Fan noise

[lowell]

Got a carvin R600 which when turned on you hear the fan rev up like a motorcycle and then it just sits there humming away while the fan is on. With fan uplugged it's quiet. There is 640mv of DC on speaker jack when fan is on and plugged in, however this DC goes away w/ fan unplugged out of circuit. I can't seem to figure out how the fan is connected and how its ac signal is injecting into the output.

[Enzo]

Wait. Is there some mv of DC on the output or is there SIGNAL on the output, which is AC? And does this offset go away with a load present?

ANyway,measure the voltage across the fan. The schematic shows the fan running on the 12v regulator VR3, and then when the amp gets hot, it turns on Q301, which shunts the 24v across the regulator. If you have 24v on the fan, it is stuck on high speed. That means either the VR3 is shorted or Q301 is shorted, or the control circuit has an issue. I am assuming the amp does not get instantly overhot at power up. Since the thermal sensor circuit also controls the speaker relay, if the relay turns the speakers on, then I must assume the thermal sensor part is OK. At least as far as A302.

[MarkusBass]

Lowell,

As Enzo has written, measure the voltage accross the fan. It should be -12V when you turn on the amp, and -24V when the amp gets hotter. If there is -24V, the problem is quite simple; either VR3 or Q301 is shorted, or the comparator A301.A has problems. The comparator can be easily verified by measuring voltages accross TP2 and TP3 (check this first).
Appart from this, you can test the fan with any 12V power supply (separately from the amp). If it makes the same noise, it is to be replaced (BTW, is there only one fan in the amp? I thought there are two - one for each power amp).
The issue with DC on the output is strange (please clarify whether this is DC or AC). The only thing that comes to my mind is that the fan has a separate ground (separate from the amp) and it is connected to the power amp ground only is one point - on the power supply board. Would it be possible that someone has modified (incorrectly) this connection?

Marek

[lowell]

I had my meter on DC setting when taking measurement. If the signal is AC, which it probably is, would the DC meter setting still read something?

The voltage across the fan is -12V. The DC voltage between TP2 and TP3 is only 1.4V.

[Enzo]

OK, so it is just a noisy fan. Put a drop of oil on the shaft bearing and see if it helps, otherwise, new fan.

If there is some small AC voltage on the output, it should not read on the DC meter. SO I am a little confused just what you measured when you wondered to us how the AC signal got on the output by taking a DC reading.

If the fan loads down some low voltage supply a hair, I suppose it is possible for that to offset the amp a hair. I would expect that to servo itself away, but maybe not. In any case, that would not be an AC signal from the fan. Seems unlikey though since the fan has its own -24v rail. Well, not its own, the relays run on it too, but it is not part of the signal circuits.

Aer you using a speaker load while taking these readings?

[lowell]

Enzo no speaker connected when measuring the DC on output. Lubed fan w/ oil, didn't help.

[lowell]

where can i get 24VDC 80x80x25mm fan? can't find one mouser or digikey.

[MarkusBass]

Have you checked farnell.com ? I can buy them in Poland, for example this one: http://www.tme.pl/dok/a22/kd2408ptb3_13(2)agn(d08016200g-00)0.pdf
It's only $8 but I assume this it to far from you :). The delivery would cost more.

Marek

[Enzo]

perhaps Carvin still has one?


Look at stock number 599-0046 at Allied, 9 bucks.

www.alliedelec.com

[lowell]

thanks!

[lowell]

well I put the new fan in, connected a speaker and the fan came on immediately and the problem was gone i.e. no loud motorcycle sound from fan. I cannot remember if I had mounted the fan in the amp yet at this point. I then plugged in and made sure the amp worked... it did. I put it back together and set it aside for a couple days. Now today I tried it and the motorcycle sound is back! Could it be just happening because of physical location reasons? Could something have happened to kill the new fan? I will try it again w/ the fan out of the amp and see if that makes a difference, I don't know maybe the fan motor is injecting noise into the outputs...??

[lowell]

i checked the output again just to clear things up. there is 24DCmv on the output without the fan and 184DCmv with the fan in and on. is this a problem? the fan still measures -12v across the connector so according to previous posts there is not a problem there.

[lowell]

if I were to bias this amp correctly what do I measure as I turn the servo pot/s? Also, could it be that the -12v measures correctly but something is still wrong w/ the ICs you mentioned? I just measured the mvDC on the speaker jack and turned the servo pot, I was able to bring the 184mvDC down to 0dc. The fan noise diminished greatly but is still there... is this how to bias it?

[MarkusBass]

Lowell,

If you look at the schematic, you will see that the voltage across R435 is amplified by A302 IC and later used as power supply for the fan. So there is direct relation between bias current of output transistors and the fan speed.
But I don't see any reason why the DC voltage on the output jack is dependand on the bias voltage. Anyway, I would just set up the bias correctly and later tacle with the fan. I don't know what is the correct voltage to be measured across R435 but looking at similar output stage in G-K you had there 5mV. So I would start with 5 mV and measure voltage on the fan. It should be 12 V and it should increase when you provide some signal to the amp. And I mean signal like 20V on the ouput. It's because the fan should get 24V only with almost maximum power. Of course it would be the best if you find a service manual for this amp - bias procedure should be described there in details.

Marek

[lowell]

Marek,
Ok I asked Carvin to send me the u-manual. In meantime what do you think I can check or test for the fan. As in previous posts I have swapped out for a brand new fan and voltage is -12vdc across fan test points. I played the amp for a while fairly loud and -12vdc did not change to -24vdc, I might not have been pushing it hard enough though. But being that there is a new fan should I test VR3, Q301, or similar?

