• 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

Sansui AU-717 by leestereo's recipe (yet another)

delboyko

New Member
Hi folks, new member here... from Croatia.

I've got my self into this in February when the Thing arrived. I bought it off ebay, quite accidentally, with unconfirmed working condition. Pictures looked fine so I sent an offer (250€) and the seller accepted. Lucky me.

I read so much about AU-717 and I knew that it needs a proper restore before judging its character.
Here's where my trouble begins. I'm have a degree in mechanical engineering and like to consider myself handy but don't really like or dig electricity (only D in college was Electricity :D) . So naturally I started looking for help. Kale and ilimzn are my fellow countrymen so I tried there first but they were booked for 6 months ahead. I then tried with local TV Video service, whom owner I know very well and were aware that his skills are praised around here. He returned the amp to me saying nothing needs to be replaced, everything works and it would be a sin to replace those fine electrolytes with what he could find in the backyard. That's the part where I realize no body wants to take my money and I got just the thing for the upcoming hot summer nights. ;)

Didn't mention it before but the amp worked. It had an awful smell that filled the room 5 min after playing. I would describe it as old electronics smell combined with something burning. Anyway, the sound was a real mess...too much bass, can't hear vocals over it or isolate instruments. A lot of hum when no music playing. To get something I can cope with I had to crank the treble to 3, and lower the bass to 9 o'clock. One day while I was pretending everything is just fine, I turned the volume to 11 o'clock, went to wash my teeth and after a minute or so music started becoming quieter and quieter. Returned to the living room and that smell was biting me, although no smoke, so I switched it off and put it aside for better days.

I read trough all the restoration projects and made a parts list on Mouser (took me a week to assemble one from leestereos guide). While I waited for the parts I found a PCB to polish my soldering skills on, and to gather all the cleaning fluids and stuff. Just when I was about to start photobucket decided to start charging for hosting, so I lost all the pics from leestereo thread. If that wasn't I sign I don't know what is. :rolleyes:

Under the hood:
IMG_8510.jpg

IMG_8515.jpg

IMG_8514.jpg


I started disassembling, inspecting, admiring, looking and taking a lot of pictures...I was really glad I involved my self in this. Spent a lot of time cleaning that goo buildup on the soldering side of PCBs. Best combo was brushing with soft toothbrush - acetone first, than isopropyl. It ended up shiny like from the factory. Besides me, nobody seems to have touched the amp. It was quite clean too...barely any dust. I decided that the owner took a proper care of it or didn't use it at all.
After cleaning, priror to desoldering.
IMG_8661.jpg

(lifted traces are not my fault, that's factory)

NEWBIE ALERT: While desoldering I learnt that you need to pull the components from the component side because if you cut them first to make it "easier" (like resistors) and pull it from the soldering side you can lift the traces. I haven't heard anybody mentioning that in "how to solder" tutorials. :angel:
Luckily, no harm was done(checked for connection) so I didn't do much about it.

I presumed that all the wires will be safely returned to its place just from pictures. Than, I noticed a lot of wires are red white or yellow and grey. That's when I had to learn to read schematics and went trough the process of marking all wires. Of course I checked it X times. :D

Since my equipment was basic, but again decent, I didn't check the electrolytes capacitance. Glued ones looked OK, non were bulged, all in all it wasn't as near severe as I've seen some around here. Although power board was black near several components and it couldn't be cleaned. Driver boards were even worse! A lot of heat I suppose but some components that were the cause of that didn't look "burnt" at all. Volume knob didn't make any noises but it was almost stuck. Fixed that with some fine silicone lube applied directly to shaft which was preheated. Repeated that a few times and now it literally flies with beautiful feedback. Balance knob produced some crackling noises though, but nothing deoxit and a few spins can't solve.

