Reference oscilator value? (DMR audio artifacts on RX only)

[G7RZU]

Hi

I am getting broken audio on RX only via my repeater, when all other radios are fine.

TX BER is around 0.3 from the GD77, as opposed to 0.1 or less on my MD380.

Also, I am very close to the repeater as it is located at my QTH.

I have been considering changing the reference oscillator value to see if this makes a difference. Is this advised?

However, I see the value being discussed as being a number centred around 1000. However, CPS shows my UHF value as -25. Is this -25 based on a nominal value of 1000?

Thanks and 73

Simon - G7RZU

[G4EML]

The previous comments about the value being around 1000 were based on a wrong assumption about the scaling of the adjustment.
It is a 10 bit number which was originally scaled from 0 to 1023 but in fact it should actually be from -511 to + 512. The CPS was changed to reflect this.
The value has not changed it is just represented differently.

An oscillator that was perfectly on frequency would need a calibration value of zero. The previous recommendation of 1000 equates to a value of -23 with the new scaling which is fairly typical for most radios.

The adjustment of frequency should not actually be that critical. The DMR specification allows for a frequency error of around 400Hz at UHF. Unfortunately many MMDVM based repeaters have problems achieving this. Adjusting a radio to suit a particular repeater may help but could equally make it perform worse on others.

Colin G4EML.

[G7RZU]

Thanks Colin, that explains things. My repeater was on frequency when it was installed, but I haven't checked it recently. Also my other radios seem fine with the repeater. I have recently been given this radio (Thanks Jon, G4TSN!) so am experimenting.

I am also wondering if, for some reason, this radio is only showing audio problems with traffic coming from the network, not local to the repeater, although this may require more tests.

Do these radios struggle significantly with strong signals? I wonder if being so close to the repeater is an issue?

Simon

[G4EML]

Hi Simon,

The radio, like many others using the RDA chip suffers when trying to listen to a weak signal with other strong signals nearby, but that is not your problem. It doesn't seem to have a problem just because the signal you are receiving is very strong. Indeed it is being used in Hotspot mode by many stations and you probably won't get a much stronger signal than from another handheld in the same room.

Colin

[G7RZU]

Well, on more testing it seems that this audio issue is only present with OpenGD77 firmware. The stock firmware seems fine. Interesting...

[VK3KYY]

Just to confirm what Colin has said.

The reference oscillator value doesn't seem to make a big difference to the operation.

In your case I don't know why you would be getting broken audio.

Does the Green LED go on and off, or is it on solidly.

There are many aspects of the hardware of the radio for which there is little or no documentation, this includes both the RF chip and also the DMR chip.

So its unlikely that this firmware may ever operate as well as the official firmware, which is probably written by a division of the Chinese company who make the DMR chip.

I've discussed what limited documentation we have for the DMR chip, with some Chinese programmers, and they can't make any more sense out of it than we could.
We only have one data sheet, which someone kindly paid for, via a dodgy Chinese file sharing site. And the file is corrupt and we can't load it all, and its clear from looking at what we have, is that this is just an overview data sheet and not in anyway the complete reference material for that chip.

Likewise with the radio chip, the only documentation we have, was found my trawling numerous Chinese language file sharing sites, and grabbing what we could, however there are still number of commands which the official firmware sends to the chip (which we mimic), but don't understand what they actually control.

[G7RZU]

Hi

Thanks for coming back to me and your detailed explanation.

As a further update, I installed the Radioddity stock firmware, ran that for a couple of days and then went back to OpenGD77 and the audio seems improved. I did notice that something to do with audio quality was in the Radioddity changelog. I am guessing that the stock firmware could have reset or changed something in one of the ROMs which has caused an audio improvement that has persisted even after going back to OpenGD77. Does this sound a feasable explanation?

73

Simon

[VK3KYY]

Hi

Thanks for coming back to me and your detailed explanation.

As a further update, I installed the Radioddity stock firmware, ran that for a couple of days and then went back to OpenGD77 and the audio seems improved. I did notice that something to do with audio quality was in the Radioddity changelog. I am guessing that the stock firmware could have reset or changed something in one of the ROMs which has caused an audio improvement that has persisted even after going back to OpenGD77. Does this sound a feasable explanation?

73

Simon

The only shared data is the Calibration data table, and the OpenGD77 firmware makes its own copy of the Calibration data to use.


It seems unlikely that Radioddity would have changed the data in the calibration table in different versions of their firmware.

However, their track record with this isn't good, because for a long time (and perhaps even now), the official firmware has a serious bug, when handling the Calibration data backup (which is also stored in the Flash memory chip)

Their emergency restore function, does not restore the backup at all.
It actually copies the current Calibration data to the backup.

So if the main Calibration data gets corrupted, then the corrupt version gets backed up, hence overwriting the non corrupted backup.

There have been multiple instances where people's radios stopped working, because the firmware screwed up the Calibration table.
I had to write a program called Flash Manager to be able to copy the Calibration from one radio to a file on a PC, then upload that data to a radio running the official firmware, which has corrupt calibration data.

However, the OpenGD77 and OpenGD77CPS supersedes this, as the calibration data can be read and written by the OpenGD77CPS and also the OpenGD77 firmware manages its own copy of the calibration data.


Overall, I think its likely to be other factors which are effecting your radio, but I can't entirely rule out some change to the calibration data.
BTW.
Did you do a full restore of the Flash and EEPROM backups, before reloading the official firmware.
If not, I don't know for sure, even whether the OpenGD77 would still have its own copy of the Calibration data, so any changes made to the official firmware's copy of the calibration data may not make any difference as its not used if the OpenGD77 copy of the firmware is still in the memory