Hi,
I just got my Anan G2 Ultra and am running piHPSDR on it. Out of the box, it would only put out 30W with the drive slider at 100%. I found the PA settings and was able to get the power up but it very non linear. I watched the yt video that is mentioned a lot, but it is using Thetis and the PA screens are not the same.
What is the proper way to calibrate the PA in piHPSDR? Is there a way to make the power output linear?
I also have a problem with 17M where, if I adjust the output for 50W when the drive is set to 50%, I can only get about 70W max. If I adjust for 100W with the slider at 100%, I get about 90W when the slider is at 50%.
Finally, I did all my calibration into a 50ohm load, when I switched to an antenna ( same swr as the load ), power readings were completely different, almost double in some cases.
Any thoughts?
Thanks
Joel
Power calibration
Re: Power calibration
I just checked my G2 Saturn Ultra and it puts out a solid 100 watts on all bands into a dummy load. It was like this from the factory. Open the menu in PiHPSDR and select Transmit. Be sure that the checkbox Tune use Drive is selected. Also Tune drive level is set to 100.
I think you will find that the Tx drive level slider isn't very linear. The power indicator on the output meter does match my watt meter between the radio and my dummy load. But, the tool tip on the TX drive slider will not always match the radio's watt meter reading or my external watt meter reading.
There is a PA calibration menu that can be selected similar to Thetis.
There are instructions in the latest PiHPSDR manual on how to use the calibration feature.
James
WD5GWY
I think you will find that the Tx drive level slider isn't very linear. The power indicator on the output meter does match my watt meter between the radio and my dummy load. But, the tool tip on the TX drive slider will not always match the radio's watt meter reading or my external watt meter reading.
There is a PA calibration menu that can be selected similar to Thetis.
There are instructions in the latest PiHPSDR manual on how to use the calibration feature.
James
WD5GWY
Re: Power calibration
I have the G2 plain, i have noticed same thing, particular on 17 and 12. Others ok, but those two not at all linear. Using Thetis.
Richard - KI4P
Re: Power calibration
When I got mine, power was not set up out of the box, the limiting to 30W was not the TX settings, I did find those and set them.
Too bad the power settings don't have the same 10W steps that the meter settings have.
Too bad the power settings don't have the same 10W steps that the meter settings have.
Re: Power calibration
Power control in both Thetis and piHPSDR is very non-linear because of the non-linearities in the RF output amplifier chain. A 1dB non-linearity, which is typical, doesn't sound like a lot, but it does appear to be quite a significant difference when plotted on a linear (Watts) scale.
For example, 50W = 17dBW. 18dBW = 63W. 16dBW = 40W. Thus, from an RF engineering perspective it might be "merely" a 1dB error, but for hams used to working in Watts it appears to be a train wreck.
In more sophisticated (and expensive) radio designs, these non-linearities are either removed by hardware (an "equalizer" stage), or by software (using a calibration table).
Ideally, Apache should be characterizing and calibrating each serial number and storing that calibration data on the PA/Filter board such that it could be read by Thetis or piHPSDR and used accordingly. Sadly, that is not the case.
In Thetis, Richie has added the means by which someone may perform their own calibration measurements and manually add the necessary adjustments for linearity error into the Thetis database.
In piHPSDR, I don't think there is an equivalent function, other than the basic RF drive gain setting per band.
If you wish to use the full power output of the hardware, then you have no choice but to calibrate 100 on the drive slider to be 100W output. This is true even if you are using Thetis and are willing to go through the entire RF drive calibration process.
For those using amplifiers, another approach is to calibrate the RF drive control at the drive level you use most often with the amplifier. This is what I do. I normally drive my amp between 30 and 40W, so I've adjusted the drive gain settings on each band to give me precisely 35W when the drive control is set to 35. I really should go through the entire drive control linearity calibration process, but I've been too lazy to do it
For example, 50W = 17dBW. 18dBW = 63W. 16dBW = 40W. Thus, from an RF engineering perspective it might be "merely" a 1dB error, but for hams used to working in Watts it appears to be a train wreck.
In more sophisticated (and expensive) radio designs, these non-linearities are either removed by hardware (an "equalizer" stage), or by software (using a calibration table).
Ideally, Apache should be characterizing and calibrating each serial number and storing that calibration data on the PA/Filter board such that it could be read by Thetis or piHPSDR and used accordingly. Sadly, that is not the case.
In Thetis, Richie has added the means by which someone may perform their own calibration measurements and manually add the necessary adjustments for linearity error into the Thetis database.
In piHPSDR, I don't think there is an equivalent function, other than the basic RF drive gain setting per band.
If you wish to use the full power output of the hardware, then you have no choice but to calibrate 100 on the drive slider to be 100W output. This is true even if you are using Thetis and are willing to go through the entire RF drive calibration process.
For those using amplifiers, another approach is to calibrate the RF drive control at the drive level you use most often with the amplifier. This is what I do. I normally drive my amp between 30 and 40W, so I've adjusted the drive gain settings on each band to give me precisely 35W when the drive control is set to 35. I really should go through the entire drive control linearity calibration process, but I've been too lazy to do it
Re: Power calibration
My usage model is both. I would like to leave my amp in standby or just off and have the slider be accurate. When I use the amp, I need 0-25W to be accurate. I would happily spend the time to calibrate it if it was supported.
My next move is to try Thetis which does have correction factors. I bought the Anan version with the display to see what I thought of it stand alone.
I have nits with both, I will have to decide which one I want to spend time on. There are tweaks to the UI I want to do in Thetis beyond the standard supported items. Now I will also look at the source for piHPSDR and see if I can add correction factor to the PA like Thetis has.
My next move is to try Thetis which does have correction factors. I bought the Anan version with the display to see what I thought of it stand alone.
I have nits with both, I will have to decide which one I want to spend time on. There are tweaks to the UI I want to do in Thetis beyond the standard supported items. Now I will also look at the source for piHPSDR and see if I can add correction factor to the PA like Thetis has.
Re: Power calibration
I did try Thetis today and I found that using the offsets gave me very acceptable results. I also noticed that even without using offsets, the power control was much better, not sure what the two could be doing different, besides the offset feature.
I also noticed that the NR2 noise reduction worked better in Thetis, not sure why that would be either.
I also noticed that the NR2 noise reduction worked better in Thetis, not sure why that would be either.
Re: Power calibration
Noise reduction 2 in Thetis has some settings you can change that will improve how well it works. You may need to experiment with them a bit to find what works best in your location.
James
WD5GWY
James
WD5GWY
Re: Power calibration
Yes, there seems to be a lot of playing to do to get the most out of the radio. I knew this going in and is one of the reasons I got it.
Though, there really needs to be an offset feature like what's in Thetis for power calibration, makes all the difference between working well and not so good.
Though, there really needs to be an offset feature like what's in Thetis for power calibration, makes all the difference between working well and not so good.