1. Time for some upgrades in server hardware and software to enhance security and take AK to the next level. Please contribute what you can to sales@audiokarma.org at PayPal.com - Thanks from the AK Team
    Dismiss Notice

Anaemic AU-111 Diary

Discussion in 'Exclusively Sansui' started by _mano, Jul 4, 2015.

  1. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    sort of Diary read.... to put you to sleep (in 2 parts)

    Serial No: 96030520 - Mar 1966
    Very Early AU-111 model with Staggered Preamp Valves.


    Highcut28, posted on AK on how to resolve the Anaemic sound being produced on his AU-111 early this year. Following a few PM's the unit landed in UK at the end of May.

    The AU-111 is one of the really early units, with 2-tone controls and staggered pre-amp valve section. The Amp was shipped with all the Tubes/Valves removed and safely wrapped inside the Amp.

    Issues:
    1) Tone controls non-effective
    2) Sound reproduction thin and anaemic
    3) Occasional drop out from the Aux input
    4) Balance not working
    5) previous Attempts made to repair and given up.
    6) Output Valves accidentally killed by incorrect speaker connection.


    The removed valves on inspection had a pair of JJ 6L6GC and a pair of metal canned RCA-6L6 and the usual newer preamp and driver stage valves. All the control knobs were also removed for transit.

    Mid-May

    With the underside removed, evidence of previous work and signal cables left cut and tucked away as well as volume control re-wired, balance control not connected, newish RS branded coupling caps and insertion of Ceramic disc caps, some expensive looking yellow/Copper capacitors in the signal path. etc.. slightly molested.

    Bit of history of the Amp suggests one of the previous owner did not like the facility of the tone control being active in the Amp and had it re-wired to bypass that section. I remember this phase of HiFi during the late 70's. (I am sure the merits of this has been discussed to death, revived and more.)

    Back to the Amp, nice little challenge. Few hours tinkering during each weekends.

    Usual Visual checks dd not show anything that may cause fire or be electrocuted. Resistance checks on the chassis isolation to mains and transformer checked all good and set to 240V.
    The main 15W on the B+ was loose and flapping as the mounting insulating washers have broken up. May be the Transit from Malaysia to UK was too much for its age.

    End Of May

    1) So powered up with a Variac slowly with none of the valves in and all the DC potentials looked good and stable. (pheww)
    2) Inserted the pre-amp valves in, powered up and left it for few mins, still good. Inserted a bit of audio into the Tuner input, audio coming out on the pre-amp output, very low and tinny. Volume control works, nothing else.
    3) Powered off, inserted the JJ and a couple of spare 6L6GC in, slowly powered up. Half way through, the 1.5A fuse on the Variac failed. Removed the fuse from the Amp, it was not the correct 2A fuse but a re-soldered lump wire for a fuse, which would never have popped. (Forgot to check this early on, early stage dementia setting in, need to start a todo list, me thinks). Inserted a correct 2Amp fuse.
    3) Tried again, this time with an Amp meter in place, the excess current being drawn is only apparent when the 6L6GC start to come alive. So, wound down the bias to the absolute minimum (cold-bias), better, however one of the JJ starts to red-plate.
    4) Removed the JJ's tried, all OK. Checked pin5 with the power on, sitting just over 52V. May be the coupling capacitor breaking down.. Checked the Capacitor, seems ok.
    5) Checked the JJ's, not matched at all, no shorts internally,lets try it again. No red-plating, all stable. Very strange... Something is not right.
    6) Replace the U link between the pre-amp and Amp section, sound is there but as HighCut28 stated not what you expected or want to listen to.
    7) Inserted the Audio signal straight into the Power-Amp section, lot, lot better. Atleast the output stage is decent and good,
    So its repairable..... and passed the good news to HighCut28 and started on the bits to replace for the suspect main coupling caps as well as all the paper-oil caps in the pre-amp and tone section.

    So a list of Orangedrops for the signal section, set of Solen's for the Coupling and a set of JJ's for the Output stage. Some Ceramic Washers for the Power Resistors.

