• 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

AU-919 Protection Bd. Transistor Help!

noveltone

noveltone

Active Member
Hello guys and gals,

It's nice to be starting a thread for the first time in a while! Recently EW said that as a matter of course he replaces the relay driver transistors on amps' Protection boards. Like a little sheep I would like to do the same to my beloved AU-919 since I'm going to go in there and try out Nicowico's fast output transistors, among other things. :zoom:

Its always blinked forever when I turn it on, so maybe this will help that issue. Anyway, I'm pretty much out of my league in determining which of the transistors on that board are the relay drivers, but here are my guesses: TR601, maybe TR604, and TR607. I've attached that portion of the schematic for your delight.

Now, since these are not even audio-application transistors, I am way out of my league in which stats are important to try to match and which aren't. :scratch2: Would one of you kind, knowledgeable folks take a look at these and see if can recommend a commonly available modern replacement? Thank you very much!

TR601: 2SD357

Original:
http://www.datasheetarchive.com/pdf-datasheets/Datasheets-9/DSA-167009.pdf

Suggested Replacement somewhere here on AK, maybe CDfixer: 2N4920
http://www.onsemi.com/pub_link/Collateral/2N4918-D.PDF

TR604: 2SD438

One thread suggests 2sc2235, but that isn't available at Digikey. Any other ideas?

Original Datasheet:
http://www.datasheetarchive.com/pdf-datasheets/Datasheets-8/DSA-158376.html


TR607: 2SC945

This one is pretty much covered by this thread: http://www.audiokarma.org/forums/showthread.php?t=129112&highlight=c945

Therefore I'll use the available 2N3904 for this application.

Thanks again!

Collin
 

Attachments

  • PROTECTOR -2.jpg
    PROTECTOR -2.jpg
    192.6 KB · Views: 79
Register to hide this ad
The 2SC945's are also used in the AU-417 that I presently work on. One 2SC945 is blown. I got hold on a few NTE-85 's, that I was told is a replacement for the 945's. Don't know how they will work? I've not installed them yet as I'm working at sea at the moment. Attached the specs. Would this be ok?
 

Attachments

Sansui tends to have their transistor holes all lined up in a row, so it's a lot easier for the tenderfoot tech to use a transistor with a collector in the middle, like the original.

Change all the electrolytic caps on the board, and swap out TR604 and TR607 with a 2SC1509(S).
 
Engels13, thanks for your at-sea reply. I don't know how to evaluate the transistor, but thanks for your suggestion. I know they're commonly available.

EchoWars, thanks for your suggestion as well. It helps to hear it from you on this, since you were the originator of this particular idea. Also, digikey stocks that transistor, which is nice of course. So TR601 is not involved, eh? interesting.....

The lead pattern I didn't even think to look at yet, but I see your point that its a lot easier than having to bend them around...

Thanks to you both.

Collin
 
TR601 is a simple regulator. As long as there's something like 23V or thereabouts on the emitter, it's doing its job.
 
My AU-919 blinks like crazy too before kicking out of protection. Call me a sheep; I am going to do this here little transistor swap too! I have replaced just about everything else: relay, electrolytics, mylars, micas, power supply transistors, and trim pots.

I listen to the amp daily with the signal path BIC FM-10 -> TU-919 -> SE-7 -> AU-919 -> AKG-701 cans and it sounds great but it does take longer than any of my other gear to come out of protection.

Thanks for the idea and information.
 
Aha! I see upon a much closer look that TR601 only regulating the power to the relays. One relay, 601, is operated manually by the speaker selector switch. This makes available the power amp's output to the A speakers. The other two are operated by the protection circuit - TR604, TR607 directly. In the event of a protection-causing fault, one relay cuts the signal to the power amp, the other cuts power to the B speakers and headphones.

I thought the B speakers had their own separately operated relay, but this is not the case. They are simply made available via switch. Of course, I trust your comments EchoWars, I was just trying to learn something from them as well.

Man, this stuff takes a great deal of time for those who, like me, don't have a ready vocabulary of circuit topology. Without microprocessor control, Sansui sure got a lot done with manual switches. Anyway, thanks for everyone's help.

Xoaphexox, the longer I own this amp, the more stuff I'm tempted to replace. It's my only good amp, so I want it to last, but it really sounds so good, especially the phono.

Collin
 
Engels13 said:
The 2SC945's are also used in the AU-417 that I presently work on. One 2SC945 is blown. I got hold on a few NTE-85 's, that I was told is a replacement for the 945's. Don't know how they will work? I've not installed them yet as I'm working at sea at the moment. Attached the specs. Would this be ok?
Click to expand...

Not trying to hijack but 2sc945 are usually easy to find and are generally priced where you can buy about ten for the price of one nte85.
Don't know about your country but parts express, B+D carry them here for .15 to .20 cents ea.
 
Nte

Engels13 said:
The 2SC945's are also used in the AU-417 that I presently work on. One 2SC945 is blown. I got hold on a few NTE-85 's, that I was told is a replacement for the 945's. Don't know how they will work? I've not installed them yet as I'm working at sea at the moment. Attached the specs. Would this be ok?
Click to expand...

