
The following should not be an issue for anyone swapping out an old version 2

Version 1 swappers should note:
The stereo wires are encased in various colors of plastic, however the wires inside the casing will be either black and white, or black and clear.

Version 1 of the 2772 uses stereo jump wires that are black and white.

Version 2 of the 2772 board uses stereo jump wires that are black and clear.

MY NEW BOARD notations refers to these stereo wires only has black and clear.
So if you are installing my board to replace a version 1, please consider the word clear to mean white.

There are many of these…but here is a sample.
upload_2018-9-4_13-28-14.png




INSTALLATION ISSUES REPLACING A VERSION ONE
First remove all of the hardware and wires from the original board. You will need these parts to install into your new board.

NOTE:
The version 1 board uses a lot of tipped wires. I’ve designed my board with through holes large enough to receive these tips. So as long as you clean off the old solder from the wire tips, you should have no problem using the previous wiring from your version 1 on the new board.


NOTE: THE BIGGEST ISSUE FOR VERSION 1 REPLACERS - THREE PRONG HEADERS
(Edit: Sep., 2018 - BIGDRIVE has found a suitable 3 prong header and plug. Examples and links for them are at the bottom of this posting. Also refer to posting #32.)

Sansui's original flawed 2772 design is fitted with two 6 pin headers.

Sansui's version 2 (the version upon which my board is based) uses just one 6 pin header, and two 3 pin headers. So one of the the 6 pin headers on the version 1 was redesigned into a pair of 3 pins headers on the version 2.

A picture to make the issue clear.upload_2018-9-4_14-27-45.png
If you are lucky you have a Sansui parts pile where you can source the headers and the plugs that fit.

There is always directly soldering bare wire to pads, or addind solder lugs to which the bare wire would be soldered. Going with the bare wire/lug route is actually a superior solution in many ways. What you lose in modular design you gain in circuit integrity.

BUT if you wished to keep the modular design aspect, but wanted to use a modern solution there may be some other acceptable options.

For example one could wire in a pair of connector Molex connectors. If I did this I might consider running the wiring underneath the 2772. This would offer a way to hide the Molex plugs while placing out of the way all that wiring which normally crowds the phono cards .


INSTALLATION ISSUES REPLACING A VERSION TWO

This process for swapping out a version 2 ismore straightforward than replacing a version 1.

Original version 2s have all wiring and hardware necessary to make this board work.
(Most specifically the three prong headers and plugs.)


SOME LARGE THROUGH HOLES
The only problem you may encounter will be the size of some through holes for the jump wires.

On the original version 2, Sansui moved away from using the tipped wiring on the jump wires which they use in the version 1. On version 2 Sansui opted to go with bare wire jumpers.


So some through holes on my board may seem unusually large, but that was a compromise I intentionally incorporated to make things easier for those who opt to use the tipped jump wires from a version 1.

Of course ... version 2 folks can...
  • use a larger wire gauge wire for the jump wires.
  • or add tips to the jump wires
  • or fold over the ends of the bare wiring culled from their original version 2. (the original bare wiring should then fill the through holes nicely.)
OTHER THAN ALL THAT…

INSTALLATION

1. Before installing into the chassis, populate your board from the center out, going from the smallest components, jumper wires, and headers, up to the largest components.

2. Install the populated board into the chassis.

3. Finalize wiring the new board into the amplifier.

4. Reinstall your Phono and EQ cards

5. Check for shorts or other flaws or mistakes as you normally might.

6. Check all voltages and adjust them to spec if needed.

7. Enjoy your more stable X1, and the new small signal relays from this century.

  • PLEASE SEE POSTING #16 BELOW IN HIS THREAD FOR PICTURES OF AND OTHER DETAILS REGARDING THE KUEHNE MODS.
  • PLEASE SEE POSTING #32 IN THIS THREAD FOR DETAILS REGARDING CONTEMPORARY 3 PRONG CONNECTOR SOLUTIONS DISCOVERED BY BIGDRIVE.
https://www.digikey.ca/product-detail/en/3-641208-4/A31678-ND/769821
index.php


https://www.digikey.ca/products/en?keywords=3-641217-4-ND
index.php


 

