Hi everyone,
I made the HF amplifier that you can see attached with the MRF101AN / BN that worked perfectly for a week.
One time, “just by applying the power supply” damaged an LDMOS.
Carrying out some measurements I discovered a short circuit between the gate and the source, measuring a resistance of 30 Ohm.
This has already happened using common power switching mosfets type IRF510 and others that were damaged “just by applying the power supply.”
The power supply is a switching.
Could someone give me an explanation and suggestions to prevent this from continuing to happen?
Thank you!
Gianpietro
I checked your schematics. It should be stable by design. However, the circuit layout and mechanical design are important to avoid unwanted parasitic feedback.
Make precautions during initial amplifier turn-on. You can discover abnormal behavior before it will damage the device. Read appnote from Polyfet "Precautions for RF Power Transistors":
http://www.polyfet.com/rftopics.pdf
The Most Common High Power RF Design Error: Inadequate Back-Side Grounding of the Source Contact Connection
https://mwexpert.typepad.com/freescale/2011/01/the-most-common-high-power-rf-design-error.html
> a short circuit between the gate and the source, measuring a resistance of 30 Ohm.
It is usual indication of Electric Over Stress or thermal run out. Transistor would fail if it's absolute maximum ratings are violated (junction temperature, gate and drain voltage, drain current).
- poor thermal management may cause fail due to high junction temperature
- overvoltage at the drain and gate pins
- overvoltage at the drain due to operation at high VSWR (no load or short). It doesn't cause direct electric damage but causes excess power dissipation and may cause thermal issues.
- parasitic oscillations occur at arbitrary frequency for which circuit is not designed. It may cause similar effects as high VSWR.
- strong voltage spike may cause latch-up of internal parasitic bipolar device. Typical reasons are spikes at the supply rail or ESD event during device operation. Duration is not important. Even a short ~nsec event may cause the problem. The key parameter is the current of the spike.
https://www.nxp.com/docs/en/white-paper/50VRFLDMOSWP.pdf
Have a great day,
Pavel
TIC
Hi and thank you very much for your reply.
Over time I learned about the problems you explained, I was wondering if some circuit precautions can protect the MOS such as:
feedback between drain and gate with 470 Ohm and 100 nF in series and also 10 kOhm between gate and ground to never leave the floating gate even for the short switching time of the relay.
I will decide whether to buy more, I have already burned two sigh ...
All my best
Gianpietro