Page 1 of 1

PA Calibration

Posted: Thu Jun 22, 2017 3:58 pm
by Jim
Decided to perform the PA Calibration due to receiving a high current fault on one band......noticed it was putting out in excess of 320 watts.

Performed the PA Calibration per the 8000dle guide [/color]looking for 10 watts into a dummy load. It seemed pretty simple.
After the calibration I see FULL output between 46 and 220 watts 6-160.

Performed the PA Calibration at 34 watts (My amp driving power for all bands). After calibration setting power out to about 34 watts it pretty well agrees with my watt meter and drives my amp just fine on all bands. I now see FULL output between 100 and 145 watts 6-160. Works best for me
since I almost always run my amp. Only thing at 100-145 watts it's not a 200 watt radio.

OK, just set 10-160 to 190 - 200 watts. 6 meters tops out at 178 watts. Lowered it to 150. Now drive setting for my amp (34 watts) is all over the place. (By band)

Any suggestions?


Re: PA Calibration

Posted: Thu Jun 22, 2017 9:26 pm
by w-u-2-o
Updated/edited on 28 June 2018 to distinguish between how the 7000/8000 work vs. the earlier radios.


What you are running into is the inherent non-linearity in the design of the transmit RF chain. Unfortunately, this is something that appears to plague the entire line of Apache Labs radios. I have personally experienced this on a 100D, 200D, and now an 8000.

On all radios except the 7000 and 8000, when you move the Drive control on the PowerSDR mRX user interface, what you are controlling is the duty cycle of a PWM output from the FPGA. This PWM output is low pass filtered to become a DC voltage, and that voltage is applied as the DAC reference voltage. Any gain adjustments in Setup > PA Settings are applied to this control mechanism. I have been told that the linearity of this mechanism at the output of the DAC is near perfect.

On the 7000 and 8000, when you move the Drive control on the PowerSDR mRX user interface, you are controlling both a digitally controlled attenuator in the TX path, as well as the DAC reference voltage as described above. The attenuator provides gross control of drive in 1dB steps, the adjustments to the DAC reference voltage provide fine control in between those 1dB steps. Since the DAC output is maintained near its maximum value this tends to improve the performance of the TX RF chain in these radios compared to the older radios.

Sadly, the gain linearity of the analog RF amplifier stages after the DAC output is not that good. Although this could be corrected in software, such calibration or compensation functionality it is not currently implemented. Thus right now it is not possible to obtain a perfectly linear correspondence from the Drive control at the RF output. As a result, there are two basic calibration methods that people tend to use because of this:

a) Put the Drive control at 100 (maximum) and make adjustments in Setup > PA Settings to achieve proper maximum RF output power for the radio on each band into a dummy load. This will ensure that the full power of the radio is available, however the Drive Control value will generally not match up to output power in Watts except at the maximum power setting.

b) Put the Drive control at the most used/important power level, for example the power level that is normally used to drive an amplifier, and calibrate for that output power in Watts. This allows for an easier, more direct understanding of the output power of the radio into the amplifier, but often will not allow maximum power to be achieved. This is essentially what you did. This is also what I always do as I never run the radio without the amplifier.

If you do require the full output power capability of the radio, then procedure (a) is what you want, and you will have to make separate notes, on each band, of other Drive settings that result in the desire RF output power levels other than maximum.



Re: PA Calibration

Posted: Thu Jun 22, 2017 10:22 pm
by Jim
Scott, thanks. That's what we decided to do today, make the PA Cal at 100 vs 10 setting all bands were set each at 200 watts except 6 meters which I got to 188 but didn't what to keep it at it's max so we lowered it to 150
watts. The other bands had a lot of head room so we felt comfortable leaving them at 200.

Again thanks for your assistance.

73 de KE4WY Jim

Re: PA Calibration

Posted: Mon Dec 31, 2018 8:04 pm
by gw3vol
Many thanks for your posting.
Having only had my new 7000 for a few days I attempted to calibrate the PA output on all bands and using my Bird Thruline as per the manual. I found that power output was all over the place and in 10% tune mode when I set the output to 100W on all bands on the PA gain page the actual value varied from 9% (set at 10%) to 35%. Also, I cannot get more than 74 watts out on 10M whereas I can exceed the 100W limit on all the other bands (not a problem as I never use 10M)
I will now recalibrate with with your option ‘a’ solution.
I do have an amp that I use occasionaly (Acom 1000).
What I propose to do is set the drive slider to minimum and then gradually increase it for the required power output from my amp. I have Thrulines in circuit at the output of the 7000 and at the output of the amp.
Thanks for your particle, I thought that I was doing something wrong so you have answered the question.
Happy New Year.
Jeremy. GW3VOL. / SV0XBJ

Re: PA Calibration

Posted: Mon Dec 31, 2018 10:06 pm
by w-u-2-o

Re: PA Calibration

Posted: Wed Jan 02, 2019 2:52 pm
by Adam

I read with interest Jeremy's post regarding PA calibration. I received my 7000DLE just over 3 weeks ago, serial No 131, and have spent most of my free time since then reading the posts on the various forums and watching the YouTube examples. I would like to give thanks to all that have given so much time and experience for shining a light on the world of SDR radio, I doubt I would have got much beyond turning the radio on without such a resource.

My unit came loaded with version 2.4 firmware that caused some initial problems, following guidance here I reverted to 2.3 and all is well. I have however a similar low output on 10m as GW3VOL reported. I performed the PA calibration and all other bands can go well over 100W, measured on my LP-100A into a Bird 8401 load, I set 100W as a max at 100% drive on each band but 10m only makes 80W, Volts/Amps show 13.3/16.0, I wonder if this is common on 7000's or if there is something I've missed or should be looking for?

Happy New year to all

Adam - G0AUR

Re: PA Calibration

Posted: Wed Jan 02, 2019 5:37 pm
by w-u-2-o

There have, sadly, been reports over the years of various Apache radio models unable to make rated power on 10M and 6M. This goes back to the first 100W model.

You have, quite correctly, looked at the voltage and current, as many people have found the problem being a bad fuse, bad power cable, or a weak power supply (you need an honest 25A supply). 16A on 10M seems low. You may want to contact the factor and/or the US repair representative, W5WC.



Re: PA Calibration

Posted: Wed Jan 02, 2019 6:13 pm
by Adam
Hi Scott

Thanks for the quick reply, I had read about the non-linearity etc but not much on a resolution unless it was a serious failure. I'm using an Astron 50A linear supply and all the other bands draw around 20A.

I'm in the UK so I'll give Apache support and my supplier a chance to comment.


Adam - G0AUR