Attachments

  • upload_2018-9-4_13-25-23.png
    upload_2018-9-4_13-25-23.png
    382.9 KB · Views: 21
  • upload_2018-9-4_13-39-31.png
    upload_2018-9-4_13-39-31.png
    592.5 KB · Views: 18
  • upload_2018-9-4_14-1-15.png
    upload_2018-9-4_14-1-15.png
    703.2 KB · Views: 17
  • upload_2018-9-4_14-9-22.png
    upload_2018-9-4_14-9-22.png
    714.4 KB · Views: 17
Last edited:
Register to hide this ad
Hi Ron,

Thanks for getting the boards made and distributed to us as well as following through with this posting. Paul Hovenga at Many Moons Audio has my X1 and will appreciate the technical assistance, I’m too uneducated to make use of it. Will your notes on the Keunhe stabilty modification appear in this post, or are they listed elsewhere and I just haven’t been able to find it? Paul would appreciate this information in English if I can get it to him. I mailed him the new phono board. Thanks to all.

Marc
 
Here once again are the Kuehne modifications.

These modifications were created by Mr. Kuehne to settle what he discovered were the sources of the famous AU-X1 destructive oscillation and DC drift on the EQ boards.

The mods were first suggested and applied to an X1 here on Mr. Kuehe's website. (It is in German, but Google translate is great for German.) http://ftbw.de/xp/amplifier-xp/sansui-au-x1-01.html

AK member Harolda was the first to discover, and report on, and discuss what Mr. Kuehne found on posting #147 on this thread http://audiokarma.org/forums/index.php?threads/the-mighty-au-x1.672322/page-8
Harolda wrote:
I read an article of a german tech who seems to have solved the instability issue of the amp by improving the grounding of the flat amp, in a same way that was done by Sansui on the Phono motherboard.
I you look at the upgrades performed to the first generation phono motherboards, one can recognize that a grounding wire from the central ground connection of the chassis has been coupled to the input grounds of the left and right channel of the phono motherboard. Additionaly both input ground planes have been coupled using an additional capacitor. I will post a picture of this. I also noticed that in the pictures i saw in an older post from Ronito.
I think the second generation motherboard is not that different from the first one, only that the modifications have been included into a new PCB. I will get back to that later

Now this grounding trick has also been applied by the german tech to the flat amp to avoid oscillations in a channel and subsequent resonance in the other channel (back and forth) that in the ends fries the power amp. So he placed a grounding cable to the (input?) ground planes of the flat amp connected via a 10nF MKT and coupled (HF) both ground planes using a third capacitor. This did the trick for him to solve all stability issues

He did some other things as well:
- remove the ceramic capacitor on the MM board that connects again to the heat sink and to the amp enclosure, if the lid is placed of the MM and MC boards
- He placed a 2 component circuit on the MM boards using to avoid the DC drift on the MM board he was measuring. This circuit should produce a voltage gain of 1.

Mr. Kuehne somehow got word and joined AK to add his 2 cents. http://audiokarma.org/forums/index.php?threads/the-mighty-au-x1.672322/page-8


MartinKuehneNew Member
"The DC drift removing trick on the MM board is: cut off the inverting input gate of the double FET and AC-reconnect it with a (small) 330nF foil capacitor. Then DC-connect the gate to the output (of course BEFORE the coupling capacitor) by a 1MOhms resistor. With a FET input that works great, hence there is no balance problem from input currents. The lower bandwidth limitation now is 0.5Hz at this point, DC drift is history.

Last edited: Feb 22, 2016"

Finally, Using Mr. Kuehne's methods ...
Harolda demonstrated the suggested flat amp ground modification http://audiokarma.org/forums/index....-output-transistor-replacements.684619/page-8
and
Hyperion demonstrated the MM (EQ) board DC drift modification here http://audiokarma.org/forums/index.php?threads/the-mighty-au-x1.672322/page-10#post-11068313

Below is a visual summation of their work.