Glue was removed with a round edges screwdriver sort off. Tried with the plastic scraper but glue bit well in 40 years. You won't scratch the PCB unless you have something sharp. Acetone helped little to nothing in my opinion. I soaked it 10 times, waited an hour, waited 5 min, it's still a pain to clean. Arm yourself with patience and something round and you're good.
Some before/after glue pics for you:
IMG_8677.jpg

IMG_8681.jpg


Driver board burnt around zener and resistor :
IMG_8706.jpg

20170723_134748.jpg


I changed the thermal paste and heatsink insulators as well. Amp is barely producing any heat. I can hold my hand on all the components except those bloody zeners (ZD02) on driver boards. I can't put a finger on them and nearby cap(C04) together with the resistor(R15) is also very hot. It's in that What's with that?
On tone control board I ordered wrong 22pF caps (C07,08 & C23,24) so I left it as it is. I can't get branded C0G caps in Croatia so they'll have to wait for some other occasion. Power supply caps weren't replaced. All other equipment was exactly according to leestereo recipe. Nichicon, WIMA, Panasonic, Vishay...everything.

While I was at it I said there must be something faulty and started checking all the resistor values and comparing them with schematics. So much fun. :)
All the fusisters were according to schematics. Although, I found at least 20(counting all PCBs) of them to be off but as much as I figured it that doesn't mean it should be replaced. I'll mention some of them here because I would really like to hear your opinion. In the process I found one that was completely dead. It was R31 (47ohm, 1/2W) fairly black, and had no continuity. Only one channel but funny there was no difference between channels while listening.

f2721.png

Both channels have the same "wrong" measurement in case of R05,06; R26,27; R32 & R39.

After cleaning the faceplate and knobs I discovered that there wasn't even a scratch on them. Just Sansui galaxy black perfection. Lucky me.
20171001_120218.jpg


Power/Protector LED is not working any more. Checked it with a multimeter and it's dead. Not sure how that happened. Strange thing though because now my input LED is blinking instead of it. Someone care to explain please.

What I would like to add is how I tested my work with the dim light bulb. There are various tutorials concerning the topic how to make one. Buy this buy that, connect here, put it in a box, make a stand, get this bulb, get that switch bla bla. What I did is super simple!
Found an old table lamp in the basement. Bought a simple socket. Cut one wire of the cord and connected it in series with the socket. DONE!
Inserted the 75W bulb, prayed to god, switched it on... Lights on than in a few seconds to barely visible light.

Connected all the wires, and the music started playing. It's like a got a baby, even delivered it. Such a marvelous feeling! :music:

The sound was really nice. Hum audible only when you crank the volume over 50% with no music. Nice balance (not perfect) with tight bass, good detailing, sound stage easy to recognize, a bit dark. Voices are a little behind but depends on the song. After a month of listening with tone and loudness off i started to have objections. I noticed that treble is falling behind, so I brought it to 3 o'clock and it's much much better now. Voices are also somewhat better but still not to my taste. Since I'm the first making such complaints on AU-717 behalf I guess something is not right. It's like I've could have done more.

I got a decent DAC and Klipsch RF-52 speakers which are quite capable as far as I'm concerned. Source is 320 kbps audio from Spotify.

It took me 20 days to finish it all. Effectively maybe 10. It's a really straightforward job if you'r just recaping. Actually, it's straightforward only if you have such great community as audiokarma behind you back. I took a lot of pics but somehow they disappeared, especially those after re-soldering. Sorry about that. :beerchug:

WARNING
If you're using "4seatPilot" parts list beware! On driver boards, R16 & R17 are claimed 27 Kohm. In case of R16 it's true but R17 is 1,2 Kohm.

Ode to the fallen:
20170725_065250.jpg


I got my hands on multimeter with capacitance meter and started measuring. Some are down to 60-70% but a lot of them is in tolerance or even measure more than stated. Every single one returned a +50% value. :rolleyes:

I'll post a Mouser order list soon. I need to correct a few things first.
 
Register to hide this ad
Nice write up sir.
If your power led cooked and is now not operating my bet is the red and white wires off the top of the power supply board (there are 2 each and they look almost identical) are reversed, especially possible if you removed all the wires from that board for cleaning and glue removal.
Open r47 is unusual, given that it is part of the offset adjustment chain (vr1/vr2) having it fail probably disturbed the offset before you repaired the amp, may have been a factory fail safe for the fet in case one of the variable pots went open, others may offer other opinions of course :)

-Lee
 
While I was at it I said there must be something faulty and started checking all the resistor values and comparing them with schematics. So much fun. :)
All the fusisters were according to schematics. Although, I found at least 20(counting all PCBs) of them to be off but as much as I figured it that doesn't mean it should be replaced. I'll mention some of them here because I would really like to hear your opinion.

