Excessive heat and venting

[hesamadman]

On the Hoffman AC30 power amp, ive noticed theres a lot more heat than my 2204 circuit. I have my 30 watt installed in a head cabinet but it is a top mount. So if I put a vent on the top, it leaves an open area for dust to get directly into the electronics. I've noticed that the chassis and switches (fender old style) are even warm. The back of the head cabinet is completely open but no top vent. I went ahead and cut a vent in the top using a marshall style vent. I have not had a chance to put the amp back in the head cabinet to see if it helped though. I also have a similar amp using the ac30 power amp in a combo setting. It is set up a lot like a blues jr. as it has a back plate that comes off. I dont see a single vent in the blues jr. Just wondering if anyone had any suggestions

[sluckey]

Cathode biased amps usually run hot, especially 4 x EL84s. This is compounded by top-mounting so the tubes point down. I'd figure out a way to put a small 12VDC fan in the cab and run it from the filament winding with a FWB and a 470 to 1000µF filter cap. This will produce about 8-9v and will run the small motor a little slower, but quieter. Leave the back open.

[PRR]

> vent on the top, it leaves an open area for dust to get directly into

I don't fret about dust.

I do worry about beer. (Coffee, Gatoraide, string cleaner, whatever.)

Tube amps often run HOT with little harm.

> I dont see a single vent in the blues jr.

Commercial amps have to meet safety code. A new commercial product has to be very inaccessible to keep the user out. And traditionally warranties are short. The maker doesn't care if it cooks its guts out as long as it lives 91 days. I'm stunned to see that a Blues Jr III has a 5 year warranty!

[dude]

I've used Sluckey suggestion of a small 12vdc fan running off the filament supply, usually on a VVR (used a heat sink and small old PC processor fan) in an 18 watt combo with great results. With the VVR usually between the PT and OT in these amps, it's closed in and gets hot.

Maybe the bigger 12vdc PC power fan would work well for you, set up pulling the heat out, not blowing on the hot spots. Depending on how you bridge the rectifier diodes, you can get anywhere from 5 to 9vdc. Just right and not too noisy.

al

[hesamadman]

Thanks PRR. That makes me feel better about the vent I already cut in one of them.


I think I will try the fan method though. id like to find one small enough so it wouldnt take up too much space.

[Mike_J]

Thanks PRR. That makes me feel better about the vent I already cut in one of them.


I think I will try the fan method though. id like to find one small enough so it wouldnt take up too much space.

I would be surprised if the vent accomplishes a lot.  The only component inside the chassis that I could see generating much heat is the power resistor tied to the power tube cathodes.  I would be surprised if it generates anywhere near as much heat as the power tubes do and they are hanging below the chassis and would get marginal benefit from a vent on the other side of the chassis.  I have used the same Marshall vents in an amp head where the tubes face up.  In that case there are four EL34s and not much room in the head cabinet so there is much logic in using vents to help the heat escape thus the use of vents. 

In Fig 2-27 of Kevin O'Connor's book the Ultimate Tone, he shows how to run a low-noise fan from the 6.3 VDC filament supply.  First you run the filament supply to a 3 A 50 volt bridge rectifier to turn the AC into DC.  Then you run the output from the rectifier through a 470 uF at 10 volts electrolytic cap to smooth the DC.  From there you run the power to a 12 VDC Silent Series fan.  I would use Mouser P/N 978-9S0812L402.  It is a 12 VDC fan which moves 23.3 CFM at 16 dB.  That would change the air in the cabinet about 10 times per minute which is more than is necessary.  However, using the bridge rectifier and cap as described above results in about 8 VDC which would slow the fan and make it even quieter.

From what I recall it is best to run the fan exhaust out of the cabinet and not directly onto the tubes.  I would find male and female wire connectors to allow for disconnecting the fan from the chassis.  I would also create tie downs for the wire so they can't come loose and move around inside the chassis.

Thanks
Mike

[eleventeen]

Power supply from a dead PC yields:


1: small 12 VDC fan and mounting screws for same.
2: bridge rectifier
3: IEC chassis AC entry connector and countersunk mounting screws for same. 


Hacksaw the power supply box and you have a nice piece of sheet metal bent into a right angle with a cutout for the fan and (surprise!) four mounting holes already drilled. 


Free.

[dude]

No need to buy a fan, old PC processor fans are small and all over the place in any old PC.

Wiring with and without CT for filaments:

[alerich]

I crammed a Marshall 2204 into the cabinet of a Park G10 practice amp combo. The chassis is top mounted and since all the tubes and transformers hang upside down and  are right on top of one another it gets HOT real fast. Hotter than I am comfortable with. I bought a 12VDC computer fan that is about 4" square and about a half inch deep. I mounted it with small L brackets so that it blows sideways across the bottom of the cabinet circulating air throughout. I power it with a 9VDC wall wart that is velcroed to the boom of the cabinet and picks up AC at the PT primary so it is only running at about 75%. It works great and is pretty quiet. The combo is now really a head since there was no room for the speaker.

No matter how long the amp is on I can reach back there and touch the power tubes without getting burned. It keeps everything really cool by circulating just a little air that does not really blow directly o anything. In the end it probably would not have hurt anything to leave it the way it was but when the chassis started getting too hot to touch I said "enough".

[PRR]

> find one small

"Small" fans are usually a mistake. Size, airflow, and NOISE are related. You want a certain airflow. (We don't know how much, but more than gnat-blow.) Unless you only play VERY LOUD, you do not want noise. This leads to an "over-size" fan worked below design voltage.

