New Hermes Lite 2 seems deaf

[Jon AB8WU]

Hello everyone,

I'm trying to determine if my new Hermes Lite 2 (August 5 2023)  is deaf or I'm doing something wrong. I've been using quisk with an RS-HFIQ to decode WSPR using an Alpha FMJ portable antenna. It's been working well so I added the HL2 to quisk and I cannot find or decode decode a single transmission save from seeing WWV weakly every now and then. 

This trouble led me to a comparative survey of the two radios using the S9 Signal Generator from WB2JLR. I've set the Generator for S9 and connected it via an SMA patch cable to the ANT port on each radio under test. Below are the Band, Signal Strength, Observed Frequency, Expected Frequency, and Frequency error for several of the S9 Generator test points.

Hermes Lite 2 with N2ADR Filter board code version 73.2 Rf LNA set to 0dB:

10M -90.89dB 28636314 28636000 10ppm

20M -88.84dB 14318127 14318000 9ppm

40M -89.00dB 7159071 7159100 4ppm

60M -88.68dB 5369297 5369300

80M -88.94dB 3579531 3579500 9ppm

Hobby PCB RS-HFIQ and Behringer UCA222 USB sound card set to 0dB

10M -36.63dB 28636268 28636000 9ppm

20M -39.38dB 14318130 14318000 9ppm

40M -43.61dB 7159074 7159100 4ppm

60M -47.02dB 5369300 5369300 

80M -49.01dB 3579533 3579500 9ppm

Any insight to what I may be doing wrong or any suggestions for troubleshooting? Is there some sort of header pin I have neglected to jumper? It almost seems to me as if the antenna port isn't connected on the Hermes.

73,

Jon AB8WU

[James Ahlstrom]

You must increase the RF gain. Attach the signal generator and vary the RF gain control "RfLna". You should see the received signal increase and you can probably observe clipping at high gain.

Jim

N2DAR

[Jon AB8WU]

I've tried that. On 10M when I turn the LNA up to 48 dB the signal is only S8 (-44.78dB) no clipping.

[Jon AB8WU]

I seem to be actually getting results on WSPR now. I think I had an incorrect mode chosen in quisk for WSTJ-X. I've since changed it to DGT-IQ. Thanks for the help.

[James Ahlstrom]

It seems that at 0 dB gain the level is -89 dB and adding 48 dB gain raises it to -44 dB so it seems the LNA is working. Maybe the filter board is set incorrectly and is attenuating the signal. Check the filter board band selection bits on the radio Bands screen.

If that doesn't work, remove the filter board and feed the signal generator directly to the HL2. If that works, there is something wrong with the filter board.

Jim

N2ADR

[Jon AB8WU]

Thanks for the help so far but I'm still scratching my head.

Filter board removed and signal generator set for S9 level output on RF3 (header pin 6).

LNA at 0dB gives  -90.5dB signal.

LNA  begins clipping at 33dB gain which gives a signal of -56.8dB

LNA at 48dB gain produces -43.72dB (S8) signal.

Hermes was powered for two tests via a Motorola 12V supply and then off a 12.38V AGM and produced nearly identical results. I upgraded the code to version 73.3 and no change.

[Steve Haynal]

Hi Jon,

You will need to set the LNA between +12 and +24 dB gain for normal operation.

Since you have a signal generator and I assume a scope, can we trace out if there is any loss of signal in the RF path? Please see the schematic on www.hermeslite.com. Please report the following in voltage peak-peak for a 14MHz test signal:

1. Output of your signal generator when connected to the HL2 header pin RF1 and then RF3.

2. RF5 on schematic

3. RF4 on schematic

4. Pin 1 and 3 of T2, there are large pads on the PCB where these can be measured.

Also, there are cases where the TR relay fails and is stuck in TX and causes receiver deafness. Please try with the signal generator connected to RF1 and then RF3 on the header as identified here:

https://github.com/softerhardware/Hermes-Lite2/wiki/IO

73,

Steve

kf7o

[Jon AB8WU]

Steve,

Thanks for some more troubleshooting steps. Sadly I do not have a working scope around. I may be able to find one.

Are you suggesting in the last troubleshooting step, which would cover a stuck relay, to use RF1 as an alternate feed point and read signal strength in software or measure p-p voltage with a scope?

With the signal generator at S9 and connected to RF1 quisk reports a visible peak in the graph magnitude -89 dB with the LNA maxed out at 48dB. With the LNA to zero it's -122dB. 

Since RF1 is the low power Tx chain that rules our a stuck T/R relay right?

I plan to dismount the board and take a close look at the solder joints.

Jon

[ON8EI, John]

Any update Jon? 

[ON8EI, John]

I came back from holidays and noticed from 10m to 20m I had few signals and lots of noise. I mentioned this to a HL2 pioneer WA3O and we went through some settings but no change.

I have 2 HL2's so I switched to my standby HL2 and it sounded perfect.

Then I changed the the new filter board with the old HL2 and it was perfect also.

Then I tried the old HL2 with the old filter board and it was perfect too.

From what we conclude, the sma plug was not connected tightly enough.

All good at ON8EI and it's great to have a spare HL2.

Thanks to Mike WA3O for going through different possibilties, he has the real ham spirit.

    de ON8EI John.