One customer brought me a amp with blowing IRFP064N. I replaced all the mosfets and gate resistors, and after 3 or 4 days it come back to me. I went back and checked all colateral components and they all checked ok, and for the second time I replaced the power supply mosfets with gate resistors , and again it came back after a few days.
These IRFP064N were giving to me a while back, but after the third time I used these transistor to repair this amp, without good result, I purchased some new IRFP064N from Digi-Key, and repaired the amp again. Now it has been more than 20 days and the amp has not come back to me.
The old transistors and the new ones looks identical the same.
I was also giving 170 pc's IRFP4004 and 90 pc's FDA8440.
The question is, how can I power test this transistor to see if they are counterfeit?
A model circuit to test them will be very nice.
Thank you.
These IRFP064N were giving to me a while back, but after the third time I used these transistor to repair this amp, without good result, I purchased some new IRFP064N from Digi-Key, and repaired the amp again. Now it has been more than 20 days and the amp has not come back to me.
The old transistors and the new ones looks identical the same.
I was also giving 170 pc's IRFP4004 and 90 pc's FDA8440.
The question is, how can I power test this transistor to see if they are counterfeit?
A model circuit to test them will be very nice.
Thank you.
The following was posted for comparing transconductance of FETs for a linear mode (class AB, etc)
http://www.diyaudio.com/forums/car-audio/163487-audison-eats-output-n-mosfet-why-6.html#post4909620
For FETs used in switching circuits, I generally look at the capacitance from the gate to the other two terminals (shorted together during testing). If all are close from a batch, that's good enough for me. For amps that are designed at the margin, I may select the closest for a given bank.
The following two groups show the FETs from an unknown source and from a reputable distributor. The first are the FETs sent with the amp (2x21844s amp) and the second are from a reputable distributor. You can see that the capacitance varies greatly in the first group but have a much tighter range in the second group. I don't know how reliable this is but I want all parallel components pulling the same load. If the structure of the silicon that makes up the FET is so great that the capacitance varies greatly, I wouldn't trust them to be closely matched.
His (gate capacitance):
1: +3.25022e-09F +1.65711e-02 <<
2: +4.50478e-09F +1.65711e-02
3: +3.84091e-09F +1.65711e-02
4: +4.02078e-09F +1.65711e-02
5: +4.77534e-09F +1.65711e-02
6: +4.76398e-09F +1.65711e-02
7: +3.25788e-09F +1.65711e-02
8: +3.98894e-09F +1.65711e-02
9: +4.85121e-09F +1.65711e-02 <<
Mine:
1: +3.45768e-09F +1.65711e-02
2: +3.46288e-09F +1.65711e-02 <<
3: +3.43843e-09F +1.65711e-02
4: +3.42203e-09F +1.65711e-02
5: +3.43284e-09F +1.65711e-02
6: +3.35477e-09F +1.65711e-02 <<
7: +3.37657e-09F +1.65711e-02
8: +3.45016e-09F +1.65711e-02
9: +3.35701e-09F +1.65711e-02
10: +3.46231e-09F +1.65711e-02
http://www.diyaudio.com/forums/car-audio/163487-audison-eats-output-n-mosfet-why-6.html#post4909620
For FETs used in switching circuits, I generally look at the capacitance from the gate to the other two terminals (shorted together during testing). If all are close from a batch, that's good enough for me. For amps that are designed at the margin, I may select the closest for a given bank.
The following two groups show the FETs from an unknown source and from a reputable distributor. The first are the FETs sent with the amp (2x21844s amp) and the second are from a reputable distributor. You can see that the capacitance varies greatly in the first group but have a much tighter range in the second group. I don't know how reliable this is but I want all parallel components pulling the same load. If the structure of the silicon that makes up the FET is so great that the capacitance varies greatly, I wouldn't trust them to be closely matched.
His (gate capacitance):
1: +3.25022e-09F +1.65711e-02 <<
2: +4.50478e-09F +1.65711e-02
3: +3.84091e-09F +1.65711e-02
4: +4.02078e-09F +1.65711e-02
5: +4.77534e-09F +1.65711e-02
6: +4.76398e-09F +1.65711e-02
7: +3.25788e-09F +1.65711e-02
8: +3.98894e-09F +1.65711e-02
9: +4.85121e-09F +1.65711e-02 <<
Mine:
1: +3.45768e-09F +1.65711e-02
2: +3.46288e-09F +1.65711e-02 <<
3: +3.43843e-09F +1.65711e-02
4: +3.42203e-09F +1.65711e-02
5: +3.43284e-09F +1.65711e-02
6: +3.35477e-09F +1.65711e-02 <<
7: +3.37657e-09F +1.65711e-02
8: +3.45016e-09F +1.65711e-02
9: +3.35701e-09F +1.65711e-02
10: +3.46231e-09F +1.65711e-02
I'm sorry, but I'm confused about the procedure, can you please be more specific.
Thanks in advance
Thanks in advance
Is the idea to short out drain and source, and measure the capacitance between the shorted legs with reference to the gate using a DMM?
Do I have to apply any voltage? in the picture I see 12v and ground.
Are these 12 volts the DMM voltage?
Do I have to apply any voltage? in the picture I see 12v and ground.
Are these 12 volts the DMM voltage?
Last edited:
The first procedure (with the link) is not in any way related to what I do.
Yes (for my testing), short drain and source and measure capacitance from gate to the two shorted legs.
Yes (for my testing), short drain and source and measure capacitance from gate to the two shorted legs.
- Status
- Not open for further replies.