A Mosfet has a built-in reverse diode that avalanches (breaks down like a zener diode) when its rated voltage is exceeded. It is rated to conduct a lot of current but if it is continuous then it causes the Mosfet to get extremely hot.
I think I understand. So the Avalanche energy is the maximum allowed energy (non-repetitive) the diode in the MOSFET can handle when its revered biased voltage rating is exceeded.
So I suppose the repetitive avalanche current would be the average avalanche current through a reverse biased diode when its reverse biased voltage breakdown is exceeded.
Thanks John.
When one person suffers from a delusion it is called insanity. When many people suffer from a delusion it is called religion.
In an effort to optimize MOSFET switching performance...
When MOSFETs are put in parallel should they be on the same heat sink or on separate heat sinks? As a understand it temperature has something to do with why MOSFETs can be put in parallel and BJTs cannot.
My guess would be that MOSFET conduction decreases with an NTC (Negative Temperature Coefficient) and BJTs have PTC. But this doesn't answer my question.
When one person suffers from a delusion it is called insanity. When many people suffer from a delusion it is called religion.
Edited by - wasssup1990 on Nov 05 2010 06:00:31 AM
The problem that I'm having is that I think if I have paralleled MOSFETs on one heatsink and at least one of those MOSFETs is dissipating more power than the others, it will cause the others to increase their RDS(on) thus reducing the efficiency of them.
However if I were to have the MOSFETs on separate heatsinks then RDS(on) for any MOSFET won't be dependent on the other MOSFET's temperature (or manufacturing tollerances, technically) that are in parallel.
Still on separate heatsinks: If one MOSFET out of four in parallel had the least RDS(on) (more current flowing through it than the others) then it's temperature would increase and so it's RDS(on) would increase to allow equal current sharing between all of it's other parallel MOSFETs. This optimizes the total current handling capabilities of the MOSFETs primarily. As an advantage I able to predict if a MOSFET is about to fail as I can measure the temperature of each MOSFET relative to the others. I cannot easily do this if they are all thermally connected (on one heatsink).
Am I missing something?
When one person suffers from a delusion it is called insanity. When many people suffer from a delusion it is called religion.
Edited by - wasssup1990 on Nov 06 2010 01:15:13 AM
A high quality amplifier or power supply manufacturer would test and sort all the Mosfets then build the product using the matched sets of Mosfets. But if you make only one product then you can't do that.
Hmm yeah that's a good idea but I can't do that since I have no need for a pile of MOSFETs. Well it's good to know some manufacturers are doing this.
I need to look into this more.
Having pulled apart an old UPS I noticed four separate heatsinks with three MOSFETs paralleled on each. I don't know if it was a conscious decision by the designers to separate the heatsinks or that it was more a space saving decision, probably the latter. However they did keep the paralleled MOSFETs on one heatsink per heatsink.
Yes I am acquiring knowledge for a project of mine but I won't post anything until it has come along (actually made something), and if I can be bothered.
Later.
When one person suffers from a delusion it is called insanity. When many people suffer from a delusion it is called religion.
Avalanche Energy
Posted - Nov 04 2010 : 06:32:33 AM
