Thank you Andersonix. I have recently built a DIY CNC router and first used it to make this project. I did have some slight chatter issues at one point, but changed the spindle speed a little and that stopped it. I'll probably try to add in the PR spindle speed function - that sounds very useful.
Here's a video about spindle speed variation from Haas. Theirs is not a random speed change btw.
Spindle Speed Variation - Stop chatter on your CNC lathe - Haas Automation - YouTube
Spindle Speed Variation - Stop chatter on your CNC lathe - Haas Automation - YouTube
Agree on all points!
Some Noctua fans use silicone 'pins' as mounting, very good for decoupling.
Neat! Btw, fearless often mean not so wise. 😉
Theoretically, class A amps, when driven, should be cooler, never tried to measure this... Are you sensing the heatsink temperature or the air inside the chassis?
Theoretically, class A amps, when driven, should be cooler, never tried to measure this... Are you sensing the heatsink temperature or the air inside the chassis?
Using an IR thermometer on the heatsink plate, on the outside. I also have done this on the inside plate, but opening the cover lets some of the heat out, and I'm reading a couple of degrees less.
I realize that the way the fan is positioned, directly under the amp, is not optimal for cooling. This blows air on the perforated bottom plate, through the cooling slots on the top cover, and allows air exchange inside the amp as well. Even so, just a small amount of air movement makes a marked difference in the temperature. The fan rarely reaches the next to highest speed and only once I have seen it hit the highest speed (when we were having a real scorcher of a day here in the Los Angeles basin).
I realize that the way the fan is positioned, directly under the amp, is not optimal for cooling. This blows air on the perforated bottom plate, through the cooling slots on the top cover, and allows air exchange inside the amp as well. Even so, just a small amount of air movement makes a marked difference in the temperature. The fan rarely reaches the next to highest speed and only once I have seen it hit the highest speed (when we were having a real scorcher of a day here in the Los Angeles basin).
Have you tried pulling air for comparison?
I think the way you have it is quite good. The only thing else you could try is installing more than one fan and running them even slower.
I think the way you have it is quite good. The only thing else you could try is installing more than one fan and running them even slower.
Yes. This is the 4th or 5th iteration.
Originally I had a different top cover on the amplifier with a fan in the cover and pulling air through. It did a great job of cooling the inside of the amp. but the heat sinks were still blazing hot.
Next, I had two fans, one on each side of the amp blowing over the heatsinks. That did a great job of cooling the heatsinks, but it was an ugly looking solution having the fans stuck on the side and did nothing for the inside.
Thirdly, I had three fans on the board below the amp. one in the center, and one under each heat sink. That did a great job, and I was happy with it, but I'm an engineer and couldn't leave good enough alone.
Next I started playing with the software on the fan controller board. If you look at the photo of the board, it is v2.0. Version 1.0 had capabilities for 4 fans (even though I only ever used 3) and three temperature sensors. This was before I put in the pseudo-random speed function. I programmed the fans so that initially only the center fan came on and only when the heatsinks came up to a high enough temperature did the heatsink fans come on. What I found was the the outer fans rarely came on. Just having a single fan under the center provided enough airflow.
At that point, I designed a new board with a single fan control and single temp sensor, the experimented with the control software, eventually adding in the pseudo-random function.
Originally I had a different top cover on the amplifier with a fan in the cover and pulling air through. It did a great job of cooling the inside of the amp. but the heat sinks were still blazing hot.
Next, I had two fans, one on each side of the amp blowing over the heatsinks. That did a great job of cooling the heatsinks, but it was an ugly looking solution having the fans stuck on the side and did nothing for the inside.
Thirdly, I had three fans on the board below the amp. one in the center, and one under each heat sink. That did a great job, and I was happy with it, but I'm an engineer and couldn't leave good enough alone.
Next I started playing with the software on the fan controller board. If you look at the photo of the board, it is v2.0. Version 1.0 had capabilities for 4 fans (even though I only ever used 3) and three temperature sensors. This was before I put in the pseudo-random speed function. I programmed the fans so that initially only the center fan came on and only when the heatsinks came up to a high enough temperature did the heatsink fans come on. What I found was the the outer fans rarely came on. Just having a single fan under the center provided enough airflow.
At that point, I designed a new board with a single fan control and single temp sensor, the experimented with the control software, eventually adding in the pseudo-random function.
How about a 3x3 grid of fans (9 total) and have them all run extremely slow (at the same speed or maybe the centre one 30% faster)
It is cool what you have though
It is cool what you have though
Last edited:
These fans look like they might also be good
SILENT WINGS 3 silent Fans for your PC from be quiet!
No idea about these, but they have a magnetic bearing, which is a worthwhile mention.
https://www.corsair.com/ww/en/MLSeries
SILENT WINGS 3 silent Fans for your PC from be quiet!
No idea about these, but they have a magnetic bearing, which is a worthwhile mention.
https://www.corsair.com/ww/en/MLSeries
Last edited:
The Silent Wings fan looks good. It is a couple dB quieter than the one I'm using at very similar performance. Probably worth a look if you are starting from scratch. I have a pile of the others from all my experimenting, so I'd have to stick with what I have.
The Corsair is much higher performance, but a great deal louder. The bearing is the first thing to go on most fans, so these may stack up really well for long term durability. But my primary spec driver was noise, so for that reason, I'd pass on this one.
The Corsair is much higher performance, but a great deal louder. The bearing is the first thing to go on most fans, so these may stack up really well for long term durability. But my primary spec driver was noise, so for that reason, I'd pass on this one.