GK Backline 600, dead fan, installed new one, smoked it immediately

[Kenrod]

So, after selling a GK BL600 to a friend, he informs me that the fan isn't working. I search the web, find that fan failure is common, order and install new fan. I turn on amp, during start-up (brief "protect" mode) the fan cycles and shuts off as soon as status light goes green.
Knowing it's a temp controlled fan, I can only assume this might be normal behavior (I have very little experience with the operation of this amp).

i step away to grab my heat gun to warm up the insides to make the fan cycle and as I return with the heat gun (20 seconds or so) I see a litttle whisp of smoke come from the fan. I power off the amp immediately, lean over and smell the distinct odor of a burnt component and the motor is hot to the touch.

I test the supply to the fan and see it as 58 VDC. This is a 24VDC fan, won't that always let the smoke out?

What gives?


EDIT: added schematic 7/27/17 9:09 CST

[The Dude]

Schematic please?

[J M Fahey]

Same thing, but until it arrives: that 24V fan may very well be fed from a 58V rail, no basic problem as long as *it* does not drop more than 24, the rest being dropped in a pass transistor which apparently gives it more or less current based on heatsink temperature, plus quite probably some ceramic resistor in series.

If thermal circuit is broken it might apply full voltage (or, say, 40V DC which is still deadly) and kill it.
Don´t think pass transistor is shorted because in that case fan would produce a hurricane for a few seconds and die.

Ar you sure you mounted a 24V fan there?
If by mistake you used a 12V one it would die even under normal operation.

[Kenrod]

Here is a snapshot of the fan supply section.
The full schematic is on my computer at work. I can upload full schematic in the am

I ordered the fan from GK, a 24VDC direct replacement

[The Dude]

You didn't maybe accidentally reverse polarity when you installed the new fan?

[Kenrod]

You didn't maybe accidentally reverse polarity when you installed the new fan?

Not unless the oem connector was wrong. Kinda of a fits-one-way sort of thing

[J M Fahey]

Here is a snapshot of the fan [ATTACH=CONFIG]44245[/ATTACH]supply section.
The full schematic is on my computer at work. I can upload full schematic in the am
I ordered the fan from GK, a 24VDC direct replacement

Thanks.
I don´t like that circuit, it´s not fail safe by any means and not what others (me included) would use.
It´s fully dependent on all parts working well or it will NUKE the fan.

If Zener D18 opens (or any track/solder/connector in its path), nothing stops the fan from receiving up to 60V DC with unlimited current.

I would calculate the proper value and add a current limiting ceramic resistor (might be chassis mounted) in series with the fan itself.
It will NOT affect operation in the least, since Q28 is a current source, but guarantees that worst case the fan passes the safe current and no more.

Please post the fan specs, it must have a small label, stating "24VDC" and either "xx W" or "yy mA" , either is enough to calculate a resistor value which allows just max rated current (or say, 10% more, which is reasonable) , then we find which resistor drops (58-24)V at that current and how much it dissipates.

[Kenrod]

[Kenrod]

added schematic to original post.

[J M Fahey]

[ATTACH=CONFIG]44254[/ATTACH]
[ATTACH=CONFIG]44255[/ATTACH]

Thanks.
So they are very generic 24V fans, you can get them anywhere, no need to specifically go to GK.
58-24: you have to drop 34V
Current is 0.16A or 0.2A , pick one.
R would be 34/0.16=212 ohms or 34/0.2=170 ohms

So, say, 180 to 200 ohms would be fine there.

Will dissipate: 34*0.2=6.8W , IF Q28 tries to murder the fan for any reason , so a 10W one is fine; chassis mounted preferred not only for dissipation but for mechanical stability.

Q28 will also work cooler

This addition is nothing unusual, extra series power resistors are often added, either in fans, or in series with 78xx type regulators, again to eas their dissipation.

[Kenrod]

is it safe to assume something is wrong around/with D18? Last fan dead, new one nuked instantly.

I suppose I should be checking the operation of the zener, and add the limiting resistor in series anyway. no need to cook another fan. I'm out the cost of one fan already. I don't know about charging for 2 fans.

The GK fan is just as cheap as anyone else (kinda surprising). Glad to see I can easily source it elsewhere if needed.

[g1]

Adding the limiter resistor should save another new fan. Also, while troubleshooting you can sub a 120R to 150R resistor for the fan.

[J M Fahey]

Agree.
A working fan is equivalent to 24/0.2=120 ohm resistor as G1 said, 150 ohms is also fine, test circuit as is with said resistor, if you measure more than 24V across its terminals and/or it starts to smoke, well, now you know.

And yes, the Zener might be open, transistors might be lossy, measuring voltages will point to suspect.

[Enzo]

Is the amp for stage use? Then ditch the whole circuit, and just put 24v on the fan and let it rock. No one will hear it. Fan noise would only matter if you had mics around it in a studio.

[cjenrick]

agree with that ^^^

if noise is an issue, they make special low noise fans, maybe just get a 120 AC low noise fan and fuse it.

had a row of people soldering, each had a fan for fumes, drove me nuts, they start hetrodyning off each other, thus the internet search for a low noise fan forgot the brand and part number,

with the full time fan, you save blowing up your amp if one of those components fails, especially the temp sensor which can drift with age,

the low noise fan will keep the studio engineer off your back. because they can be OCD.