It is very unusual for resistors (other than fuse resistors), to drift - especially the large numbers of resistors, and by the large amounts you have recorded. This leads me to ask the question "you do know that to get an accurate resistance reading from resistors you need to de-solder one end of each one and lift it out of the PCB"?

Measuring them 'in circuit' just means you are measuring the resistor and everything that is connected in parallel with it. :) - which could explain the readings you were getting.
 
Last edited:
Delboyko, welcome to AudioKarma, Exclusively Sansui. Thank you for the detailed description of your restoration. Please post the serial number for the database. Click on the link in my signature block.

- Pete
 
Delboyko,
Really enjoyed the story, especially "Sansui galaxy black perfection."

One question to the experts, shouldn't you replace the fusible resistors as a general rule?

Mark
 
shouldn't you replace the fusible resistors as a general rule?

In a word, yes - the vast majority (including me) replace with the same wattage Metal Film equivalents, even modern fuse resistors being considered too unreliable. It appears to me that they were part of a belt-and-braces safety drive by manufacturers, to include 'safety components' to be 'extra safe', but which actually just interfere with, and compromise system operation in many cases.
 
Last edited:
Hi guys, I just recapped my AU 717 power supply bd (not the main 4 caps ) with Nichicon Fine Gold caps. now I'm concerned, Should I pull them out & swap for the long life low ESR 105 degree Panasonics while its apart, or should the nichicon FG (shorter life, 85 degree) caps work out fine...???.
(also, I used the Microcare Tidy pen adhesive remover to remove glue, glue turned to gel in minutes and wiped away with pencil eraser).
Please advise if the FG caps are really not suitable in PS bd. What would you do?
Thanks
Denis
 
Hi guys, I just recapped my AU 717 power supply bd (not the main 4 caps ) with Nichicon Fine Gold caps. now I'm concerned, Should I pull them out & swap for the long life low ESR 105 degree Panasonics while its apart, or should the nichicon FG (shorter life, 85 degree) caps work out fine...???.
(also, I used the Microcare Tidy pen adhesive remover to remove glue, glue turned to gel in minutes and wiped away with pencil eraser).
Please advise if the FG caps are really not suitable in PS bd. What would you do?
Thanks
Denis

Denis, you should post a new thread for this question as your post constitutes a 'thread crap' - the rule is 'new problem or question = new thread'. It might seem logical to post in a thread where people are talking about an AU-717 and rebuilding - but actually it technically contravenes forum etiquette and comes across as bad manners.

The particular question you have posed could get many replies, how do you think the OP would feel about loads of PSU capacitor choice posts appearing in his AU-717 restore thread, and thus effectively 'hijacking' it?
 
Here is proof why. A recent bunch from a little HK PM635i integrated I restored. Have a look at the rated value and the measured value and a % error. There was one only (1)! I would regard as OK.
Interesting post John.

Does this mean that fusibles generally increase in value when they "drift", can they go the other way? I have read a few posts about where people have reported fusibles as being open - so I just wondered.

Cheers
 
Does this mean that fusibles generally increase in value when they "drift", can they go the other way?
If I may give an answer to this, under fault conditions they are intended to drift high - always, this is to ensure they are always reducing current flow, obviously if the fault they are protecting against is serious enough it will cause the limits of the resistor to be exceeded and it will 'open'. Also if they reduced in value under fault conditions thus increasing current flow, this will simply cause the damage to spread and this would negate their original purpose.
 
Does this mean that fusibles generally increase in value when they "drift", can they go the other way? I have read a few posts about where people have reported fusibles as being open - so I just wondered.
Open is high - infinite resistance. "The other way" would be a short, zero resistance. I've never seen that happen with a fusible resistor.

- Pete
 
Denis, you should post a new thread for this question as your post constitutes a 'thread crap' - the rule is 'new problem or question = new thread'. It might seem logical to post in a thread where people are talking about an AU-717 and rebuilding - but actually it technically contravenes forum etiquette and comes across as bad manners.

The particular question you have posed could get many replies, how do you think the OP would feel about loads of PSU capacitor choice posts appearing in his AU-717 restore thread, and thus effectively 'hijacking' it?
Thanks for the Education Hyperion, i see your point & agree. i'm a bit green on these forums. but learning.
 
Back
Top Bottom