NTE semiconductors are not looked upon favorably here, poor quality.
Consider other sources for a more reliable part.
Casey
 
I've bought some stuff from those guys in the past, and the labels looked convincing, but that project never worked out well enough to tell if they were fake transistors or not. I know they sure stock a lot of semiconductors that haven't been made in a decade or so. Makes you wonder. For myself I'd rather stick with what a Digikey or Newark stocks, in current production, by a reputable manufacturer. To each his/her own....

Collin
 
nosirrah said:
NTE semiconductors are not looked upon favorably here, poor quality.
Consider other sources for a more reliable part.
Casey
Click to expand...

Thanks Casey. Yes, I was not aware and after reading Echowars comments I've put the NTE's over the side and followed his good advise to order some 2SC1509(S).
Thanks guys, Pieter.
 
I replaced TR604 and TR607 tonight with two new 2SC1509(S) from Digikey.

The unit takes 31 flashes of the protection light to kick in - the same as it did before replacing these transistors.

I had a slow KA-9100 recently that was remedied by replacing the caps on the protection board. This AU-919 has had all it's electrolytics replaced too but it didn't shorten the pause.

This unit takes longer than any of my other units to come out of protection. Could there be another malfunctioning part on the protection board or does this unit simply just take longer than others to turn on by design?

FWIW the unit was run for over an hour and then it's bias and offset were adjusted to spec.
 
It might simply be taking that long for the offset to settle to a value acceptable to the prot. circuit.
 
EchoWars said:
It might simply be taking that long for the offset to settle to a value acceptable to the prot. circuit.
Click to expand...

So I'm guessing it is caused by an issue in the driver board then. Is this something to worry about or should I just ignore it? I think it sounds good right now. I just don't want to risk losing those original outputs.
 
Watch the offset yourself with a meter if you want to comfirm that this is the case. No need to guess or wonder.
 
EchoWars said:
Watch the offset yourself with a meter if you want to comfirm that this is the case. No need to guess or wonder.
Click to expand...

That would make too much sense! :)

I hooked a meter up on each channel. As soon as the unit comes out of protection:

L channel: 74mV (and falling)
R channel: 16mV (and steady)

When I recently replaced the trim pots and silver micas I dialed in the values and both channels settle in around 0mV but I will check next to see if this is still the case.
 
xoaphexox said:
That would make too much sense! :)

I hooked a meter up on each channel. As soon as the unit comes out of protection:

L channel: 74mV (and falling)
R channel: 16mV (and steady)

When I recently replaced the trim pots and silver micas I dialed in the values and both channels settle in around 0mV but I will check next to see if this is still the case.
Click to expand...
Just when you thought you were making sense...

Obviously the offset is low once the relays close...we want to know what the offset is before the relays close. ;)
 
EchoWars said:
Just when you thought you were making sense...

Obviously the offset is low once the relays close...we want to know what the offset is before the relays close. ;)
Click to expand...

D'oh!

Okay well after having the unit on for 1.5 hours while I played a video game, the DC offset are a few mV in both channels as well as both channels of the flat amp.

Echo, are we looking for a specific value here? What I am getting at is; if I was able to pull this apart and get the value in question (let's say 250mV for sake of argument) is it going to help narrow down the faulty part or is the fact that the offset is that high to begin with enough information?

Is the offset controlled by the transistors TR19 & TR21 (TR20 & TR22) on the driver board? I am going to take some voltage readings to make sure it matches what is on the schematic.
 

Attachments

  • Clipboard01.jpg
    Clipboard01.jpg
    142.8 KB · Views: 18
Well at least now I have some "hard facts" that something is amiss:

TR21:

-3.86V (-3.5V in schematic - should be OK) collector to chassis ground
-52.6V (-52V in schematic - should be OK) emitter to chassis ground

TR19:

4.3V (3.5V in schematic) collector to chassis ground
51.1V (52V in schematic) emitter to chassis ground

Any wisdom you could share would be good. Should I take more readings?

I noticed the unit came out of protection about the time that TR19 reached 50V if that means anything
 
Anyone please correct me if I'm wrong here. In order to do what EW is saying, you have to read the offset prior to it entering the relay, which won't conduct until it closes, of course.

The way I read the schematic (found in the first post, but only partially shown) is that the output from the driver comes into the protection board on wires 1 and 2. These test point are shown in the attached image of the board. If you measure between each of the two points and chassis, this is the reading before the relay. Number 1 is the left channel, having the highest reading you measure, so maybe do that one first. If you turn the amp on, watch your DMM and you'll probably see the offset drop from something high to around the 74 mV you measured, then click over. I'm curious what how it fluctuates. Please tell! :)

Collin
 

Attachments

  • protect-tp.jpg
    protect-tp.jpg
    111.5 KB · Views: 26
You must log in or register to reply here.
Back
Top Bottom