Leakage Dissipation in IRF2804

[electrosteam]

Hi all, my first post.

I have some battery controllers for a ride-on railway locomotive that were rated to 36 V, but operated for the last few years with batteries totaling 24 V.
Today I tried them with batteries at 36 V and found the IGBTs were getting hot, with a dissipation of about 10 W each (2 controllers for 2 motors, each controller has 2 devices in parallel).

The IRF2804 is rated at 40 V, 75 A, 330 W peak.
A DC check shows about 250 mV Vgs, which seems way lower than any of the published characteristic curves.
The CRO could not find any AC noise.

Is Vgs 250 mV too much ?
Could the device be simply avalanching ?

They were cheap kits, so counterfeit parts is a possibility.

Thanks in advance for any comments or advice,
John.

 

[electrosteam]

Sorry, the devices are obviously Power MosFETs, not IGBTs.
John.
(I couldn't find an edit function to correct my first post)

 

[electrosteam]

Still learning about how this forum works, I had selected the 'Solved' button before I had actually posted the following and lost it, so for the record my conclusions.

In circuit, the devices have a DC leakage current of 1 A at 36 V with 10 mV on the gate.
The datasheet says less than 250 microA at 40 V.
Under test, the devices are leaking 2 mA at 33 V.

The IRF2804 in my cheap kits are either faulty or counterfeit.
I will source some genuine higher-voltage equivalents to enable operation at 36 V.
John.