0 for 2

[Danny Walters]

Hello

I have a couple of Hermes Lite boards that I need some assistance with.

The first, I purchased and assembled earlier this year for a friend.  All went together like clock work, and I spent the weekend making contacts all over, especially on 10m.  It was great fun, and worked a treat.  I had my friend over and showed him around it, made some contacts, packed it all up and sent him home.  It never made RF output again.  Had him bring it back.  Sure enough, dead as a mackerel. The receiver works fine.  I walked the circuit with a scope and I can find no trace of any RF output anywhere on the board.  I replaced the output transistors, with no positive effect, so I put the old ones back.  I'm relatively certain the that the filter board is intact and has no visible defect.

In an effort to get him up and going, I just ordered in a replacement Hermes board, figuring I'd troubleshoot the first one at my leisure, and maybe build up a second system.  That one went up in smoke the second I applied power.  U16 is damaged, and literally flamed out when I powered the system up.

Same power supply as used successfully by me on the first board.  Regulation and ripple are good, and has sufficient current available.  Will run the first board's receiver just fine, and will control the filter board correctly.

I contacted Makerfabs, who told me about this group.  I'd appreciate any assistance.  I'm $700 in on this system, and only have a nice metal case to show for it.  I'm more than a little baffled on both of these failures, and more than a little frustrated by it all.

[Steve Haynal]

Hi Danny,

Sorry to hear of your problems. Almost 2000 HL2 have been sold now, so I can assure you that you are just unlucky or there is some systemic problem in your setup.

Failures like this can be due to stripping off components when inserting the board and heatsink into the enclosure. People do not always orient the heatsink correctly.

https://github.com/softerhardware/Hermes-Lite2/wiki/Final-Assembly#heat-shim

Please remove the boards and provide pictures of the top and bottom near the PA LDMOS devices. For the board which receive still works, can you transmit into a dummy load briefly when the board is outside of the enclosure?

Was it U16 or U19 which smoked on one board? Please provide a picture of this.

Also, have you reviewed this wiki page?

https://github.com/softerhardware/Hermes-Lite2/wiki/No-Transmit

73,

Steve

kf7o

[Danny Walters]

Hi Steve

Thanks for your reply.

I completely believe that this is just a run of bad luck, although I concede that stuff like this drives me insane sometimes.

I was meticulous in placing the heat sink shim, and did it the same way on both boards.  I read previously that this is a common vulnerability.

First failed board PA section.  (Not my prettiest rework, but the replacement devices I tried looked a lot better than this)

I'll have to put this together an transmit into a dummy load a bit later today.  I tend to suspect it will be zero, but I'll gladly go through the exercise.

Here are photos of the other failed board.  It is U16 that went, immediately upon application of power.

I reviewed the no-transmit notes prior to seeking help.  

[Steve Haynal]

Hi Danny,

For the first board with no power out, if you disable the PA in software, set software for CW operations, and key via the external input, do you see power on the RF1 low power output? This bypasses the PA and can help identify where the problem is. I don't see any damaged or missing components around the PA.

For the blown U16, are you interested in attempting to repair this yourself (we can walk you through the steps) first, or are you seeking approval for a warranty exchange? When did you purchase the second board?

73,

Steve

kf7o

[Danny Walters]

Hi Steve-

Let me get it all reassembled and try to get something out of the RF1 output, and I'll report back.

Regarding the bad U16, I'd prefer not to attempt a repair.  Something killed it right off, even something like maybe a defective layer, and if I don't get it all, I'm pretty sure it will just blow again.  I'd rather just warranty it, and truthfully, I really only need one working board and would ultimately prefer to just return it for refund.  I purchased the second board March 9, 2026 to get my friend up and running while I figure out what to do with the first one, and I really don't need it.

[Steve Haynal]

Hi Danny,

The boards are powered on and tested before leaving Makerfabs, so I am surprised that something shorted the power rail immediately.

I will contact you via e-mail regarding a warranty exchange.

73,

Steve

kf7o

[Danny Walters]

Hi Steve

Received your email and replied.  Thanks for doing that.

Regarding the first board, with the PA turned off, there is indeed a tiny signal out of the RF1 connector, around -90dbm into 50Ω.

[Steve Haynal]

Hi Danny,

That is much lower than expected. How are you triggering TX? Via an externa CW key connected directly to the HL2 when the software is in TX? Can you report the Voltage peak to peak measured at pins 4 and 3 DB3 relative to ground with an oscilloscope?

The schematic is here:

https://raw.githubusercontent.com/softerhardware/Hermes-Lite2/master/hardware/hl/hermeslite.pdf

You can find more information on this measurement by searching for DB3 in this group:
https://groups.google.com/g/hermes-lite/search?q=db3

73,

Steve

kf7o

[Danny Walters]

Hi Steve

Apologies for the protracted delay.

Yes, keying from external connector.  

Signal at DB3 pins 3 & 4 is ~200mv p-p. (187-ish?)  

I also followed that through to U9 pins 3 & 5 and it appears more like ~800mv p-p.  I do not appear to have any meaningful signal at DB2 1 & 4 however.