Out of curiosity, would you be able to measure the temperature of the mosfet itself? Maybe the center pin of it? That seems to be the limiting factor. Good work! I like your ductwork. That was a good idea.
Nice result. Smart design for the cooling circulation flow so the internal components get max cooling.
If you want some further experiment suggestions read on but otherwise ignore 🙂
1. Use some smoke (eg from an incense stick) to trace the air flow and see if hot air out of the fins is being sucked back into the fan intake. If it is a baffle on top for rear air intake might be an option that also shields the fan noise. A cardboard box taped on top would be a quick and easy mod.
2. It would also be easy and informative to cover one fan and measure the temperature increase. If the cooling limit is the volume air flow restriction out the bottom plate one fan might be enough.
Did anyone try to replace the refrigerant in old freon pipe heatsinks? What would be your knowledge about the best refrigerants used in this type of heatsinks or the best heatsinks using refrigerants?
What would be your best pick out of all these? I'm thinking of refilling some server heatsinks if they don't perform well enough with my new class b germanium amps as I wanna try the coolest setup possible
https://www.findthisbest.com/best-refrigerants
Unfortunately i always felt uneasy with picking up low noise fans...I don't know any that I can't hear in between two songs and lower speed fans efficiency needs some refrigerant help in my opinion although the good old natural convection through a 4 meter pipe looks bizarre enough for an all out audiophile approach.
https://www.findthisbest.com/best-refrigerants
Unfortunately i always felt uneasy with picking up low noise fans...I don't know any that I can't hear in between two songs and lower speed fans efficiency needs some refrigerant help in my opinion although the good old natural convection through a 4 meter pipe looks bizarre enough for an all out audiophile approach.
I’ve used cpu cooler/Noctua fan combo’s on a few of my class A amp builds with a single or pair of output devices with excellent results. But, in my opinion, I could not get a satisfying cosmetic look with this setup. But your experiment here looks great! This setup could be made aesthetically pleasing and have the capability dissipate some serious watts. Thanks for sharing
lhquam,
Nice work - the second approach you have taken by forcing the air up through the fins from beneath is an interesting one. As Vunce has mentioned, typical fan mounts don't always look the best, and it might be possible to achieve a more acceptable aesthetic by doing it this way.
Is there any audible fan noise and/or duct/port noise?
Nice work - the second approach you have taken by forcing the air up through the fins from beneath is an interesting one. As Vunce has mentioned, typical fan mounts don't always look the best, and it might be possible to achieve a more acceptable aesthetic by doing it this way.
Is there any audible fan noise and/or duct/port noise?
For my warmer amps, I use a couple fans blowing directly into the big sinks. Not exactly pretty, but very effective.
https://www.amazon.com/gp/product/B00JLV4BWC/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1
These are quiet and have rubber grommets on the corners.
I know they don’t need the extra cooling when there’s open air around the sinks, but I figure cooler temps equates to less stress on any components.
However, now that I think of it, my F5 was biased when everything was cooked in and stable after an hour or so with the sinks at ~ 50C… I guess I should forego the fans with the F5.
https://www.amazon.com/gp/product/B00JLV4BWC/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1
These are quiet and have rubber grommets on the corners.
I know they don’t need the extra cooling when there’s open air around the sinks, but I figure cooler temps equates to less stress on any components.
However, now that I think of it, my F5 was biased when everything was cooked in and stable after an hour or so with the sinks at ~ 50C… I guess I should forego the fans with the F5.
Attachments
1. I did a test directing the exhaust sideways and could not measure any difference in temperature.
2. I have not tested with one fan, but I have tested with the fan speeds reduced, which resulted in a temperature increase, due to the reduced volume of air per hour.
Attachments
Using the DYNACO Stereo 400 heatsinks with the fins facing down and the NOCTUA fans blowing into the tunnel makes for a SINGING BUSH which at best gets warm at the top (bottom in my case) of the fins.
Some real heat can be felt right next to the devices but not enough to not allow a finger to linger for as long as you are willing to hold it there Only right behind the devices is it HOT but I would think that is to be expected.
Using four amps I have to say four NOCTUAs are not silent even when slowed down but once the music plays they are inaudible. I sit about eight feet away from them,
I wonder if I will have to turn the fans up a bit for the Georgia summer? Might be a good time to build that fan controller ...
