Hi all,
I had access to a FLIR thermal imager, and couldn't resist testing various household items.
My F5 is still in "development" and only mounted to a heat sink. Unfortunately, the aluminum doesn't show up very well, since it is so reflective.
and a fellow audiophile...
I had access to a FLIR thermal imager, and couldn't resist testing various household items.
My F5 is still in "development" and only mounted to a heat sink. Unfortunately, the aluminum doesn't show up very well, since it is so reflective.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
and a fellow audiophile...
An externally hosted image should be here but it was not working when we last tested it.
Dat cat's got them voodoo eyes....!!
Too bad we can't see the F5 heatsinks with more definitive detail.........
Too bad we can't see the F5 heatsinks with more definitive detail.........
So, I would guess you were looking at an aluminum (natural) surface behind the board? It shows cold. Natural shinny metal surfaces do not reflect proper emissivity for an IR camera. You will need to put something like a spot of thermal silicone goop on the shinny metal and read the temp of the goop. Unless you have black anodizing.
It also looks better if you adjust your low and high Limits to something normal like 25 -85C. Room temp would be black and 85 would be white.
😀
It also looks better if you adjust your low and high Limits to something normal like 25 -85C. Room temp would be black and 85 would be white.
😀
Last edited:
Yes, as you can see the device is cunducting it's heat to the board via the leads. All around the 3 leads you see yellow(increased temp) but, the soldered, exposed, shinny, metal of the solder joints to the output device appear cold (Black). Not likely, but, unimportant here...
What I do see is a heat producing component to the left of, but just below the level of the tansistor, most likely under the board though? What's that a source R? Only 1?
This is a very good tool for optimising, if you have access to one. Also for deguging problems!
That is probably the 0.47 Ohm resistor, R11/R12. They get warm, but are sized appropriately. Both should be visible, but one might be running hotter.
I'll put some silicone goop at various places on the heat sink, and re-image it.
Just for all your information, the heat sink is a 10x12, with 2" fin. I was hoping to get both F5 channels on a heat sink of this size. Right now (for testing purposes) I have a single channel mounted in the center of the right half-side of the heat sink. I don't think it will work, and I'll just stick with one channel per 10x12", unless I got to some active cooling.
Hahaha.......
I have a planter outside my front door, one of the neighbours cats LIKED to relive its self in there, till i doused its 'lurking evil' with a pail or 'holy dishwater'.
......or did it become the first liquid cooled cat?
I have a planter outside my front door, one of the neighbours cats LIKED to relive its self in there, till i doused its 'lurking evil' with a pail or 'holy dishwater'.
......or did it become the first liquid cooled cat?
Good pictures and a good approach to determine the thermal transition resistance between the dies of both MOSFET and the heatsink. The value of this thermal resistor must be very large cause bad quality of the micas or silicon platters.
What happens, if you use separate heat sink devices for each power MOSFET ?
Perhaps you can make additional pics under this conditions (it is hard to believe for me, that the reflection character of your heatsink is solely responsible for this visual impression by the pictures from post #1).
Last edited:
The emissivity problem with "shinny metal" is common. Every material surface has a different IR emissivity rating. I beleive you would need to set an emissivity variable in the camera software to something like 25% to correct the aluminum temp error. That will however mess up all the other visible material temps in the image. Black anodizing the surface of the aluminium will give very close to correct readings. I used to use a thin layer of White Out typewriter correction fluid on metal sufaces to get normal readings. Now I use creamique thermal compound.
I think I'll start with a few thin blobs of silicone, at various points on the heatsink...
Good quality black electrician's insulating tape M3 or Nitto works very well for providing a realistic measurement point on shiny surfaces, set EM to .98 and you could get very close to a realistic values.
But please don't use cheap tape the results could be quite varied.
But please don't use cheap tape the results could be quite varied.
Just keep in mind there is an insulating property to anything you cover the device with. A good thermally conductive substnce would be the best. A thin layer might be a nebulous statement here. How thin is thin? It can be to thin. the white out or the ceramique I use can both be thin enough that the metal is still reflecting through. If your eye can still see the device, put another layer on.
A good indication is when the device creating the heat is hotter than the surrounding surfaces. I ussually use a thermal couple also to back up the IR set-up.
A good indication is when the device creating the heat is hotter than the surrounding surfaces. I ussually use a thermal couple also to back up the IR set-up.
The question in my previous post (#11) I must ask more exactly:
What happens, if you use separate heat sink devices for each power MOSFET without mica or silicon platter stuff for electrical isolation - i. e. only mosfet and heatsink and "Heat sink compound" ?
Buy PCB Compounds and Adhesives Heat sink compound,20 ml tube RS ERHS20T online from RS for next day delivery.
All the devices will run at different Ts and thus begin to operate with different Tc and Tj. Eventually you could have one device taking the brunt of the abuse and blowing up.
One could use a Uni-Posca pen for anti-reflection spots, cheap and available in arts and crafts stores.Uni POSCA Markers - Cult Pens.
With little practice you can get a very thin coating which can be removed with a gentle rub with a fine scourer
With little practice you can get a very thin coating which can be removed with a gentle rub with a fine scourer
- Status
- Not open for further replies.