Are Mosfets designed to run hot? I'm waiting for some heat sinks to come in the mail before I put my TSE in its chassis. The Mosfets seem to get very hot to the touch. I just got an inferred thermometer, and measured the temps. After running for 45 min. the mosfet heat sink right behind the mosfet , the temp settled at 106 C. It didn't seem to get any hotter. My current regulators which I mounted on the bottom of the brd, are running at 78 C.
When the heat sinks come in I will package the amp up with the mosfets and their heat sink sticking out the top of the chassis. I was then going to attach a piece of aluminum heat sink to the mosfet heat sink.
Can you put to much heat sink on a mosfet and make it run cooler then it was designed to run at? I would think that the cooler it runs at, the better. Am I wrong in thinking that?
When the heat sinks come in I will package the amp up with the mosfets and their heat sink sticking out the top of the chassis. I was then going to attach a piece of aluminum heat sink to the mosfet heat sink.
Can you put to much heat sink on a mosfet and make it run cooler then it was designed to run at? I would think that the cooler it runs at, the better. Am I wrong in thinking that?
In general, parts are designed to run at 20 o 25C; 106C which you can see from the outside mans it's probably close to 150C inside, the damaging temperature, so cool them down now.
There is not such a thing as "too much heat sink".
There is not such a thing as "too much heat sink".
In Firstwatt amps we try and run the mosfets at 55C on the heatsink with around 65C max on the middle leg of the mosfet. Your mosfet is running right at the max for the mosfet I ordered for my build which is 150C. I would not run it without more heat sinks unless you have some extra mosfets for replacement handy.
I am a little confused. I have the SE board in hand and the mosfets are positioned in the center of the board where it would impossible to use a large heat sink. Are you talking about the U2 regulator IC and the D1 rectifier instead? They look similar other than the regulator IC has more legs.
I mounted the current regulators on the bottom of the board and the mosfets on the top of the board. This way they are not back to back, and there is room to attach more heat sinks.
This is the first I have heard about the mosfets themselves needing more heatsinking. I am sure George and other builders would have reported failed mosfets because of excess heat if it was a problem. They are running much hotter than customary for what I have heard the temperature of mosfets should be but I would not be concerned and just order an extra pair and enjoy. Do not blame me if they go out though. I am not going to do any more heatsinking on the mosfets with my build. I may order an extra pair though with my next Mouser order.
wdecho,
What output tubes are you going to be using. From what I have heard, the brd. was originally designed for 45 tube. If you build it for 300B's, they draw more current, and everything runs hotter.
What output tubes are you going to be using. From what I have heard, the brd. was originally designed for 45 tube. If you build it for 300B's, they draw more current, and everything runs hotter.
I have my Gold Lion PX300B already here. I am waiting on my PS transformer from Edcor. I am not against using a fan if necessary or installing another heat sink but as you have said if a heat sink is installed you must install the mosfets at the bottom of the board. I think George has tested the amp with 300B's sufficiently enough to proceed without extra heat sinks other than the ones specified. The one pictured in the Lexan case is a 845 without much ventilation. I do not specifically know how much difference the 845 vs 300B but I think you and I both will OK with the small aluminum heatsinks. It all really depends on how the mosfet is biased on hot it will run.
The mosfet I am using the NDF02N60ZH which is rated at max 150C which is within spec's for the temperature you have measured. The temp is higher than I like to see which will shorten the lifespan theoretically from all I have read but that is what it is rated for 150C anything over 150C would be an immediate concern but you should be OK. Just give the amp good ventilation.
The mosfet I am using the NDF02N60ZH which is rated at max 150C which is within spec's for the temperature you have measured. The temp is higher than I like to see which will shorten the lifespan theoretically from all I have read but that is what it is rated for 150C anything over 150C would be an immediate concern but you should be OK. Just give the amp good ventilation.
Last edited:
I wouldn't run any component at its ABSMAX spec on any parameter, including temperature. In particular for temperature, the reliability goes down dramatically when the temperature increases. I tend to keep heat sinks below 60 ºC, especially for external heat sinks, where you risk burning someone's fingers.
One thing, though: Did you put thermal grease between the MOSFET and the heat sink? If you didn't, I suggest getting a bit of Wakefield 120 or 126 thermal grease and giving the MOSFET a very light coat on the side that mates with the heat sink.
Tom
One thing, though: Did you put thermal grease between the MOSFET and the heat sink? If you didn't, I suggest getting a bit of Wakefield 120 or 126 thermal grease and giving the MOSFET a very light coat on the side that mates with the heat sink.
Tom
Tom,
No grease, but I have, I'm not sure if its Keratherm , but it's a TO220 insulator/thermal pad. I used them for years working for NASA. They replaced the mica insulators and grease. No mess.
The pads are indeed less messy. They're also less thermally conductive than the thermal grease. If you're using an isolated package anyway (TO220TF) I suggest using the grease rather than the thermal pad. The NDF02N60ZH MOSFET mentioned in Post #10 is in the 'TF package.
....or get the TO-220 (non-isolated) version of the device and use a thermal pad and shoulder washer.
Maybe your super-duper NASA pads are different that the Bergquist SilPads that we mere mortals can buy. 🙂
Tom
Unfortunately, even at NASA we have to buy parts from the same suppliers as everyone else. So our thermal pads are not super-duper. I just hate that grease. It seems to get all over everything, no matter how careful you are. I never seemed to have problems with using the pads, but I will give the grease some thought. Thank you.
I do not think it is so much the grease but mica that is the better transferor of heat. The grease acts to fill in the microscopic holes between the two substances. The pads do a good job from my understanding but mica is still superior. Wear gloves.
I spent some time stuffing my board today and when I got to the mosfets I read that an insulator is not needed. Sure enough there is no metal showing on the back of my mosfet. I put on a tiny amount of grease and installed it on the heatsink. If I were you I would ditch the insulator and bite my tongue and put some grease on it and see what the temperature is then before adding anymore heatsink material.
Well, I removed the thermal pad, and replaced it with thermal grease. Then after mounting my brd. to the chassis with the mosfets and heat sink sticking out the top, I added a 60mmx60mmx10mm heat sink to the back of the mosfet heat sink. Now I get a reading of 65C after the amps heated up. I'm happy now. That's a reading right off the body of the mosfet.
- Status
- Not open for further replies.