HL2 low output power

[Cheuk-Yan Au]

Dear all,

Can you assist with the issue of having low Vbias, Vop and Vpa causing low output? What other components should i check? Which components are causing this low power issue?

Here are the outputs:
Band, Measured output(W), Reported on Thetis
10m, 0.5W, 1W
12m, 1W, 2W
15m, 0.5W, 1W
17m, 0.6W, 1W
20m, 0.5W, 1W
30m, 0.6W, 1W
40m, 0.6W, 1W
80m, 0.5W, 1W
160m, 0.7W, 1W

Test setup
Hermes lite 2 - HL2 with N2ADR filter board, removed AL block that binds it to case for heat dissipation 
SWR/PWR meter - Nissei RX-503 (PEP power)
Dummy load 50 Ohms, good to 200W (with points to measure voltage)
Digital Multimeter

Software: Thetis v2.10.3.11 x64 (10/19/25) HL2 (MI0BOT) on Windows 11.

Straight morse key testing using CWU at Drive=0db

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

1. Ensure your software is configured properly for transmit. All compatible software has similar configuration.

•     The PA must be enabled. Check
• Filters must be set to not attenuate your transmit signal. Check
• Software settings which may attenuate transmit per band must be adjusted to maximum. 
• The HL2 has a 7.5dB hardware gain. This should be adjusted to maximum.  Check

Audio to the software must be configured correctly and set to the proper sampling rate (48kHz) and levels. Some software such as Quisk will not transmit if no transmit audio device is specified.

2. Configure your software for CW and ground the tip to shield of the front KEY/PTT stereo connector on the HL2.

• Does the HL2 switch to TX as indicated by the TX LED illuminating? Yes
• Do you hear relays changing when engaging transmit? Yes
• Do you measure any power out on the ANT SMA connector (not the RF1 SMA connector) when terminated with a proper 50 Ohm load? Change in output voltage seen when TX (measured on dummy load)
• With the radio set in full duplex mode, do you see any RF signal (single spike) on the spectrum graph? No RF signal seen on spectrum graph on Thetis when TX-ing
• Do the temperature readings from the HL2 reported by software increase? What is the temperature range you see during TX? Yes 37->39C when TX held for ~2secs. 

3. If there is no power out, check for these physical problems.

• Do you have the N2ADR board properly connect to the main HL2 board with the small 20x2 jumper board? Yes
• Confirm that the center pin of the ANT SMA connector and your external coax make proper contact. Yes
• Check that you did not damage or short any components when installing the heat shim. In particular, J31 and J32 on the bottom of the HL2 and near the heat shim have been scraped off in the past and cause TX failure. Also, B106 and L34 on the top of the HL2 near the bolt and nut hole can be damaged.  No visible damage to these components

4. Check for expected voltages and bias current

Configure your software for SSB and engage TX via software or external PTT. Software should report bias current in the 180-220mA range. There is no need to set your bias current as this is configured at the factory and stored in nonvolatile memory on the HL2. This YouTube video shows how to use Quisk to read (and set, which you don't need to do) the PA bias current. Other software has similar functionality.

• With TX engaged, is the DC voltage measured at Vpa 8.0V? 6.17V
• With TX engaged, is the DC voltage measured at Vop 10.15V? 6.74V
• With TX engaged, is the DC voltage measured at Vbias 10.0?  6.96V

5. Disable the PA in software and then engage CW transmit again as described in step 2.

• Do you measure any output when a proper 50 OHm load is connected to RF1, the low power output? No change in output voltage
• With the radio set in full duplex mode, do you see any RF signal (single spike) on the spectrum graph? No spike seen in Thetis spectrum graph when TX
• If you have an oscilloscope, configure software to 80M and measure any RF output on DB3 pin 4 to ground and DB3 pin 3 to ground during CW transmit. What is the peak to peak voltage, and is it a clean sinusoid at the 80M frequency? DB3 is in the lower middle section of the HL2 between the AD9866 (chip which takes the smaller heat sink) and 6 blue inductors as seen in the HL2 with voltage measurement points labeled. picture. No oscilloscope, did not test

6. Check for an intermittent or poorly grounded AD9866.

• With TX engaged, apply finger pressure to the top of the AD9866 down towards the board. The AD9866 is U7 and takes the smaller heat sink as seen here. While applying pressure in various amounts and slightly different directions, does RF power out appear? No perceivable difference when pressed at different position and forces
• On the bottom of the HL2 board, there is a larger hole which exposes the thermal pad of the AD9866. The thermal pad is soldered to ground at the factory and you will find it hard to notice any solder if done properly. Still, there have been cases when improving this ground solves transmit problems. Try filling this hole with solder to provide a possibly better ground to the AD9866. It will take a bit of time for the hole and surrounding ground plane to heat up enough to melt and take solder. Tested continuity and it is connected to the ground place (against the copper edges)

Best regards.

Chuck 9V1AD

[Reid Campbell]

Hi,

Could you check the power supply voltage when you transmit? It is possible that your power supply is not giving the required 3A for transmit.

Cheers

Reid
Gi8TME/Mi0BOT

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/hermes-lite/ef7bfec7-98e9-4517-91b7-eae04f33c494n%40googlegroups.com.

[Cheuk-Yan Au]

Dear Reid,

OMG!!! THANK YOU VERY MUCH!!!

I think my power supply is kaput... It was good then for 100W on the FTDX10 which it blew my FTDX10 finals (the HL2 + amp is its replacement). I have a 12V battery plugged it in and voila it runs at around 5W (intended)....

Best regards,

Chuck 9V1AD