Does a tube amp cooling fan make a difference? Yes! (quick study results)

[runmikeyrun]

Lots of debates online about the efficacy of a cooling fan in your tube amp. How much does it really help? Is it worth the time/money installing one? I decided to do a quick experiment using an IR temp gun.

I finished rehabbing this Peavey Mace earlier today. It's a "tubes below the chassis" design and with some amps (like my Ampeg V4) the radiant heat through the chassis gets so hot that components have melted and fallen off the PCB! Excess heat also shortens the life of the components. I didn't want to deal with the extra heat so I installed a 3" fan behind the front panel pushing air out of the amp and drawing cool air in from the back. I've done this on a few amps now but was never sure how effective it really was.

THE EXPERIMENT:

I checked the temps in two locations: the back panel of the chassis and black plastic tube base. I did not read the temp of the tube itself because the glass gives varying readings depending on the tube's interior component the laser is hitting. The plastic is a more accurate reading of the average temp.

I turned the amp on and turned the standby off. I let the amp idle in play mode with nothing plugged in for 10 minutes to heat up the chassis and components so one reading wouldn't be starting with a room temperature amp.

Then I unplugged the fan and let it idle for five more minutes. The chassis was 108F and the tube base was 192F.

I plugged the fan back in and let it idle for another five minutes. The chassis was 98F and the tube base was 121F.

I realize that the amp being played will increase the temp of the tubes and transformers. However, this would require more time to test than I had to invest (setting up a 1khz test tone to feed the amp) and this test demonstrates that cooling the amp does have a measurable temperature benefit. 192F to 121F is pretty drastic!

How I installed the fan: First, I had to do some math. Ugh. I have a 0.2 CFM fan and I needed to make sure the vent in the back would allow enough air flow. Too small of a vent will be hard on the fan and won't allow maximum airflow. I used some online calculators so I won't get into the math here, but my vent size is MUCH larger than necessary, even accounting for some flow restriction from the metal grating.

Getting 12vdc power for the fan: I tapped the 6vac heater voltage and ran it through a simple full wave rectifier (4 diodes and a 100uF smoothing capacitor) and usually that gives you about 5vdc. If you want to run your fan on that, you can. It will turn much slower, be almost silent, but moves way less air. Since i'm not concerned about fan noise in a live setting, I added a $5 adjustable buck converter to step the DC voltage up to 12v. If 12v makes the fan too loud, you can adjust the voltage down. Some amps have dc heaters and a rectifier isn't necessary.

Using a 12v brushless DC fan means you will not introduce any electrical interference, but you need the rectifier and buck converter. Using a 120vac fan can create extra induced noise through your speakers but doesn't require a rectifier/converter and is easy to tap right off the AC line inside the amp. Another option is a 5vdc fan, removing the need for the buck converter.

Anywho, that's what i got. As always, YMMV.

 

[Wasnex]

A fan can definitely make a difference. I have a boutique amp that has way too much resistance in the bias supply and grid network. Because of this, the output tubes go into thermal runaway after about 15 mins of operation. With the tubes in bias probes and a voltmeter on the bias supply, you can see the tube current increasing as the bias is pulled down. Eventually the bias collapses and tube current breaks over into full on. The final transition is rather sudden .

I did some experiments with my carpet blower and found the amp holds bias fine with a bit of air flow. I installed a couple of low speed fans blowing directly on the tubes. The amp has an aux AC outlet on the back, which is a convenient place to plug in a wall wart for the fans.

I think this is the fan kit I used: Coolerguys Dual 80mm Fan Cooling Kit

 

[agedhorse]

This is exactly many new amps that need to comply with safety regulations MUST include a fan, it's that effective.

I recommend leaving the internal circuitry alone and use a small plug-in wall wart type supply.

I also hope that you have a typo, .2CFM is virtually zero and will be ineffective.

