Hot Azur

For the three and more legged things
User avatar
floppybootstomp
Old Hand
Posts: 1255
Joined: Wed Feb 18, 2009 10:37 pm
Location: Greenwich

#1 Hot Azur

Post by floppybootstomp »

I have a Cambridge Azur 640A amplifier that overheats. It has a protection cutout circuit that shuts it down when it gets too hot. I have to unconnect and reconnect the mains lead to power it up again.

It is in a cabinet but I have moved it to a better ventilated area than where it used to reside. It does get extremely hot in use (not passive) and my solution up until now has been to let a household fan blow cool air over it from the rear.

What I'm wondering is why. It did cutout a few times in the first six months or so but it has become worse.

Here's a picture of the internal:

Image

I have tightened the bolts on the 4 output devices. The heatsink is merely warm at the front of the amplifier and becomes too hot to touch at the rear.

This would indicate, perhaps, that the output for one channel is overworking/drawing too much current/faulty. I don't know.

The amp works perfectly and sounds fine.

I am considering fitting an 80mm mains fan in the front right hand side angled at the main heatsink in a bid to cool things down.

The three (what I assume to be) mains regulators to the left of the amplifier also get very hot but not as hot as the output device heatsink.

So if anybody can offer a possible cause and/or cure I'd be grateful as I'm currently running it with the lid off so not to have to use the mains house fan.

And just out of interest, I don't have a circuit diagram, so am wondering what that small heatsunk semiconductor is between the two output pairs?
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#2

Post by Mike H »

Suggests quiescent current is too high for one channel or the other. Or both. How you would adjust that I don't know, if there isn't a preset for it.

Dunno what the supply V is but suppose it were a modest 40V total, and the quiescent were 100mA, that's 4 Watts while it's just 'idling' silent.

I see there's what looks like a small power transistor in the middle of the heatsink bracket, maybe this is the temp sensor part of it :?: Though usually there's a separate one for each amp. Unless there's another one I can't see. Or maybe it's part of the thermal shutdown gizmo.

.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#3

Post by Mike H »

Are those output devices actually chips, not trannies?


.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
floppybootstomp
Old Hand
Posts: 1255
Joined: Wed Feb 18, 2009 10:37 pm
Location: Greenwich

#4

Post by floppybootstomp »

Mike H wrote:Are those output devices actually chips, not trannies?


.
Dunno, they're not wearing stockings, dresses and makeup :D

In all seriousness, thanks for your reply, I shall check out the number on the packages and look them up.

Looking at the pic I note they're five legged devices so likely IC packages.

I'll also try and find out if I can lay my hands on a cct diag.

Thermal sensor? Quite likely, good call.

The amp is rated at 80 wpc so I would expect it to dissipate a fair amount of heat, but you could fry eggs on this thing.
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#5

Post by ed »

Hi Tony

devices are darlington SAP15s and I believe they do have temp detect....
If you want the workshop manual pm your email address and I will send it...

Ed
There's nowhere you can be that isn't where you're meant to be
User avatar
floppybootstomp
Old Hand
Posts: 1255
Joined: Wed Feb 18, 2009 10:37 pm
Location: Greenwich

#6

Post by floppybootstomp »

Ed, you're a gent :thumbup:

E-Mail pm'ed to you.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#7

Post by Mike H »

Ah-hah!! They're just complementary darlingtons, but wiv diode ladders. There's cunning. According to datasheet still need a preset and should be adjusted for 40mA idling current.

.
Attachments
sap15p-sap15n.gif
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
floppybootstomp
Old Hand
Posts: 1255
Joined: Wed Feb 18, 2009 10:37 pm
Location: Greenwich

#8

Post by floppybootstomp »

Yes, Ed very kindly e-mailed the workshop manual for the amplifier and a datasheet for the Darlingtons for which I thank him.

This from the manual:

To set the bias on the 640a amplifier:

"Measure the voltage across R22 and R100 (left channel) individually; they should be adjusted using the preset R116 so that the voltage measured on each resistor is greater than 13mv.

This procedure should then be repeated for the right channel by measuring the voltage across R52 and R69 whilst adjusting preset R68 so that the voltage measured on each resistor is greater than 13mv."

So that will be first thing I do, sometime tomorrow.

Still a bit of a mystery to me though, whilst one pair should run hot and the other pair relatively cool. And also the fact the amp is sonically faultless.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#9

Post by Mike H »

Most likely one of the presets has gone dirty track and increased resistance.

Before turning it on suggest you make a note of each one's angle, then twiddle it lock to lock a few times to restore contact. Then back to where it was, then do the adjustments with the power on.

Edit or, if you know which end position gives minimum resistance, start from there.

I'm surprised at this actually, what happened to the "fail safe" version that we used to see so popularly about 30+ years ago?


.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#10

Post by Mike H »

Appendix
Has anyone else noticed that in the above diagram there are more diode junctions than there are transistor junctions?


.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
floppybootstomp
Old Hand
Posts: 1255
Joined: Wed Feb 18, 2009 10:37 pm
Location: Greenwich

#11 Eventually...

Post by floppybootstomp »

Finally got round to doing this thing and it's cured :)

Bias on the right hand channel was acceptable, 15mv, but the bias on the left hand channel, the one getting very hot, was 150mv.

Adjusted it to 15mv and no more heat. It's a little warm but the heat is now uniform on the heatsink serving both output sets. Previously the heatsink was literally too hot to touch.

The presets, in the picture, are the square blue multi-turns between each output pair and the resistors where bias is measured are the large grey pair near each output pair.

Thank you chaps and special thanks to Ed for supplying the manual.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#12

Post by Mike H »

Oh yes I see, with the adjuster screw on top.

Well done Image

Still doesn't explain why it should have gone off that much


#
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#13

Post by Andrew »

Could the preset have been wrong from the outset 15mV .... 150mV human error when reading a meter?

-- Andrew
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#14

Post by Mike H »

That's a possibility, decimal point in the wrong place?

#
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#15

Post by Andrew »

At first I thought why wouldn't they have a special test harness built to help set the bias, you know red light, green light affair, but then wondered if they would bother actually building something when a cheap meter would suffice...
Last edited by Andrew on Sun Nov 01, 2009 8:08 pm, edited 1 time in total.
Post Reply