Output power control Thetis v2.9.0.6 HL2

[Ivan Medvedev]

Hi!

I am a new HL2 user. Everything works fine, all that's left is to get HL2 to automatically control my XPA125B :)

Question about output power control. I am using Thetis v2.9.0.6 HL2. Slider Drive reduces power from 0 to -7.5dB or from 5 to 0.9W. It doesn't work correctly. At the setting -0.5, -1.5, -2.0, -4.0 and -5.0dB the power does not decrease. More details in the picture. What is wrong?

73, Jonas, LY1O

[Larry W8ER]

Jonas ... I also am driving an amplifier and ran into the same issue. I am using Thetis 2.9.0.8. The attenuator in the FPGA chip that controls the transmitter power has 7 steps, not the smooth variable control that you would expect. With an Anan radio the Drive control is smooth but it's different radio, different chip, same software with different result. You may be able to use the PA setting tab to achieve finer control. Find one of the plateaus that's close to the power that you need, and go to the PA tab and reset the gain value to achieve the exact power output that you require.    73 -- Larry W8ER

[DL1YCF]

> Am 08.06.2023 um 20:57 schrieb Larry W8ER <larr...@gmail.com>:
>
> Jonas ... I also am driving an amplifier and ran into the same issue. I am using Thetis 2.9.0.8. The attenuator in the FPGA chip that controls the transmitter power has 7 steps, not the smooth variable control that you would expect. With an Anan radio the Drive control is smooth but it's different radio, different chip, same software with different result. You may be able to use the PA setting tab to achieve finer control. Find one of the plateaus

The PA settings just help to calculate the „step“ from the drive slider.
BTW there are 15 steps.

> that's close to the power that you need, and go to the PA tab and reset the gain value to achieve the exact power output that you require. 73 -- Larry W8ER
>
> On Thursday, June 8, 2023 at 11:14:43 AM UTC-4 me2...@gmail.com wrote:
> Hi!
> I am a new HL2 user. Everything works fine, all that's left is to get HL2 to automatically control my XPA125B :)
> Question about output power control. I am using Thetis v2.9.0.6 HL2. Slider Drive reduces power from 0 to -7.5dB or from 5 to 0.9W. It doesn't work correctly. At the setting -0.5, -1.5, -2.0, -4.0 and -5.0dB the power does not decrease. More details in the picture. What is wrong?
> 73, Jonas, LY1O
>

> --
> 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 on the web visit https://groups.google.com/d/msgid/hermes-lite/3b17b1c6-d83f-44e2-b46a-5181ed08f52fn%40googlegroups.com.

[Ivan Medvedev]

Thanks Larry, but that doesn't seem right. The AD9866 Data Sheet says that there should be a step of 0.5dB. There should be 15 power levels.

Jonas

["Christoph v. Wüllen"]

To the whole list, especially SDR developers:

Of course it is possible to use TX IQ down-scaling, and then you will get a very smooth behaviour
and can regulate the output power smoothly down to zero.
And of course, you can compensate for this by up-scaling the TX DAC feedback samples and
"save" the PURESIGNAL algorithm.

However, the TX feedback level is already very low for the HL2 (0.233), and having it even
smaller (and then up-scaling it) may raise issues with ADC/DAC resolution. Is it worth to
try the combination of

TX IQ sample downscaling / TX DAC feedback sample upscaling

to get smooth TX output power level adjustment AND a working predistortion? In the suggested
setup, namely having a very small output power that drives a very big PA, the resolution of
the TX feedback signal may be a problem.

Yours, Christoph DL1YCF.

[Ivan Medvedev]

Thanks Christoph, but I'm not interested in other gain control methods. 

Let me clarify the question. As we found out earlier that the AD9866 can control the TX gain from -7.5 to 0 dB in steps of 0.5 dB. This is in 15 steps of 0.5 dB. In Thetis version v2.9.0.6 HL is only 10 step. More detailed picture in my previous post.

Is it just me? Maybe other apps are gain control differently? Maybe I have a hardware problem?