"Oversize" because fan specifiers look to the air-flow first. (They won't be living with their product.) So fan makers rev-up the fans for big airflow, damn the noise (within limits). So for many living-with, working-with, and recording-with purposes, we take a "12V" fan and run it on 9V or 6V.

I had a net-hub made a hell of a racket. Meant to go in a hot server closet. But I had it in my office. Two 12V fans parallel on 12V. I re-rigged them series, they got 6V each. Were much quieter and the cooling was ample for office-temp air.

For the last 6 years I have had a Pentium PC, stock Intel fan cooler. Pretty darn loud. I'm re-purposing it, and the noise finally got to me. I've discovered it never gets ANYwhere near design max temperature. I pulled the fan and it didn't hot-up frighteningly fast.

I just left it run for hours with 90 Ohms in series with the fan. That gives 6.1V, about half RPM, and MUCH less noise. It shows 14C rise, and 63C rise is allowable (22C room, 85C max inside CPU). I just bumped the resistor to 135 Ohms (I had a 4-pack of 270r) which makes it *silent* (to my ear) and will see how toasty it gets. I suspect <20C rise, assume 30C in summer, 50C is still far below the CPU's limit. I wish I had done this 6 years ago. (But I still had Intel's warranty then.)

This is a nice 3.5" fan and probably a 65 Watt CPU. This is not out of sight of the heat in a AC30 (perhaps 90 Watts), and tube-stuff can certainly live long and prosper at max CPU temperature. So I'd be thinking of 3" fan, perhaps 2" if in a real tight spot and I was not sleeping in the same room. And of course, run the fan on lower voltage. If you only have "rated voltages (as in my PC), a handful of 270 or 330 is a good trimmer resistor. As noted, 6.3VAC rectified on a cap is 7V or 8V, which is a good starting zone for less-loud.

This good CPU cooler fan drew 0.12 Amps at 11.9VDC. Seems to be 0.068A at ~~6.1VDC. For hasty guesswork, pretend it is 100 Ohms and use divider-law. Small 12V fans will be higher "resistance". 5V fans will be lower "resistance". Teeny 5V fans may get back near 100 Ohms, but for any useful airflow a teeny-tiny fan is more noise than you want.

[dude]

I crammed a Marshall 2204 into the cabinet of a Park G10 practice amp combo. The chassis is top mounted and since all the tubes and transformers hang upside down and  are right on top of one another it gets HOT real fast. Hotter than I am comfortable with. I bought a 12VDC computer fan that is about 4" square and about a half inch deep. I mounted it with small L brackets so that it blows sideways across the bottom of the cabinet circulating air throughout. I power it with a 9VDC wall wart that is velcroed to the boom of the cabinet and picks up AC at the PT primary so it is only running at about 75%. It works great and is pretty quiet.

Great idea, no fooling with a fan rectifier, no current draw from the PT just tap the a/c and a Wall Wart, if voltage is too much there are plenty that give less.

As far as the small fan, that was used just for a VVR. Heat sink on bottom of chassis over the VVR and a small 12vdc processor fan strapped to the heat sink running at 6 vdc. Can't hear the fan, it trickles just a tad of circulating air but it's pulling off the heat sink, just enough to keep the VVR cool in my 18 watt Marshall combo.

A bigger fan running on low dc, would be better for cooling a hot amp with the tubes hanging down.

al

[hesamadman]

So for many living-with, working-with, and recording-with purposes, we take a "12V" fan and run it on 9V or 6V.

Im going to go this rout. Wont be much more work at all. I am going to find some kind of means of quick disconnect. I dont mind to do the crimp on male and females. I'll just make sure I have wiring secure inside.


I guess the main thing to consider would be amperage. Ive been looking at 12v fans that range from 130-240 ma

[eleventeen]

"I've been looking at 12v fans that range from 130-240 ma"

Mine is .19A = 190 mils.

I think you'd actually have to look pretty hard for a wall wart that could NOT handle that. None of the ones in my junk box are under 500 ma. Even my teeny cel phone one (though only 5 volts) is 700 ma.

[hesamadman]

I'm actually hoping to do this without using an additional power source.

[eleventeen]

That would be my choice. It's no big deal to pull another 200 mils from your PT. Just be sure to insulate and construct whatever means you choose to take the heater voltage outside the chassis carefully.

[hesamadman]

That would be my choice. It's no big deal to pull another 200 mils from your PT. Just be sure to insulate and construct whatever means you choose to take the heater voltage outside the chassis carefully.

Man I wish I could get flexible conduit like I use in homes small enough in diameter for this reason.

[PRR]

Heater power should not be terribly dangerous. About like the power out of a wall-wart, and they often use extremely cheezy cable.

You can sometimes cut-up a dud outdoor round extension cable, use as-is, or pull the core out and shove in new smaller core. I used outdoor cord on a recent fusebox monitoring display. The power was maybe half-Watt, but some chance of breakdown to 240V, it was "permanent", and it was at the main entrance to my cellar. I didn't want anybody, at any time (maybe me when I am old and wobbly), getting hurt.

[shooter]

fusebox monitoring display

I always wire at least 1 room light to each breaker, poor mans monitor

[PRR]

My Intel-stock CPU fan ran at 2,700 RPM at full 12V.

With 135 Ohms in series, it is a bit under 6V, runs at 1,110 RPM, and is MUCH quieter. The CPU wanders 35C to 46C which is far below the 85C limit of this CPU. I soldered it in.

FWIW: I tried 270 Ohms. The fan would run once started, but would not start on its own. That would be below 4V, on a "12V" fan, which is probably where you do not to go near.

Also FWIW: this old Pentium 3 may be the 87 Watt series. Which is sure in the area of a AC30's heat output.