Some real heat can be felt right next to the devices but not enough to not allow a finger to linger for as long as you are willing to hold it there Only right behind the devices is it HOT but I would think that is to be expected.
Using four amps I have to say four NOCTUAs are not silent even when slowed down but once the music plays they are inaudible. I sit about eight feet away from them,
I wonder if I will have to turn the fans up a bit for the Georgia summer? Might be a good time to build that fan controller ...
I did some additional, more careful temperature measurements comparing upward exhaust vs. sideways exhaust and found that there was between 1 degree C and 2 degree C reduction in heatsink temperature with sideways exhaust. Not a lot or difference, but it does confirm a degree of mixing of the exhaust with the intake.
I have some really really old fan speed controllers that were intended for PCs. From what I remember they start at a higher voltage for a few secs, to make sure the fan starts turning, then they lower to the final voltage/speed.
Not sure if they still make what I have, but I did find this
Fan with speed controller
Not sure if they still make what I have, but I did find this
Fan with speed controller
I followed ihquam and did a similar experiment using my Pass M2 clone amp as the test platform. The chassis is a common size, Modushop 4U 300mm, this amp already has a PWM fan controller installed to power a small Noctua fan mounted on the Micro Audio SMPS.
I cut a piece of 3/16” plywood for the top cover, then cut a hole to mount a single Noctua NF-A20pwm fan. The fan is mounted to pull in air from the top. Baffles were made from cardboard to direct the air up through the heatsinks. The fan controller is adjusted to deliver 40% fan speed when max demand is called for.(Temperature sensor is mounted to SMPS transformer)
In stock form, the M2 heatsinks run 50°-52°C depending on ambient temperature.
After 2 hours this setup is running 40.5°C. Significant reduction in temperature!
A baking rack for a fan grill will keep my curious Labrador from getting her nose clipped
I cut a piece of 3/16” plywood for the top cover, then cut a hole to mount a single Noctua NF-A20pwm fan. The fan is mounted to pull in air from the top. Baffles were made from cardboard to direct the air up through the heatsinks. The fan controller is adjusted to deliver 40% fan speed when max demand is called for.(Temperature sensor is mounted to SMPS transformer)
In stock form, the M2 heatsinks run 50°-52°C depending on ambient temperature.
After 2 hours this setup is running 40.5°C. Significant reduction in temperature!
A baking rack for a fan grill will keep my curious Labrador from getting her nose clipped
Attachments
This fan moves a lot of air even at a slow speed. A great recipe for a silent fan. No, I can not hear it. An intake filter should be used for long term use.
Noctua NF-A20pwm fan is an excellent choice. With the larger fan and lower RPM the noise can be reduced for the same airflow. Why didn't I think of that before cutting my metal today for two 120mm fans?
I have this Noctua NF-P12 redux-900 in my THF-51S 50W follower amps - slow speed and very quiet:
https://www.amazon.ca/redux-900-Ult...82c5c&pd_rd_wg=OeOzR&pd_rd_i=B07C5KZX85&psc=1
lhquam, If you have your metal cut, this fan is 120mm.
https://www.amazon.ca/redux-900-Ult...82c5c&pd_rd_wg=OeOzR&pd_rd_i=B07C5KZX85&psc=1
lhquam, If you have your metal cut, this fan is 120mm.
I've been using Fractal Design dynamic X2 GP-14 fans for many years. They fit nicely inside my amp, don't make any noise and blow air across the transformer. With the lid on, the temperature equilibrium is achieved after 20 min or so. This keeps the inside temperature under control during hot summer days. I turn them off during winter. The fans run on an external little linear 12V power supply.
.... the sole reason being 4A of quiescent current.
.... the sole reason being 4A of quiescent current.
Last edited:
If you'd be using the right type of heatsinks no active cooling would be needed...Many high power amps in the 80's used the type you only get in computers now .They used Freon at the time, but I'm not an expert in refrigerants to tell if newer refrigerants are inferior orbetter at cooling.
https://quirkaudio.com/wp-content/uploads/2019/02/IMG_1255-768x1024.jpg
https://quirkaudio.com/wp-content/uploads/2019/02/IMG_1255-768x1024.jpg