High BER on Hotspots

Post Reply
VK3KYY
Posts: 7481
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

High BER on Hotspots

Post by VK3KYY » Sun Jan 02, 2022 11:46 pm

Anyone using V2 hardware will experience problems with hotspots on any power level except High power

Colin G4EML informed me that this is because of a hardware design fault in these older radios, where the PA could not turn on fast enough and the beginning of each DMR pulse was not being transmitted correctly.

There does not appear to be an easy way for people to fix the hardware in these old radios, and there is also probably not a complete solution to this by changing the firmware

For simplex hotspots, we could add a setting, so that the radio would leave the PA turned on for both TS 1 and TS 2, effectively making the radio transmit like a Tier 1 DMR radio.

But we don't think we can change the PA drive timing so that it was turned on early enough to not cause problems, becuase this would possibly be transmitting over the end of the previous TS

VK3KYY
Posts: 7481
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: High BER on Hotspots

Post by VK3KYY » Mon Jan 03, 2022 10:51 pm

Colin has an idea to set the Tx power to maximum for the first millisecond of each TS pulse, then change back to the actual power level,

This may be a partial fix for the problem with the old hardware which has the slow PA response.

Actually we may need to do something more complicated like a sawtooth waveform, with the leading edge of the PA drive being a vertical increase to absolute maximum PA drive on the old radios, but its definitely worth investigating.

I1KDO
Posts: 27
Joined: Sat Nov 16, 2019 7:49 am

Re: High BER on Hotspots

Post by I1KDO » Sat Jan 22, 2022 4:18 pm

From the tests done in these days with the power set 100 mW at 433 Mhz and HS Open GD77 I have never had difficulty the error is 0.2 BER on Pi but always perfect.
Probably there are differences in the radios I have 3 GD 77 that do not behave in the same way in the latest FW releases maybe different calibration or other ....

73

Lorenzo

G4EML
Posts: 919
Joined: Sat Nov 16, 2019 10:01 am

Re: High BER on Hotspots

Post by G4EML » Sat Jan 22, 2022 4:24 pm

The problem with Hotspot access was fixed by correcting an error in the frequency calibration.
There remains the issue with the PA ramp up time, but this will normally only affect use with distant repeaters, Local hotspot access should now be OK.

Colin.

I1KDO
Posts: 27
Joined: Sat Nov 16, 2019 7:49 am

Re: High BER on Hotspots

Post by I1KDO » Sat Jan 22, 2022 5:53 pm

:D
ok

iu2dlc
Posts: 8
Joined: Fri Mar 06, 2020 6:29 am

Re: High BER on Hotspots

Post by iu2dlc » Tue Feb 22, 2022 7:18 pm

Well i don't know if this can help, at least it is just to say. I had the same issues with an hotspot made with an old GD77 and OpenGD77. In few words RX freq. was never tuned it means that i had ber always >1 oftentime.
I tried in several ways. The first workaround i found was setting +1000hz offset in my Anytone 878UVII it drastically solved the issues but it is a "dirty" fix at all. I tried again and again changing pi-star as well and the results were acceptable but obviusly the freq was also indicated as 433.674 instead of .675. Anyway ..in the end i found another more deep trick that worked. It's a bit dirty as well but at least it make me able do not touch offsets both in pi-star and 878...well just changing the reference oscillating tuner in calibration of opengd77 putting it to -20. It is clear that the radio was not perfectly tuned..strange becaouse i never touched cal. previosly.
Anyway a mistery remain: the small Radioddity GD-73 is not able to work with my hotspot...i tried all the workaround written before. Unfortunatelly the gd-73 does not have a fine offset tuning as the AT878 but anyway even with "playing" with pi-star never was able to make the small radio works with it. TBH i don't care except for the mystery itself. I do not have any analyzer on top of this gd-73 aerial is built-in so i guess nothing to do even discoryng the issues. That's all ..maybe the part belonging to gd77 could be a light for someone even with the new experimental fw on 9600.

User avatar
F1RMB
Posts: 2518
Joined: Sat Nov 16, 2019 5:42 am
Location: Grenoble, France

Re: High BER on Hotspots

Post by F1RMB » Tue Feb 22, 2022 7:25 pm

Changing the reference oscillator value change the frequency alignment,
Of course, your GD can be a bit off frequency, so you can improve that using measuring tools, like a good frequency counter.

About the GD-73, IIRC Roger already posted a comment about it.


Cheers.
---
Daniel

SA0BUX
Posts: 544
Joined: Tue Jul 05, 2022 8:50 am
Location: JO99ah, Stockholm, Sweden
Contact:

Re: High BER on Hotspots

Post by SA0BUX » Sat Jul 09, 2022 3:48 pm

I did some tests with the June releases and my md-9600 v2.1 HW.
I have both a simplex and a duplex mmdvm Hotspot and my tests towards parrot wasn't stable until I used 1 W or more.
  • Duplex Hotspot, radio didn't signaled channel setup until I used around 1 W, radio displayed timeout.
  • Simplex, you could start talking but it was unstable with lower power from parrot
Last edited by SA0BUX on Sun Jul 10, 2022 8:22 am, edited 1 time in total.

G4EML
Posts: 919
Joined: Sat Nov 16, 2019 10:01 am

Re: High BER on Hotspots

Post by G4EML » Sat Jul 09, 2022 8:54 pm

The MD-9600 hardware was never designed to work with less than 5W output. And even then it didn’t work well on early radios.
Although we are allowing lower powers to be selected there is no guarantee that they will work correctly.

Colin

Post Reply