Build This MoFo!

Member
Joined 2014
Paid Member
Yup, probably a good idea.
Any mounting advice for a CPU cooler?
I was thinking of designing a custom PCB with a standard motherboard mount and use a standard intel back plate with the mosfet bent up flat against the PCB with cooler bolted on top?

Check out this post #1,368
Each channel is powered by 24V 5A Meanwell bricks
The mosfet is clamped to the cpu cooler copper pad using aluminum angle stock. The aluminum cpu cooler base was drilled for two M4 fasteners that secure the mosfet clamp.
The MoFo pcb has a couple holes drilled in strategic locations to line up with the cpu cooler mounting bosses for solid attachment so the stress of mounting was not solely on the mosfet legs.
CLC filter on separate psu feeding the PWM fan controller and fans.
Going down memory lane has me wanting to pair my new FE2022 preamp with Big MoFo 🤩.
 
I think I'm gonna have to get the old ms paint out to demonstrate the cooling technique I am thinking off (I am not a graphics guy) :rolleyes:. As for PWM and DC control of the cooling fan. I was thinking of a DC control as this would amount to a simple voltage divider linear emitter follower (12V fan about 3.5W max at 300mA with 24V supply), but I am somewhat familiar with working with micro controllers, PWM would be best and could also provide other features such as a delayed relay start at the same time, something like the 8 Pin ATtiny85 (plus +5 regulated) would be great (I like through hole). As for separate PSU for fan supply I have found in admittedly 2 or 3 experiments, as long as careful consideration is paid to ground and nice big caps are present, noise isn't much off a problem, especially the back though a drop down voltage component.
 
Last edited:
Good luck with the fan and all that. It's a valid approach and it's been used for a myriad of applications with considerable success.
But nothing beats a fully passive, massive heatsink.
When I was servicing satellite transceivers and RF transmitters in the field (long time ago) the most common problem was fan failure and consequently dead transceivers / transmitters.
Only because most clients didn't want to invest in a more reliable solution. Which was going to be cheaper in the long run.
 
Member
Joined 2014
Paid Member
Your point of long term reliability is valid and understandable, but home audio is a different type of use case. A Noctua fan will probably last 5 lifetimes if used to cool a home audio amplifier, barring no user installation errors.
My biggest issue with CPU cooler/fan use is aesthetics. Mad Max looks aren’t WAF friendly 🤣.
 
OK, I definitely don't want to shift the discussion to "active vs passive cooling pros and cons" one.
But home audio is an amp idling at 50 - 100mA per channel, class AB.
Here we have a Single Ended Class A Beast idling at 2A and dissipating 40 - 50W per a single Fet.
And yes, that fan can last 5 lifetimes and also can last only 5 minutes.
So everyone is free to decide what's the best way to deal with the issue.