The Flat amp ground modification.

index.php

index.php

index.php

index.php

index.php


MM (EQ) board DC drift elimination modification.
index.php

index.php


index.php


And another - to give a better view of the connections.
index.php
 
FINAL NOTE ON AU-X1 MODS:

The instability and oscillation problems for which the Sansui AU-X1 was famous for had many suggested solutions.

Some were amazingly extreme.

So for fuller appreciation of Kuehne solutions,
I present the solution suggested by (in sincerity) a truly great tech, AK member BeatleFred.
His suggestions were likely very good, but too drastic in my view.

BUT back when he wrote this on July 23, 2003, there was a lot less information about the AU-X1 from other experienced techs.

So to me, it is amazing the extent he was willing to go in order to make his X1 stable.

BeatleFred wrote posting #8 on http://audiokarma.org/forums/index.php?threads/rugged-sansui-amps.8123/
"Sansui AU-X1 Modification.

The Sansui AU-X1 is designed around a circuit that became known as the Diamond Drive DC circuit. It had already been applied in the AU919 and AU719 integrated amplifiers but Sansui took the design one step further in the AU-X1. This amp used even faster transistors and achieved a slew rate of no less than 260 V/uSec. The power bandwidth is hardly limited and reaches over 100 MHz. It is therefore an amplifier that is very vulnerable to almost any kind of HF influence, be it speaker capacitance or inductance or even speaker cable capacitance or inductance. I found that one other factor of influence is pcb connector corrosion, but this factor becomes only apparent after so many years. About this I will comment later. First, a description of how to modify the power amp section to get rid of its instability once and for all.

For the principle of my modified version of the AU-X1 power amp I looked into the AU5900 and basically found what I was looking for. If you have the diagram of the AU5900 available you will find that most of it can easily be applied on the X1 power amp board. The current source powered differential FET input amp can stay as it is, but it is modified to directly feed the pre-driver stage TR10 and TR11. To do this you have to remove the voltage source of ZD01, C02 and R03. This last resistor is replaced by a bridge wire. Replace resistors R04 and R05 by 330 Ohms 0.25W and replace diodes D01 and D02 by Zener diodes of 47V 3W. Notice to reverse these diodes! C04 to C07 can remain in place. I have replaced C05 and C06 by 47uF 63 V. The current source around TR01 can remain unchanged, be it that I replaced TR01 by MJE340 and R45 by 15 KOhms to increase stability. On the input I have replaced R01 by 47 KOhms and C01 by 47 pF polystyrene. The entire Diamond Drive circuit, consisting of TR02 to TR09 and all its collector and emitter resistors R16 to R28 is removed. TR11 is to become a current source and to do this, R28 is replaced by two general purpose diodes in series, like the 1N914 or 1N4007. R34 is replaced by 100 Ohms and R15 is replaced by 15 KOhms. You can either solder this resistor directly into the collector hole of TR09 or make a wire bridge from base to collector. Now a bridge wire has to be made between the base of TR04 and the base of TR10 or the collector of TR03. Remove C09 and C10 as well as C15 and C16. C13 and C14 can stay, but if you want to further reduce ringing and slew rate, remove them. The feedback circuit consisting of R38 and R12 is modified as well. R38 is replaced for 27 KOhms paralleled with a 47 pF polystyrene capacitor. R12 is replaced for 470 Ohms. Originally it is DC-coupled to ground, but for increased DC stability I have chosen to put an electrolytic capacitor of 100 uF 63 V in series with R12. The most convenient way to do this is to remove an existing wire on the board and put the capacitor in. TR11 and TR13 are replaced by MJE340 and TR10 and TR14 are replaced by MJE350. I have also chosen not to use the separate stabilized power supply for the driver stages, but simply use only the power supply straight from the main capacitors. To do this, connect pcb points 06 to 14 and 08 to 10. The ground wire of the connectors, the center ones, should in that case be cut, stripped, and soldered to ground, otherwise the flat amp power supply is not grounded. (Don’t forget to do this, otherwise you’ll have a bit of a problem!) The + and – of the connectors can stay disconnected. If you do want to continue using the stabilised power supply it might be wise to cut the connector as well and solder the wires straight into the pcb. This is of course not strictly necessary, but it will increase stability. For that reason I have also bridged D03 and D04 and increased R41 and R42 to 56 Ohms. This will considerably reduce the heat dissipation of TR15 and TR16. I have replaced TR15 for 2SD1138 and TR16 for 2SB861. These driver transistors will still allow you to use three output pairs, but I have removed one pair. I have also replaced the power transistors by MJ15022 (TR701, TR705) and MJ15023 (TR703, TR707). At last, make sure you replace R39 and R40. They can remain of the same value, but I have found that that particular type of resistor is unstable over time. The ones I came across read values of over 1 KOhm.