    1st Wknd June:
    Lets start on the Power Amp, section.
    1) Removed the existing RS badged Coupling caps. Instead of the values being 0.33uF, they were 0.22uF. Adds to the anaemic sound issue. Replaced them with the set of Solens.
    2) Repowered up, to check, red-plating again on the same position. Power-down, wait for a while and started to poke around the socket where the red-plating occurred. The 1K resistor feeding pin 5, 0.25W carbon resistor of that time, where the resistor barrel is held by compressed caps with leads to solder. The compressed cap was just being held by the tension of the soldering on the lead. Replaced all the resistors in this area of the bias.
    3) while I was plumbing the Solen's I spotted a real serious issue. As this Amp has gone through a few repair and modification cycles, solder has been used excessively, while the amp was upside down and as a result, the solder being applied has flowed down from the 9pin valve socket tags and bridging to the chassis. Easy to do, difficult to spot. Sucked it up, checked around all Valve bases, specially where new solder has been applied.Further overflow's were found, re-heated, sucked up and fresh solder applied.
    4) Powered back up, all stable, sound good just being driven from the pre-amp/amp split sockets.

    2nd Wknd June:
    1) the amp is a 2-pot bias scheme. With authorisation from HighCut28, changed to 4 pot version, however used some Vishay panel-mount trimmers and test-points, these are now presented so, it can be checked and adjustable from the top of the Amp. No new holes, used the existing venting holes around the 6L6GC area. Should not affect the ventilation too much. Yes used a bit of vero board, easily removable for restoring the originality, however, its a well worth useful tweek/deformation.

    3rd Wknd June:
    1) OrangeDrops are now being exchanged for the oil-paper caps. The OrangeDrops, just about able to squeeze & Tuck it into position. Really fiddly job, Cutout old, remove and clean old lead, cleanup/suckup excess solder on tags. Wholesale change of caps as HightCut28 wanted this to be the last time and wanted the amp to last.
    2) This is a time consuming activity, each capacitor replacement can easily take up to 15 to 20 mins of slaving over the Amp.
    3) For the Coupling caps, I used fibre based high temperature sleeving. Great, only problem with these are they start to fray at the cut ends and will be a potential issue in tight areas. Looked around, and found a suitable rubber/silicone based sleeving which melts at +180C. Had to wait a few more days for this to arrive.
    4) Finished replacing all the orangedrop
    5) Re-wired the tone section
    6) Made up a replacement Balance control using 23step attenuator. The Original was replaced with a standard dual pot. Which is not the same as the original pot in-terms of how the resistive track is made up. The originals are 500K pots, however, the 500K was only half way up the potentiometer track and then remaining was Zero Ohm and opposite layout for the other channel. Very hard to find these, so 23 step attenuator re-worked.
    7) Re-wired the Volume, Balance and Tape Monitor input.

    Tested the outcome, blast, still anaemic, but Tone control's now work, no more spurious discharge noises, tone control pots quite.

    8) OK, get the scope out, 200Hz Signal, signal through C17, fine, via C19 attenuated very badly. This is one of the expensive looking Yellow bodied capacitor with a copper foil around it. Whip it out, remove the copper foil, .001uF instead of 0.5uF. Stuff in a temporary .33uF had lying around, sound million times better. back to the shopping list, identify the values on the Circuit Diagram, order some more orange-drops to replace the yellow/copper caps in the signal path.
     
    smurfer77 likes this.

     

    Please register to disable this ad.

  2. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    pt 2:

    4th Wknd June
    1) While waiting, found intermittent issue of drop outs due to the signal level trimmer-pot at the back of the Amp. Bypassed.
    2) Aux channel issue, took out the RCA socket strip, excess use of solder shorting signal to ground, and the signal cables have been shortened so much they needed to be replaced if any decent connection had to be made. Also, the RCA sockets were very loose so, they all need to be removed and replaced. Time consuming activity.
    3) Finished one set of the RCA strips, required new signal cables being used, used some of the long lengths to replace the short runs.
    4) Centre- channel output needed to be re-wired with chopped components replaced.
    5) More resistors falling apart as soon as you touch them, so ordered a 0.25W resistor pack,, almost 1000 resistors of common values, for any I break as I work through the Amp while pulling at components. Its tight in there.
    6) Pre-Amp Done, lets bias up the tubes correctly and run it up and here the result.. Home-run

    Not quite, as I wound up the Volume to 12'o Clock about 8W RMS, the right channel started to make intermittent thunder/clap noises. Got to be a valve, swopped pre-amp valves, no change, still the right channel, swopped output Valves no change. Lets power up the scope, silence, 2Amp fuse popped on Amp. c**p.

    7) Left it to cool down, changed the 2Amp fuse, Amp meter in circuit and powerup using variac. Wow, current drawn far too excessive with less than 20V AC being fed into the Amp, short circuit!!!.