73! Jonas LY1O

[DL1YCF]

I have not *measured“ if I have step 15, since the PA is already over-driven then.
And I do not have Windows, therefore I cannot speak for Thetis, but with
piHPSDR and the PA calibration value set to the minimum value (38.8 dB), then
when moving the drive slider from 0 to 100 the „TX drive“ byte in the protocol
changes from 0 to 240 in steps of 16. Increasing the drive slider one watt at
a time, I see all 16 possible value. Only the upper 4 bit form the data for
the modem chip, and this varies from 0 to 15, which translates to -7.5 to 0 dB.

I could make a measurement at the „TX low power output“ and see whether the
*modem chip* produces the full scale. But first I would do the following:

in my github account github/dl1ycf/pihpsdr, there is a „hpsdrsim“ program
which simulates a radio. Invoking it with „hpsdrsim -hermeslite2“ it
shows itself as a HL2. An the program documents on the screen all settings
and their changes. You could equally well use wireshark etc., but then you
need a clever program to extract the relevant information from the data stream.
This would check if Thetis does the job correctly.

Yours, DL1YCF.


> Am 09.06.2023 um 21:28 schrieb Jonas <me2...@gmail.com>:
>
> Christoph, do you have step 15? I checked at PA Gain 38.8 and 40.3 the result is similar.

[Steve Haynal]

Hi Jonas,

To eliminate any questions about the Thetis implementation, can you try the same measurement with other software, for example Quisk, SparkSDR or SDR Console? Also, please disable the HL2 PA and report the low power output at RF1 not as dB but as voltage peak to peak into a 50 Ohm dummy load. The output steps may not each be exactly 0.5 dB but you should see change in voltage peak to peak measurement per step. 

I am away from my lab right now but will check and report measurements later.

73,

Steve

kf7o

[Ivan Medvedev]

Hi Steve, thanks a lot for the advice. Successfully run HL2 with SDR Console v3.3. The result is good. Drive is controlled from 0 to 100 and in the HL2 output I have:
0 - 0.034W
10 - 0.056W
20 - 0.095W
30 - 0.161W
40 - 0.270W
50 - 0.455W
60 - 0.75W
70 - 1.33W
80 - 2.2W
90 - 3.9W
100 - 6W.
I measured with SMORF from microHam.
No problem with an additional amplifier.
Unfortunately, this app does not support PureSignal :(
73, LY1O

[Marcos X]

I wanted to ask you a question, for example... in version 2.8.11 with drive 100 ADC1 Overload did not appear, with 2.9.0.6 it appears. My question is, does something happen to the hardware, is it bad for the ADC? or absolutely nothing happens

thankyou

[Reid Campbell]

Hi all,

I can confirm that Thetis 2.9.0.8 HL2-beta1 is not producing linear stepping. I'm currently working on integrating the IO board, so this will be fixed in that release.

Please make sure that you are running the latest release of the Thetis software and it has been modified to run with HL2. There are several release of Thetis which are mainly for the Anan range of transceivers. You can get the most up to date version at:

https://github.com/mi0bot/OpenHPSDR-Thetis/releases

and the current one is v2.9.0.8-beta1

The upshot of this is that the output will be able to be lowered more as the first 3 steps of the attenuator were being missed. There were also some missed steps as well.

Thanks to those that found the fault.

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 on the web visit https://groups.google.com/d/msgid/hermes-lite/1809123d-9161-41d6-81af-23a2c25c1c76n%40googlegroups.com.

[John Williams]

Reid,

 

May I also ask that you test the low power output as well after you apply your fixes? It also seems to not be quite right. Probably the same issue. I see sometimes that it will go to full output regardless of the slider. A restart of Thetis will usually correct it.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/df47f4cc-bb4a-3953-d1a6-988f405d7f7a%40gmail.com.

 

[Ivan Medvedev]

Many thanks Reid for the detailed explanation. We are patiently waiting for a fix.
73 Jonas  LY1O