fail of MasterGan4

[ZHT]

I find the wavefrom of SW terminal(the connection of the bridge) is wrong. The frequency is consistent with the input signal but the Amplitude is 3V3. But there's no 3V3 power supply. It should have been the VIN I offer. And it's exactly a square wave. What may cause this case?

[Peter BENSCH]

Welcome @ZHT, to the community!

Please forgive my lack of understanding, but the MASTERGAN4 has no terminal SW.

• Could it be that you mean a ready-made board?
• Is this board from STMicroelectronics, e.g. an evaluation board?

Regards
/Peter

[ZHT]

I'm sorry for the unclear description. Actually, I bought a chip MasterGan4TR and degisn the circuit with the reference of your evaluation board. Here's the specific circuit. The VIN input is 12V. I'm just confused about why it works abnormally like what I described before.

[Peter BENSCH]

Well, the MASTERGAN4 is a half-bridge driver which usually handles up to 600V, but you want to operate use it with VIN at only 12V, right?

• Please check whether VBO UVLO protection is already active.
• Please also check at GH, where the gate of the HS GaN is led out, whether the drive level is sufficiently high and reaches at least 5V with respect to VOUT.

Regards
/Peter

[arduino12]

Hi,

I also designed a full bridge using 2 MASTERGAN4 components,
I use them to switch an air-coil between 20kHz..200kHz using phase-correct PWM.
I monitor the coil's current and set the duty-cycle so the current is about 1A (software PID at 10Hz).

It all worked well at 50V VS, but since I changed it to 80V - the GaNs started to burn.. :(
I monitor the coil voltage using a differential probe - its about 160V..200V not near the "600V max".
I monitor the coil current using a current sense clamp probe - its about 0.8-1A RMS (depend on frequency..)

The circuit works for 1 hour - temperature about 30°C most the time-
But sometimes the magic smoke escape (from the high side pad area of the MASTERGAN4)-
I suspect a cross-conduction - but it shouldn't happen - the internal gate-drivers should protect this case..
Another failure is that the MASTERGAN4 just stops working - it is enabled and receive complementary signal from MCU-
but it's output is not switched.. - its internal bootstrap diode is fine - I see 5V on the boot capacitor (still 30°C).
I suspect a hardware flaw in its internal "over temperature / UVLO" protection...
Sometimes after half an hour it starts working again...

I don't understand why they keep burning "randomly"...
1. Is there a way to protect them?
2. I tried probing their gates but didn't help.. any more things to probe on a non-working MASTERGAN?
3. Is there a replacement component with built-in over-current protection?

Thanks and looking forward to help,
Arad :)