Experimental Baofeng RD-5R version

VK3KYY
Posts: 7486
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental Baofeng RD-5R version

Post by VK3KYY » Sun Feb 07, 2021 7:56 pm

do1jml wrote:
Sun Feb 07, 2021 5:49 pm
G4EML wrote:
Sat Feb 06, 2021 10:28 pm
I would suggest looking at the output amplitude envelope with a diode detector or a fast scope. It sounds to me as if the RD5R is not controlling the output power fast enough and doesn't like the 30ms bursts. The bursts could be very distorted if the power rise time is slow.
How would the original firmware do it, then?
Probably don't turn off the finals

But this would require a lot of changes to the firmware if that is the solution, because the entire Tx system in controlled together.

Normally you don't leave the finals turned on then there is no driver signal, as the finals will amplify any noise on their input, including noise.

Not switching the antenna switch to Rx would not cause any spurious emissions, but any change would be a hack, and I don't have the necessary test equipment to look at the results of such hacks.

The last thing I need is people using this firmware on the RD5R causing interferance.

do1jml
Posts: 53
Joined: Fri Sep 11, 2020 10:56 am

Re: Experimental Baofeng RD-5R version

Post by do1jml » Sun Feb 07, 2021 8:42 pm

I can understand that. What kind of test equipment could be use to determine if there is interference? Would some kind of SDR do?

VK3KYY
Posts: 7486
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental Baofeng RD-5R version

Post by VK3KYY » Fri Feb 12, 2021 10:29 pm

do1jml wrote:
Sun Feb 07, 2021 8:42 pm
I can understand that. What kind of test equipment could be use to determine if there is interference? Would some kind of SDR do?
I've had a look at the signal power envelope using a 350Mhz oscilloscope, and the RD5R is transmitting the same envelope voltage on DMR as on FM.

If you have access to an oscilloscope, it doesn't need to be a super fast one, probably 100Mhz bandwidth would be enough.
Connect an antenna to the input of the oscilloscope.
Transmit on 2m, and adjust the timebase and input voltage sensitivity, and you can clearly see the DMR signal voltage etc.


The results you are seeing on those power meters is obviously dependant on the quality of the modulation, and the RD5R signal modulation quality / spectral purity is not good.

There may be ways to improve the DMR transmission on the RD5R and the legacy source code is available for download etc, so you could compile the firmware yourself and attempt to make changes to the way the firmware controls the hardware.

But you would need to be very careful, because if you make a mistake in the code and turn on both the PA and the Rx at the same time, you'll destroy your receiver section.

do1jml
Posts: 53
Joined: Fri Sep 11, 2020 10:56 am

Re: Experimental Baofeng RD-5R version

Post by do1jml » Sun Feb 14, 2021 10:23 am

I have access to a 100 MHz oscilloscope and I could indeed check the voltage levels. But the question was about interference and I don't really see how an oscilloscope would check that. Why do you say that the modulation is not good, how did you find out?

VK3KYY
Posts: 7486
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental Baofeng RD-5R version

Post by VK3KYY » Sun Feb 14, 2021 10:36 am

Unless you plan on building the firmware from source, there is no point in looking at the pulse modulation on an oscilloscope.

The problem is likely to be something to do with the PA activation / deactivation speed, and will require some fundamental changes to the Tx / Rx switchover, which is controlled by an Interrupt Service Routine triggered by the C6000 DMR chip.

Unless you know what you are doing, I would not recommend messing around with the code in this area, because if you get something wrong in the programming you could destroy the Rx in your radio, by enabling the Rx Pre-amp while the PA is enabled.

do1jml
Posts: 53
Joined: Fri Sep 11, 2020 10:56 am

Re: Experimental Baofeng RD-5R version

Post by do1jml » Sun Feb 14, 2021 6:14 pm

Roger, I am very grateful for your open firmware efforts. On my side as a user, I think I noticed a bug, so I am reporting it because I understood that this is what is wanted. I would appreciate if you could look into it, but I am not asking. If you tell me that you don't have the time now, I can wait. If you are telling me that you cannot fix it, I'll just go back to the manufacturer firmware, no hard feelings involved.

But if you tell me that the device modulation is not good or that the firmware could produce interference, I would like to understand what is going on. Or if you tell me that you need some measurements on a device I own, which may be behaving in a different manner than the device you tested the firmware on, I ask how I can help. I hope this will help to clear what appears to be the beginning of a misunderstanding.

VK3KYY
Posts: 7486
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental Baofeng RD-5R version

Post by VK3KYY » Sun Feb 14, 2021 8:31 pm

No worries

There are some more urgent problems I mut fux before I can investigate this problem.

I also need to make a special dummy load cable to connect the radio directly to my oscilloscope because using coupling via an antenna connected to the oscilloscope does not give clean results.

I do not own a spectrum analyser, apart from a TinySA, so I don't know if the spectral output with the OpenGD77 firmware is worse than the official firmware.
The spectral output is not clean even with the official firmware.

Probably the OpenGD77 is no worse for RFI, but the 4FSK modulation at the start of the DMR pulse can be improved, because the amplitude of the Tx power is not very stable at this time

do1jml
Posts: 53
Joined: Fri Sep 11, 2020 10:56 am

Re: Experimental Baofeng RD-5R version

Post by do1jml » Mon Feb 15, 2021 7:57 am

VK3KYY wrote:
Sun Feb 14, 2021 8:31 pm
I do not own a spectrum analyser, apart from a TinySA, so I don't know if the spectral output with the OpenGD77 firmware is worse than the official firmware.
I have a spectrum analyser. The output from the GD77 did not look much different that the output from other radios I tried. But if you tell me what you need, I could do the measurement for you.

VK3KYY
Posts: 7486
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Experimental Baofeng RD-5R version

Post by VK3KYY » Mon Feb 15, 2021 8:37 am

do1jml wrote:
Mon Feb 15, 2021 7:57 am
VK3KYY wrote:
Sun Feb 14, 2021 8:31 pm
I do not own a spectrum analyser, apart from a TinySA, so I don't know if the spectral output with the OpenGD77 firmware is worse than the official firmware.
I have a spectrum analyser. The output from the GD77 did not look much different that the output from other radios I tried. But if you tell me what you need, I could do the measurement for you.
You could try loading the official firmware and doing some benchmark tests, on FM and DMR.

DMR will be difficult to capture on the spectrum analyser unless you have a way to sync the scan with the data burst

do1jml
Posts: 53
Joined: Fri Sep 11, 2020 10:56 am

Re: Experimental Baofeng RD-5R version

Post by do1jml » Fri Feb 19, 2021 8:53 pm

I tried that today. As I am not quite sure what kind of measurement you need, I was a bit in the dark, but maybe the results will be interesting.

I had the Baofeng and the GD77 connected to a dummy load and transmitting with 1W. A quick test at 5W did not show much difference.A quick check with a nanovna shows the dummy load, cable, connectors, etc... to be reasonably adapted at 50Ω with a SWR of 1.1-1.2. Tests done on UHF.

First the FM tests.

The Baofeng and GD77 FM modulations are similar and contained with a 16 KHz band. This are the good news.

This is the GD77
fm_rad-20Ks.jpg
fm_rad-20Ks.jpg (31.43 KiB) Viewed 3267 times
This is the Baofeng
fm_baof-20ks.jpg
fm_baof-20ks.jpg (45.46 KiB) Viewed 3267 times
You will notice there are additional spikes in the Baofeng spectrum. Next message.

Post Reply