Tech 21 - Trademark 60 Problems!! Any Help Here?

[bogner100b]

I've got a Trademark 60 that sounds amazing!!!  However after 15 minutes she shuts down.

1) Caps?
2) OP Amps?
3) Cold Solder joints?
4) OT?

Any one who can point me in the right direction would be awarded with two thumbs up :dbtu:

[Enzo]

is anything getting overly hot?   DO power supply voltages remain or go?

What does "shut down" mean?   All signs of life go off?  Lights stay but sound goes?  Display freezes?

[Roly]

1) Caps?  - unlikely
2) OP Amps?  - unlikely
3) Cold Solder joints?  - unlikely
4) OT?  - highly unlikely; it appears to be solid state and therefore almost certainly won't have an output transformer.


Guessing doesn't work.

First we identify what is actually wrong, then we fix that.

Being solid state my first guess from your definite "after 15 minutes she shuts down" (which I assume is 15 mins of flog, not just sitting idle or quietly noodling) would be that it has a thermal cutout.  If it isn't getting hot internally there is still a chance that a thermal cutout itself has gone faulty and operating at a lower temperature.

Being a 60 watt amp with no obvious heatsink, and only a tiny vent on the back, reinforces my view.  I wouldn't build an amp like this without a fan (but then I wouldn't build a 60 watt amp in this manner).


Does it come alive again if left to cool down?

Are you running it straight, or have you added an external speaker cab?

Please pull the chassis an post a couple of overviews of the internals.

[joecool85]

...I wouldn't build an amp like this without a fan (but then I wouldn't build a 60 watt amp in this manner).

Interesting you say this as I've always been curious to try out one of these amps.  How would you build one/do it differently?

[Roly]

So the heatsink(s) can breathe.

An early incarnation of my Twin-50 with three Mullard 55D heatsinks on the back;



A heatsink is not a magic devices that makes heat simply vanish, it's a coupler to the air, and if the air it is coupled to is trapped inside a chassis then all that is going to happen is that the whole shebang is going to get muther-hot.

It is a common design error to under-estimate the amount of heatsinking required, driven by the facts that heatsinks tend to be expensive, and mounting them properly, fins vertical in free air, can be difficult or conflict with a desired overall layout or look.  I have struck this thermal "optimalism" so often in commercial gear it's not funny - e.g. a steel chassis makes a truly lousy heatsink (and even an ali one isn't much better).

And they are rated to operate in 25 degree C free air, not confined inside a box.  You can model heat in sims like LTSpice and it requires a thermal gradient (like a voltage gradient) for heat to flow (like current) from the chip to the case, case to heatsink, heatsink to ambient air - if the air is, say, 65C, then everything back to the chip dice will be hotter still.

In this case there seems to be only one tiny little vent to get rid of about 60 watts worth of waste heat (60W amp, 50% efficient), and I suspect that the ultimate answer here will be to fit a small fan over that vent inside to draw in cool air.

It might be easier to visualise if you imagine that chassis with a 60 watt light globe belting away inside.

[bogner100b]

Hi Guys, thanks for the questions.  The amp does not power off after 15 minutes (all lights are still on) but the amp "farts" and then output stops.  Then if I turn off the power switch off/on it powers back up. 

Take a look at some pics i've attached

[Roly]

Those are nice detailed shots, but not an internal overview.

1 - Okay, let's try this another way - is there a round capsule (thermal cutout), about 5/8ths inch in diameter, with two leads coming off, attached to the heatsink somewhere?

2 - How hard are you hitting it when the fault happens?

3 - Is it always about 15 (+/-5) mins, or sometimes much longer or shorter time?

4 - How hot is the amp getting (back, near vent)?

5 - Does it still play up if just left idle for half an hour or more?

6 - Are you driving an external cab?

7 - Does it *always* recover on power down/up?

Try running it with the chassis out and open and see if it does the same thing.

[bogner100b]

The amp still operates outside of the cabinet, even with the "cooler temp" out of the cabinet it still shuts down.  As far as a (thermal cutout) the design didn't include one in the layout of the amp (poor oversight).

[tonyharker]

I expect this amp uses the LM3886 IC which has a heat sensor/thermal cutout built in. The symptoms you are getting points to this shutting down.
In the picture raul.jpg what is the brown material under the IC under the metal clip? Looks like cardboard?

[bogner100b]

Tony - in another forum a member noted that the Part Number is LM3876T and that when he swapped it, the problem went away.  But you think it could possibley be a LM3886?

And yes it appears to be cardboard from factory, to maybe protect against grounding out?


Thanks
B

[tonyharker]

If it is cardboard, then there's your problem.  It should be a special silicone rubber/fibreglass washer (Sil-pad) to transmit the heat.  Cardboard is a heat insulator!
LM3876 or 3886 there's not much difference between them.  They have similar pin-outs, but 3876 is 56W and 3886 is 68W outputs.  Both will shut down if overheated. 

[bogner100b]

Tony - I'll check what it is when I pull it and replace appropriately



Thanks!

[gbono]

The output chip in your amp is the LM3876T and it is available from Digikey for $2.75 without shipping. There is a "pad" that Tech 21 puts on the heatsink to keep the backside pad on the chip from shorting to the heatsink (chassis ground). If you have an IR thermometer you can check the temperture rise on the device to see what is happening. Most if not all of this board is surface mount technology (I don't remember seeing anything smaller than 0805).

Heatsinks are interesting creatures - I have seen several demos using FEA to show that there is a "diminishing point of return" on the number of fins needed for a given application. Not enough and there is a lack of heat transfer and too many (or wrong geometry) and you get turbulence that diminishes heat flow....

[joecool85]

I'll be interested to see/hear what happens...I've been itching for some Tech 21 gear and its always good to know what I'd be getting myself into :-)

[Roly]

Like Tony I'm also suspecting thermal shutdown of the power output chip, but from the pic it looks like an impregnated thermal pad to me, and if it were really cardboard I don't think it would run for 15 seconds never mind 15 minutes.


Just a tip bogner100b, I'm a tech, but you are my eyes and ears, and if you don't actually answer the questions that are asked I/we are operating blindfolded and confined to guesswork.

It's still doing it when the case is open, but how hot is the heatsink getting (cool/warm/hot/untouchable)?

Direct a fan into the innards so it has bulk cooling air and see if that makes any difference.

gbono
Done a lot of experiments with heatsinks.  When depending on convection alone the air right next to the fins has to get quite hot before it moves out of the way to allow cooler air to have contact.  This explains why there is such a large difference between fins vertical and fins horizontal.  Even the smallest forced draft has a profound effect, and in fact the difference between a gentle breeze and a howling gale is less than from still air to a gentle breeze.  This leads to the rule of thumb with fans that 90% of the cooling effect come in the first 10% of the rev range.

http://www.ozvalveamps.org/techsite/thermofan/thermofan.htm