    8) Isolated B+ to OPT, short gone. Separate the B+ feeds to each of the OPT's. Damn, one of the OPT primary is shorting to ground. Something has cooked the OPT before and when I started to put decent load into them, it has broken down.

    9) Advise HighCut28 of the bad news, some more history of the Amp requested. While doing this, started to ring around in UK transformer repair companies, no luck at all. As the OPT transformers are slightly unusual like the McIntosh type, there was no confidence even to have a go. Absolute tragedy.

    10) Scoured the web and Including sending an email Hasimoto and Isaso of a replacement - no response.
    11) Out of sheer desperation, lets see if there are any scrappies on yahoo.jp. They are still going for lot of money, high demand.
    12) Then I remembered about an article by linux-guru who tried to re-create the AU-111 using toroidal transformers for the Output and had promising results. Sent an email, Linuxguru responded, however, the toroidal showed early promise, but had some issues with high frequency stabilisation. He also stated, that he is in the process of designing a new OPT which he expects to be ready in September. However with the above workaround he also stated rewinding/repair facility is a viable industry in Latvia and offered to help me out.

    July
    1) OPT transformer received by Linuxguru, he has dismantled it and looks like its repairable. Brilliant news.
    2) Started to work on the 2nd row of RCA strip.
    3) been using the Amp on the table, yes one channel only, very good, sound as should be.
    4) Did a THD check, almost 45W RMS with THD between .7 to 1.2% (Yes High, but only single channel being driven) With both channels being driven it will be lower RMS by a touch.
    5) write this up

    Todo:
    Finish the 2nd RCA strip and wire-up.
    Complete the Centre Channel work as per mod.
    wait for the OPT
    Install new Pre-amp valves and balance setup


    Now for some pictures....
    Bad bits, notice the Fuse... Seems the previously replaced capacitors are almost 10th of the value of the design intent...
    [​IMG]

    Before and After change of components
    [​IMG]

    Mod for the lack of Balance Pot
    Balance

    [​IMG]

    The Sansui Iron stripped, curtsey of Linuxguru
    OPT Transformer
    [​IMG]

    Reference info for RCA strip details for the AU-111
    RCA-STRIP
    [​IMG]

    Will post some more, finding time to fix and write up the documentation is a general weakness on me.
     
    smurfer77 likes this.
  3. bluesky

    bluesky Addicted Member

    Messages:
    7,607
    Interesting!! And a Very Early AU-111 too.

    :thmbsp:
     
    BLAH BLAH likes this.
  4. audiobee

    audiobee Active Member

    Messages:
    178
    Mano:
    Did you made a new
    RCA strip for the AU-111 in the pictures?

    I need 2 RCA replacement strips as mine were broken.

    can I get a set from you ?.
     
  5. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    These were the existing original set. I tried to replicate these using 5mm plexiglass/perspex. However, it shattered during the hole making process. Then I stopped trying.
    I may try this again using newer bakealite like Tufnol strip and using a reamer to enlarge the holes. http://www.directplasticsonline.co.uk/Tufnol_Sheet.html
    I will buy a strip and try. Will let you know.
    Mano
     
  6. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    Yes surprisingly the OPT has the date of Year 61 !.
     

     

    Please register to disable this ad.

  7. highcut28

    highcut28 Active Member

    Messages:
    181
    61 huh? year of my birth !
    Amp put together about '65 probably.

    What a saga its been already...
    It was perhaps a silly decision to buy this amp knowing that certain circuits had been bypassed, but i had agonisingly lost an earlier attempt to buy one, and this amp actually sounded ok.
    First restoration attempt was really a non starter.
    It sounded anaemic after that.
    It then sat at the bottom shelf looking forlorn.
    On a later attempt to hook it up for a soundcheck i messed up badly and connected + and + to one spkr and that blew a pair of power tubes and the OPT i think.
    I probably got onto AK looking for solutions for the 111 but it was some time before i hooked up with Mano.
    How awesome is that ?
    This brilliant forum and the way it connects people across the world.
    The amp has has been shipped from Malaysia to the UK and the sick OPT is now in Latvia
    So its fingers and toes crossed that Linuxguru can mend the OPT.
    Or else i ll have a mono amp :scratch2:
     
    BLAH BLAH likes this.
  8. audiobee

    audiobee Active Member

    Messages:
    178
    there is a pair of output transformer just listed on the Bay right now.
    the seller has 0 feedback.. and it is sold as it no return..

    item # 261955436598...
     