Bunch converters lower the voltage. I think k you mean boost converter,

Brushless fans can be VERY electrically noisy, I don't know where you got that information. But so can boost/buck converters.

 

[BoogieZK]

IMHO only needed when the tubes are below the amp and PCB mounted.

What should be monitered here is the PCB (if there is any) and transformers, because tubes are meant to work at high temperatures.

Maybe a tube expert could barge in here but if you cool tubes down, the glass will struggle between the internal heating and the outside maybe resulting in an integrity failure in time.

Also maybe the air-flow should be removing the heat from inside the amp instead of blowing on tubes?

 

[stevensivak]

So why did so many amps put the tubes upside down and allow the heat to gently broil delicate electronic parts?

 

[LSM07]

So why did so many amps put the tubes upside down and allow the heat to gently broil delicate electronic parts?

To keep our beloved amp techs employed, of course

 

[goran]

One thing I've noticed on my Mesa 400+ that has a factory installed fan (that blows sideways) is that the air around tubes closer to the fan is much less hot the air around the tubes on the opposite side, which is expected. I was wondering would that difference in temperature mean that the "outer in the back" tubes will have shorter life-span?

And another thing is that 6L6 tubes un Mesa run much much hotter than 6550 tubes in Trace Elliot Quatra Valve.

 

[beans-on-toast]

Using common sense, if a tube amp seems to be operating too hot, blow an external fan into it. It works wonders. Don't put the back of the amp up against a wall where air circulation is poor.

If a tube amp is running hotter than normal, check that the cabinet impedance match is correct and check that the power tube bias is optimally set. Bias, plate voltage, cathode current, and output transformer impedance contribute to how much heat is generated. It depends on the design.


The capacitors can get too hot in some amps which is justification enough for a fan. Tubes are designed to withstand heat. For example, the maximum tube bulb temperature for a 6550A is 250 degrees C (482F).

 

[Jefenator]

There's some solid state gear out there that could use a little help sometimes, too.
I don't have my little clip-on fan anymore [that I got to keep my old fan-less Stewart World 1.2 power amp happy] but maybe I'll get another one & use it on my beloved tube amps during the summer.

 

[Wasnex]

One thing I've noticed on my Mesa 400+ that has a factory installed fan (that blows sideways) is that the air around tubes closer to the fan is much less hot the air around the tubes on the opposite side, which is expected. I was wondering would that difference in temperature mean that the "outer in the back" tubes will have shorter life-span?

And another thing is that 6L6 tubes un Mesa run much much hotter than 6550 tubes in Trace Elliot Quatra Valve.

You might want to have the bias checked in the Mesa as factory bias is fairly cold.

 

[agedhorse]

IMHO only needed when the tubes are below the amp and PCB mounted.

What should be monitered here is the PCB (if there is any) and transformers, because tubes are meant to work at high temperatures.

Maybe a tube expert could barge in here but if you cool tubes down, the glass will struggle between the internal heating and the outside maybe resulting in an integrity failure in time.

Also maybe the air-flow should be removing the heat from inside the amp instead of blowing on tubes?

There are other components that are heat sensitive as well, and if the chassis itself gets too hot it will not pass touch temperature compliance.

 

[goran]

You might want to have the bias checked in the Mesa as factory bias is fairly cold.

Doesn't "cold" mean that it's fairly low voltage? Or that directly translates to how hot the tubes will be? Just trying to learn something . Your Mesa 400 tubes aren't hot to the touch?

 

[Oldfartguy]

From my old times as an engineer at a huge AM Tx station... I remember valves need an even flow to avoid a hot spot inside the tube. I mean, if you install a fan directly toward the tube when is hot , you will create a hot spot on the other side because you are forming a hot turbulence behind, and shorten the valve life. That's why mostly high-power valves are encapsulated in a constant air flow tube to preserve tube life. Of course, in a bass amp, the power handle is way lower, but keep in mind the fan position and air flow pressure around the valve. Another point is that immediately after you turn off, the valve tends to do a higher temperature pick. On professional installation, the fan is adjusted slightly higher for a short time to avoid this problem. Just my 0.2 cents

 

[runmikeyrun]

This is exactly many new amps that need to comply with safety regulations MUST include a fan, it's that effective.

