Blown MOSFETs?

[Maurizio Brameri]

Hi,

first view at AN1819 board rev B after the disaster.

I will desolder then Q1 and Q2 to check them.

R10 and R12 are at 99% blown. Probably R7 too.

Which components do you suggest to check and substitute?

Best 73s,

Maurizio I2NOY

[AC1I]

When mine blew it was enough to replace R10 and R12 along with the MOSFETs. Wouldn't hurt to check the diodes though.

When I replaced the MOSFETs, I used high temperature Loctite on the screws (but others have used lock washers successfully).

It appears that the thermo-cycles eventually loosen the screws which results in the MOSFETs failing.

Patrick AC1I

[av8to...@yahoo.com]

Hello Maurizio:

If you are using these VRF2933’s you can substitute for VRF2944’s but this comes at a cost. At Mouser the VRF2933’s are running about $165.00 USD per device, and the VRF2944’s are running about $198.00. Both components are in stock. 

I am running the VRF2944’s, but I bought them before the price got so outrageous. They will produce more power, but the Hardrock-500 protective circuitry will not allow you to take full advantage of the upgraded devices. I bought them because of my need to run digital, thus I opted for the more robust devices. That being said, they were not this expensive at the time.

As for the resistors you can replace them with any same type and package component. There are many options, so no worries on those. I really hope Q1/Q2 are OK.

73

Rich, AJ3G

On Sunday, August 10, 2025 at 10:53:35 AM UTC-4 maurizio...@gmail.com wrote:

[jcve...@gmail.com]

The diodes are supposed to protect the rest of the circuitry if the MOSFETs fail. So far they seem to be working. All you should need is R10 and R12. Remember to set the bias pots fully CCW so the bias is completely off when you power it up the first time.

Also, take it slow. See how the amp takes 0.5W drive then slowly increase from there. you should see 40-50W output. If there is something wrong there won't be any damage.

73,

Jim WA2EUJ

[Maurizio Brameri]

Many thanks to everyone for suggestions.

I have ordered a couple of VRF2933 and I am waiting them.

I let you know if I will solve with new MOSFETs and resistors.

73,

Maurizio I2NOY

[Robert Cheek]

I had one of the very first Hardrock 500s and have used it without any issues, except for the PTT lag, for years now.  

After updating the HR 500 to the third party firmware I suddenly blew the MOSFETs.  It was operating under the same conditions is had been for years and the MOSFETs suddenly failed.  I have the replacement parts now and when repaired I am going back to the original firmware because I am now suspicious of the new firmware ability to protect the MOSFETs.  

I have had instances with the original firmware where the amp dropped out because of an SWR condition and the protection was very fast acting.  After installing the new firmware is when I had the MOSFET failure.  It seemed like second or more delay and the MOSFETs blew before the SWR protection kicked the amp off line.   It never went off line.  

Bob N4RFC

[Maurizio Brameri]

I am happy not be the only people with blown MOSFETs just after the flash of VE7VXO firmware.

Flashed back original v. 3.4b and in next days I will solder new MOSFETs just arrived from Mouser.

I am in touch with Joe VE7VXO for other cosmetic matters of his firmware and I will submit this issue of blown MOSFETs, as now we are in 2 people experimenting this failure.

73s,

Maurizio I2NOY

[Andrew Catanzaro]

I will say that as of yesterday I was communicating with Joe about these MOSFET failures.  Seems he has been using his own firmware for over three years now with no MOSFET failures.  There are no lines of code that would affect any operating parameters of the MOSFETs.  The advantages of his firmware are too great to abandon because of only an uncorroborated suspicion.  Being a CW addict myself, I was ready to sell my HR-500 because of the keying issues and his firmware eliminated the unneeded lines of timing routines that kept the amp from responding to PTT needs appropriately and immediately.  It’s also very useful now to have a power output meter that actually works.  Joe is as much  of an RF engineer as he is a computer programmer and I would not dismiss his work.  He’s convinced that the most likely culprit in these MOSFET failures has been inadequate torque in the mounting screws.  Since you are installing new MOSFETs, Maurizio, don’t forget to go back and periodically retorque the mounting screws even if you use lockwashers. 

 73 – Andy W9NJY

[Maurizio Brameri]

Hi Andy,

it was not my intention to put Joe's firmware under accuse.

I was only surprised by the coincidence and Bob confirmed I am not alone.

When I disassembled MOSFETs from PCB I found the 4 screws well tightened and I used a threadlocker liquid to prevent that screws lost tightening.

I was thinking by myself what was the cause and in my future tests I will consider deeply every parameter displayed on TRX and linear amplifier.

As I use the BT interface made by HobbyPCB to send TX command to linear amplifier and the band I am transmitting in, maybe some wrong decode of CAT commands sent to Hardrock-500 with firmware 6.xx could make some strange behavior on switching relays that could cause issues to final stage components.

I hope that Joe can make some tests and code examination to be quite sure the issue that me and Bob noted is only a strange coincidence.

Best 73s,

Maurizio I2NOY