  9. highcut28

    highcut28 Active Member

    Messages:
    181
    Ta, will keep an eye on it
     
  10. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    Thanks Audiobee, I've also had another PM from canuckaudiog. Yes will bid. Just in case.

    Mano
     
  11. vigman

    vigman SUBSCRIBER

    Messages:
    4,029
    Location:
    Valencia CA
    MANO BRILLIANT.. standing ovation !!!!!
     

     

    Please register to disable this ad.

  12. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    Mike,
    Appreciated, wish I had the time to restore this to similar to Johan Potgeiters eye catching AU-111work by canukaudiog. During my youth training days in GPO (now British Telecom), any work completed for assessment needed to be like Johan's.
    Mano
     
    BLAH BLAH likes this.
  13. highcut28

    highcut28 Active Member

    Messages:
    181
    wow the pair of transformers went for $360
     
  14. audiobee

    audiobee Active Member

    Messages:
    178
    Mano:
    The dimensions for the RCA strip was wrong.
    the spacing between the holes did not add up to the total length of 94 mm.

    please verify ?
     
  15. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    Audiobee,
    Will do, I have a unmolested original now, I can take the measurements on once I remove the old RCA connectors. Also, I have purchased a sheet of tufnol to play with. Should be able to advise on the measurements during this weekend. If the luck stays on my side and not shatter the tufnol during the hole making process, I may have couple of strips you wanted.
    Mano

    p.s. I think from the edge
    it should be 5mm to the fixing hole
    then 16mm to the first RCA socket
    then 15.5mm all the way through to the 5th RCA
    then 16mm to the other edge.
    then 5mm fix hole to the edge.
     
  16. Hyperion

    Hyperion Roobarb & Custard Subscriber

    Messages:
    48,601
    Location:
    Hertfordshire, UK
    Use a hand drill or electric drill on s-l-o-w, use a low feed rate (don't press too hard). Try to drill from both sides if possible, make a pilot drill hole but not too big.

    Good luck.
     

     

    Please register to disable this ad.

  17. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    Finished the RCA socket wiring. This Amp was playing mind games all the way through the RCA socket replacement. The previous attempt of repair resulted in the output wiring to the speaker terminals being wrong way around. Instead of the Right Channel being on the bottom, it was the Left channel.

    Also, have an Output Transformer ready to be plumbed in from a donar unit, I acquired for spares.

    [​IMG]

    Todo:
    Powerup with replacement OPT,
    Test and measure the Signal Tubes and place them for balance.
    burn it in for a while and then Check THD at max RMS.

    and hopefully start to order the shipping back to Highcut28.
     
    BLAH BLAH likes this.
  18. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    John,
    Yes will do, thanks,
    When I tried it with plexiglass 5mm thick plastic, electric portable drill, slow - very slow was fine with the holes up to 5mm. When it was being enlarged to 8mm for the RCA holes using drill bits .5mm size bigger at each stage, the stress was too much for the Plastic.
    I have plenty of Tufnol now, also found a hand reamer in the tool box, hopefully the tufnol will be more forgiving to the gentle abuse.
    At one stage, I was contemplating being gentle but have the drill speed at max.. instead of slow. . time to play this weekend.
     
  19. Hyperion

    Hyperion Roobarb & Custard Subscriber

    Messages:
    48,601
    Location:
    Hertfordshire, UK
    Regarding drilling plexiglass, the problem is most of us use HSS drills - where the cutting angle of the drill is set for metals. To drill plastic (and tufnol) I think I am correct in saying that you need special bits, where the cutting angle is much less aggressive.

    When I drilled some transparent plexiglass a while back I was forced to use flat wood cutting bits, (and a hand drill), due to the large-ish diameter holes I needed. Almost by accident this was quite a good solution as the cutting angle of those bits is much less aggressive than HSS bits, and just what is needed.

    BTW: Good thread Mano ;)
     
    Last edited: Jul 15, 2015
    BLAH BLAH likes this.
  20. _mano

    _mano Active Member

    Messages:
    456
    Location:
    London, UK
    John,
    Thanks, yes... HSS bits are far too aggressive. I think the answer will be wood cutting bits, will look for a 8mm flat wood bit. Excellent tip.:thmbsp:
    Mano
     

Share This Page