To be on the safe side I used a variable auto transformer for the first power up of the amp. Short circuit the input (do not connect the preamp section of the unit!!) and apply no load to the output. When DC-offset can be adjusted to a few millivolts and bias current does not run away and can be adjusted to normal values, apply a load of 8 Ohms to the output and an adjustable signal generator to both inputs. Connect a two-channel oscilloscope to the output and adjust the signal generator for a signal of approximately 50 mV. Look on the oscilloscope for any abnormalities or oscillation. Turn down the level of the signal generator and switch it over to square wave. Adjust the frequency to 10 KHz and gently turn up the signal level. On the oscilloscope the signal should have little or no ringing and a nice symmetrical rise and fall curve of an acceptable steepness.

Now that you have the power amp modules in working order again it is time to take a closer look at the preamp section. The four phono preamp boards are situated on a mother board with some relays to switch from mm to mc input and each have their own stabilized power supply. First, check all four power supplies for the correct voltage. They should read + and – 10 VDC and + and – 46 VDC. If not, remove the power supply board and repair it. Look for any bulging electrolytic capacitors. Best is to replace them all, except maybe for the large electrolytics on either side of the power amp modules, but the number of small caps runs into the 60 or 70 or so. Remove one of the connectors from the mother board and look at the pins on the mother board. Usually these pins are severely corroded. I have chosen to cut the connectors from the wires and remove the pins from the mother board and solder the wires straight into the mother board because I found that the corroded contacts cause DC-jumps on the outputs of the phono preamps and, even worse, cause oscillation. This counts for all connectors of this type everywhere in the amp. Removing all these connectors and soldering all wiring straight into the the pcb’s should keep you busy for a few hours, but in the end you will find that nearly all oscillation coming from the preamp section is removed.

At last, you will have an AU-X1 in good working order once again; technically stable and reliable for years again, and most of all, a pleasure to listen to."
 
Last edited:
mjs1 said:
Hi Ron,

Thanks for getting the boards made and distributed to us as well as following through with this posting. Paul Hovenga at Many Moons Audio has my X1 and will appreciate the technical assistance, I’m too uneducated to make use of it. Will your notes on the Keunhe stabilty modification appear in this post, or are they listed elsewhere and I just haven’t been able to find it? Paul would appreciate this information in English if I can get it to him. I mailed him the new phono board. Thanks to all.

Marc
Click to expand...

The cavalry have arrived.
 
Last edited:
EDIT: SEPTEMBER 30th 2018
SEE ENTRY #32 IN THIS THREAD FOR INFORMATION REGARDING A CONTEMPORARY CONNECTOR FOUND BY BIGDRIVE THAT FITS THE 3 PRONG CONNECTOR POINTS.

THIS ENTRY WILL REMAIN AS A HISTORICAL POINT.

Another note: EXAMPLE THREE PRONG CONNECTOR SOLUTION if TRANSITIONING FROM A VERSION ONE F-2772

Here is my favorite THREE PRONG connector to use if I sought to transition from a version one 2772 to my version two (if I wished to keep the ability to plug and unplug connectors to my new 2772).

This connector is very high quality and would serve well as a replacement for the three prong original connectors.

upload_2018-9-22_23-38-50.png
upload_2018-9-22_23-41-3.png
upload_2018-9-22_23-42-13.png
index.php


index.php
 

Attachments

  • con together.jpg
    con together.jpg
    37.7 KB · Views: 161
Last edited:
You must log in or register to reply here.
Back
Top Bottom