[MarkusBass]

I think that it is important to understand how this part of the amp works. The voltage across R435 (available on T1 and T2 test points - there is a possibility to set up bias using these test points) is amplified by A302.A. Amplified signal is available on TP3 test point. Later it is compared (by A301.A working as a comparator) with the voltage on TP2 test point and if it is greater, positive voltage is provided through D311 diode to Q301 transistor'a base. Only in this case the fan gets 24V supply. I think that loud playing is not enough, you should provide a signal from a signal generator to see whether it works. Of course in this case you shouldn't use a speaker but rather some resisitor (>200W). Alternatively you could simulate this situation by providing to pin 3 of A301 a voltage that is greater than the voltage on TP2 but I would do it very carefully - possibly with a resitor like 47-100k. This should force Q301 to work. If it does, you don't need to check the transistor.

Marek

[lowell]

Marek,
I got 35vac on the speaker jack through a resistor and left it going for 10minuts. I monitored the TP2 and TP3. TP3 measured -12vdc and TP2 was -.5vdc. The fan voltage did not change from the -12vdc throughout this test... then I saw some heat or smoke... couldn't tell if it was smoke or not, hopefully just some compound or dust burning a bit... oops. Does this mean Q301 is bad? Or could A301 have a problem?

[MarkusBass]

Lowell,

I wouldn't stress the amp for such a long time. The voltage change on the test point should be immediate after you increase the input signal to the amp.
I still suggest that you try to understand how this part of the amp works. I must admit that I was wrong when I tried to explain this in my previous post; the first OpAmp is not just a simple amp but it is configured as integrator. Here: http://www.wisc-online.com/objects/index_tj.asp?objID=SSE5303 it is explained quite well how it works (press the Next button to go to the next screens).
I'm not sure whether Q301 is suspected. I would say rather that it does not get proper voltage from A302.A. The voltage on TP2 should be -5V (-15V divided by 20k and 10k resistors) and it should be always the same. But the voltage on TP3 should change when you increase input signal. Would it be possible that A302.A op amp does not get the signal from R435? This is not correctly shown on the schematic but I would say that the signal (AC) on R435 should be available on H9.1 and H9.2 and later on T1 and T2 (don't confuse it with TP1 and TP2). Is it possible that between H9 and T1/T2 there is some kind of a plug and it is disconnected (or there is no contact)?

Marek

[lowell]

Marek thanks for the link... will read on. However, the amp finally kicked into 24v mode on the fan and the noise went away! I then tried running the amp with VR3 disconnected and the amp works and there is no noise. I am going to replace VR3 and see if that's the problem. If I'm on the wrong track let me know. I will read that link info you sent me, thanks again.

[MarkusBass]

Without VR3 there is no -12V on the fan so it's no wonder that there is no noise. VR3 provides -12V to the fan and you can easily check it by measuring its output voltage. If it's -12V, it works correctly.

Marek

[lowell]

Marek... when I disconnected one lead on VR3 the fan still had -12v and there was no noise...??? don't know why or how. I'll replace it anyway and let you know what is happening after that. Thanks for the integrator link that helped me understand the circuit alot. I don't understand how these circuits (comparator and integator) would cause this noise though and only at the -12v and not the 24v.

[lowell]

so the replacement of the fan v-regulator took care of the fan noise... however when biasing the amp I'm not getting a reading for the bottom power amp (across H8). Carvin said to bias to 3-5ma across the ballast resistor. The upper power amp I was able to bias but the meter reads 0amps across the lower power amp ballast resistor. I tried adjusting the VBE multiplier pot with no difference. I did not want to fry anything so I shut it down. I tested the 2 150ohm resistors on the VBE mutliplier and they're fine. ??

[MarkusBass]

I suggest to measure voltages on Q206 (base, collector and emitter). Are you measuring the voltage directly on emitter resistors of the output transistors? And are you doing this using milivolt range of your DVM?

Marek

[lowell]

Marek,
I misread your post and replaced Q206. There is now bias current flowing, however it is rising rapidly after turning on the amp. It starts around 1ma across the emitter resistor and rises. I shut the amp off after it hits 5ma.

[lowell]

Ok so I was sitting over the amp on my bench stewing over why this thing was "running away" on me. I then thought about how the heatsink plays a role in the circuit - I did not have the heatsink attached to the power module. Once attached, the amp worked great. I slowly raised the bias until it both amps were running around 3ma. I walked a way for a minute and heard the mains fuse blow. When I ran back the entire power amp that I had just serviced was shot. All output xsistors measured shorted. At this point I had not placed any "bird poop" on that side of the board. Could running the amp for 5-10 minutes w/o this white goop have caused the power amp to overheat and blow? This is the only conclusion I can draw being that it was working ok until I turned up the bias after about 5 minutes of having the amp on. It seems extreme to me though considering that it was such a short period of time and I was not running any signal into it. I will replace everything again, luckily I ordered double everything in case this happened. Then I will retry w/ thermal doodie and heatsink attached.

[lowell]

Ok the heatsink compound made a big difference as the amp is running for a while now w/ no problems.

I replaced everything in the top power amp (see schematic) for a second time around, added thermal compound and mounted everything to heatsink. Solid state amps have so many interracting circuits it's hard for me to keep track of what's going on. I guess I'll get better at it the more I get my hands dirty. Tube amps are much easier and simple for me... but that won't stop me from learning more about ss as I'm very interested in that as well. The whole idea of thermal coupling and solid state biasing circuits i.e. servos/vbe multipliers has been a new thing for me. Do you ss experts find that most amps you see for guitar, bass etc that are solid state operate in similar class? I'm finding AB to be the most common that I've seen, however the GK100RB I worked on was in Class G I believe... called so I think because of the lower and higher supply rails.