I recommend leaving the internal circuitry alone and use a small plug-in wall wart type supply.

I also hope that you have a typo, .2CFM is virtually zero and will be ineffective.

Bunch converters lower the voltage. I think k you mean boost converter,

Brushless fans can be VERY electrically noisy, I don't know where you got that information. But so can boost/buck converters.

lol yeah it was definitely a typo! I was thinking about the 0.2A current draw when typing the CFM. I think it's 25-30 CFM

 

[agedhorse]

From my old times as an engineer at a huge AM Tx station... I remember valves need an even flow to avoid a hot spot inside the tube. I mean, if you install a fan directly toward the tube when is hot , you will create a hot spot on the other side because you are forming a hot turbulence behind, and shorten the valve life. That's why mostly high-power valves are encapsulated in a constant air flow tube to preserve tube life. Of course, in a bass amp, the power handle is way lower, but keep in mind the fan position and air flow pressure around the valve. Another point is that immediately after you turn off, the valve tends to do a higher temperature pick. On professional installation, the fan is adjusted slightly higher for a short time to avoid this problem. Just my 0.2 cents

Transmitter tubes are considerably different than audio amplifier tubes, they operate at higher temperatures and use different materials.

Hot spots (within reason) are not an issue for audio tubes.

 

[Wasnex]

Doesn't "cold" mean that it's fairly low voltage? Or that directly translates to how hot the tubes will be? Just trying to learn something . Your Mesa 400 tubes aren't hot to the touch?

Cold bias means low current. I assume this is reflected in the tubes running cooler. However, output tubes will still be scalding hot. I certainly don't plan to put it to the touch test .

Pretty sure the bias spec for the Hexa is 30mA per tube. Plate voltage is probably around 690V.

I don't know the factory bias spec for the 400+, but the plate voltage is shown as 480V on the schematic.

 

[bassguitarist85]

I put these on my Classic 30 and VB-2, but they are powered from an external USB supply with a L-M-H speed controller using an IEC Male/Female adapter that splits off to a NEMA 5-15R for the USB supply. 120mm brushless fans. No noise electrically through the amp. Typically run on low speed and no issues with excessive fan noise.

Important to note, the fans PULL hot air out and away from the power tubes. They do NOT push cold air across the tubes, this can risk thermal shock I was told. Chassis temps are far more cooler so they do work very well.

 

[agedhorse]

I put these on my Classic 30 and VB-2, but they are powered from an external USB supply with a L-M-H speed controller using an IEC Male/Female adapter that splits off to a NEMA 5-15R for the USB supply. 120mm brushless fans. No noise electrically through the amp. Typically run on low speed and no issues with excessive fan noise.

Important to note, the fans PULL hot air out and away from the power tubes. They do NOT push cold air across the tubes, this can risk thermal shock I was told. Chassis temps are far more cooler so they do work very well.

View attachment 7299186View attachment 7299187

Except that the cool air enters on the other side of the tube, either direction causes one side to be warmer or cooler than the other side.

 

[Wasnex]

Important to note, the fans PULL hot air out and away from the power tubes. They do NOT push cold air across the tubes, this can risk thermal shock I was told.

I have several mass produced tube bass amps that disregard this advice.

 

[bassguitarist85]

Except that the cool air enters on the other side of the tube, either direction causes one side to be warmer or cooler than the other side.

I'm guessing this is unavoidable regardless of fan direction?

I have several mass produced tube bass amps that disregard this advice.

You should look up the various threads on the SVT for its fan direction, should it blow frontwards out of the amp, or intake and blow backwards towards the tube. It's highly debated.