AU555 - Driver transistors heating up

afroaudio

Active Member
Hi gents,

I have an AU555 that I have recapped. Also replaced 2SC458s and flying saucers as well as output transistors (MJ21194G) trim pots and most resistors on main amp board. All thanks to others here that have gone before me and documented the process :)

After solving another problem with the preamp I now have a working amp which sounds great...Bias has been adjusted but not clipping (no scope)
My problem now is my replacement driver transistors are heating up very quickly as soon as signal is put into the amp. The flying saucers were replaced with NEC 2SA1220A and KSC2690A. After a minute or two I can smell that something is getting very hot and traced the heat to those driver transistors, (can barely touch them)

Can anyone point me to where I should look for faults here? I am probably going to take the amp to a tech tomorrow but thought I would ask here also.
555 main board by bregtje_lush, on Flickr
 
Last edited:
just playing spot the difference and wondering why tr803 and 804 are installed differently ? look where the tabs are pointing .
 
Since nobody has responded, I'll take a stab.

I assume you tested all the new trannies and diodes before installing them. Were you firing it up in stages as you replaced components? Are your new drivers correct replacements? Do they have different pin outs from the originals?

I'm not familiar with that unit, but, after triple visual checking all my work for bad solder joints, shorts, and damaged or incorrectly installed components, I'd start by checking voltages against the schematic, with the preamp connected and then disconnected. Maybe run another preamp into the amp and seeing if it the drivers overheat. Maybe pull the finals and see if the drivers still overheat.

And what was the problem with the preamp?
 
Sounds like too much bias current - or an unknown 'mistake'. (are we sure all those nice new resistors are the correct value?)

How are the output transistors for temperature?

Quite neat work, from the topside view ;)
 
Last edited:
Hey guys thanks for replies! TR803 and 804 are original on the board and werent touched. I figured the fact that the amp as a whole is functioning, sounding good and that bias adjusts as expected (set to 24mA as per service manual) means that at least I have everything oriented correctly. New driver transistors are definitely good replacements that many others have used and had success with: 2SA1220A and KSC2690A. Output trannies are good. Preamp is clean and working perfectly. I will be sitting down with my tech tomorrow hopefully to check voltages against schematic and try to troubleshoot further. My electronics knowledge is shall we say basic, but I can solder and I can read Audiokarma threads til the wee hours.. which is where I hopefully learnt from those that went before. I am confident all the new bits are correctly chosen.. just hope I put them all in correctly! Just seems weird for everything to sound perfect but to have this heating up going on... maybe go triple check my resistor replacements
 
Last edited:
The only thing I could think of if the drivers heat up too much is if you did put them in but forgot to bolt the TO-3 transistors properly into the sockets.

Then you have the drivers supplying all the power while base and emitters connected of the TO-3 transistors, which can go unnoticed if you do not tweak up the volume.

I know, this is a very remote possibility....but I remember doing such myself once with some kind of amp. Without any thinking what I was doing, I bolted them with small plastic insulators on the bolts, which was the wrong thing when using sockets, of course, resulting in collectors not connected.
 
as above ...good theory ..but if bias is correct on the outputs they should be dealing with the current .
 
Board doesn't seem to crowded - but have you checked separation of the traces for the new installs - it is easy to have one solder blob spill over to the "neighbor" - If any doubts use a magnifying glass or continuity checker on every install.
 
Is it the driver transistors on the board getting hot, or the outputs? It's hard to imagine the outputs having the correct bias and no problem adjusting it, if there were anything wrong with the driver transistors. Check the resistor values in series with the drivers with a meter to confirm they aren't off by 10X or something. Do all the devices overheat on both channels, or is there a difference? Another possibility is oscillation. The amp could appear to work correctly, but overheat. Usually oscillation will get the outputs first. Also check voltage drops on resistors to be sure an individual output isn't shorted to ground or something.
 
Conrad I think he said the driver transistors were heating up. What biases them?? Did he change TR-815? I started to look at this but got sidetracked. That silly Class AB sub amp I was working on last month had a bad Darlington hidden on the board which was overdriving the .... Out of my drivers and smoking the outputs and drivers alike.
 
maybe take voltages across the driver resistors and see how much current they are drawing .
another thought see if the drivers get hot without a speaker load
 
I can see one of those TO126 transistors incorrectly installed.....

Transistors will heat up if you get the pinout wrong, or you put a PNP where there should be a NPN or vice versa.....
 
Hey Kevzep, amp is with my tech now and not in front of me but certain those TO126s are correct.. they are all ECB pinout. They are KSC2690A Fairchild and 2SA1220A NEC.
Otherwise checked for solder bridges and the ON Semi MJE21194 outputs were tested and I believe correctly installed. The 2690 drivers on both channels were the ones getting hot. This only happened when the amp was driving speakers after a few minutes.
Hope its not oscillation... I did manage to pull some 2SD180s (original outputs) from a receiver that got parted recently and could try those depending on what my tech finds. I made the mistake with this job of changing too many components at a time and not checking my work in stages. Definitely learning a lot though.. will report back soon
 
If you say they are ok - then they are ok....

But I have to say the one above D801 looks wrong, having gazed at it for a while, and at first I thought the one above D802 was wrong as well - but not sure now.

The difficulty is seeing the metal plate on the transistor body for orientation. :yes:
 
Hyperion yes thought that might be what you were looking at... confusing the issue is the KSC Fairchild's dont have a metal plate exposed while the 2SA NEC's do.
Thanks for all the suggestions guys... if my tech doesnt quickly find my screwup I have a list of things to check :yes:

Big lesson for this project is that I should be patient and do small changes and then test. Problem was this amp was not working to start with so I went ahead and did a full recap, replaced 2sc458s, saucers and outputs (Right channel outputs were bad) instead of first trying to isolate what the problem was.

Will get there in the end! Thanks guys :thmbsp:
 
Don't know if this helps, but driver transistors will overheat when the outputs are installed, but the collectors of the outputs aren't connected. This can be to bad wires, problems with the socket the output is plugged into, or, in one case, because they used insulated shoulder washers on the screws to the outputs. The connection is through the mounting screw.....
 
Back
Top Bottom