Fans on amps.

@ieales. NEVER BLOW AIR ON / INTO AN AMPLIFIER.  I never knew that ! I thought I was doing something good for the amp. While I was listening to music the other night for a few hours or so I went to feel my amp like I do on many occasions and like always it was so hot u could not touch it for more than a sec. U literally u could of fried an egg on top of if. So I placed a fan right next to the right side of the amp for about a half hour or so and it significantly cooled down the right side of the heat sinks so much that they were only warm now. I didn’t know if that was a good idea or not so I took the fan away and decided to ask you guys for any suggestions about that. I was even tempted to buy another fan and place it on the left side like I did to the right side to have even cooling from both sides of the Krell Fpb 600. Thanks a lot for your advice. 

I run a fan on an amp at such a low rpm and airflow rate, that if it’s power supply is unplugged, it stalls out. It’s a small 12V fan running at approx 4.1V DC. When I plug it in, I have to touch one of the vanes to start it by hand. That is what you call the ’just above stalling’ voltage for the 12VDC fan. Each will be different.

It’s ’just enough’ air flow to move the air away from the heatsinks, and no more. The running temp drops by a good 20-30 degrees.

You can also buy a 12V fan speed controller from a computer shop. One that is all passive, just resistors and a few switches. That would have the least level of interference in your audio equipment.

Do not, I repeat, do not buy a PWM electronic controller of any sort. Buy a small unpowered passive one that uses resistors and is meant to go ’in line’ (in series) with the DC power line for the fan, and uses simple switched resistors to vary the loading or current/voltage that the fan has access to.

The problem is that the passive/resistive/switched fan speed controllers have all died away for the more expensive and ’egotistically more stroking’ expensive electronic ones with switches, big panels, tons of parts, LCD displays, lights, alarms, knobs, etc. This kind of poorly implemented (pulse power supplies, PWM voltage control, and electronic added noise) device is fine for computers (where the entire box is almost pure electrical and RFI noise) but absolutely terrible for high end audio.

You’ll have to search out a simple passive switched fan speed controller, that will have only a few speeds available to you, via the switches, and that’s usually slow, medium, and fast. they are difficult to find as ’passive’ type speed controllers are not cool and complex, and are considered passe and cheap. There is a good amount of them out there, but their specs and build are mostly hidden away from eyes, due to human perceptions about quality.

This kind of simple device:
https://www.newegg.com/p/376-0020-00007

Which is fine, if the cooling considerations you reach for... does not have an electrical system that does not destroy the sonic qualities you are trying to hold on to.

There are some that have simple amplifier circuits and are not PWM modulated, but this will take some discernment on the part of the buyer, to understand which is which.

Wow, excellent information. I was mistaken, I had 105’s installed. Could you clarify your last sentence? Thanks!

Installing 105°C caps is usually only a cost issue unless one takes the time to evaluate all parameters.

Old electrolytic caps increase ESR and decrease in capacitance. Loss of bass indicates capacitor coupled [boohoo] stages.

A high pass filter is a cap in series with a resistor. Typical design would place the corner one or two magnitudes below the minimum frequency of interest. For 20Hz, corner range would be 2 to 0.2Hz. It also blocks DC [A/C coupled]. As caps age, capacitance decreases, raising the corner frequency. The caps were shot long before a loss of bass would be perceived. The ’boohoo’ is because everything is audible and electrolytic capacitors tend to smear the sound. All other things being equal, which they seldom are, direct coupled can sound better. Bottom line, NAD weighed the trade offs and made fine sounding equipment for reasonable prices.

I run a fan on an amp at such a low rpm and airflow rate, that if it’s power supply is unplugged, it stalls out. It’s a small 12V fan running at approx 4.1V DC. When I plug it in, I have to touch one of the vanes to start it by hand. That is what you call the ’just above stalling’ voltage for the 12VDC fan. Each will be different.

IMO, this is a bad idea. Line voltage varies all the time and unless one is using a well regulated supply, the fans will stall when the line voltage drops. Additionally, a fan may run when cool, but stall when it warms up.

You can also buy a 12V fan speed controller from a computer shop. One that is all passive, just resistors and a few switches. That would have the least level of interference in your audio equipment.

Not so. If running multiple fans, resistors and switches are a PITA. The adjustable supply is a small transformer and an adjustable voltage regulator. The transformer is several feet away and injects ZERO noise. I know, I measured. Most 12v fans will run within about 0.5v of one another. If one fan is overly fast, add a diode [1N914 or 1N4001] in series to drop the voltage slightly.

For added flexibility, a small control panel http://ielogical.com/assets/Audio/FanCtrl.jpg

I do it my way, you do it yours.

My particular 12VDC fan with the 4.1vdc power supply has been running for over 12 years. 24/7.

Yes, is a fine example, not an outlier. But the power supply and fan were decided upon --together. Tested. Those numbers will not fit all situations.

I’m using it in a situation where no fan is actually needed, not in a situation where a fan is required to not fail. That would be like ..uhm..a closed cabinet with no visual access to the fans to check on them--combined with an amplifier that is likely to die if left in a hotbox like a sealed cabinet. In that scenario the fan is ultra critical. Not a good idea to make the life of the amp dependent on a $5-$20 fan and given power supply for said fan...

As for the rest, we can do it ourselves and arrive anywhere we want or with as much engineering rigor we may desire.

Most folks (95% plus, as a guess) want pre-fab solutions, so that is the reasoning behind indicating commercially available devices and critiquing commercially available devices.

85°C caps rated for 2000h in an amp running @ 35°C [95°F] have an rated life of 8000h.

Oops. 2^5 = 32. 2000h rated are 64,000h @ 35°C. 32k days @ 2h/d ≈ 88y.

However other factors limit the life other than temperature, so 15y is a safe limit.