Better frequency accuracy with FW 1807A for DV10

[F5HPE]

For few day, AOR had launch a new DV10 firmware with detail below

1807A 2018/07/13
Feature improvements:
- Better frequency accuracy through additional automatic temperature compensation by FPGA/CPU
- Better P25 and DMR decoding
- Better SSB audio quality
- Better SD card compatibility for recordings.
- CTCSS, DCS, reverse tone, voice scramble user settings now displayed during reception.

Bug fixes:
- Head of digital signals sometimes cut
- D-STAR only working in AUTO mode
- SI command behaved erratically.
- LCD contrast sometimes turned dark during charging.
- Dial selector knob sometimes irresponsive in MemCHEDIT and SearchBankEDIT.
- Other minor fixes

 

[TMac20]

Hello,

Can you or any other users outline what they have noticed by way of any improvements, what is fixed, what is still not fixed what is better and what has not changed.

Thank you

 

[F5HPE]

Hello,

Can you or any other users outline what they have noticed by way of any improvements, what is fixed, what is still not fixed what is better and what has not changed.

Thank you

I'm working on.

 

[TMac20]

Hello All

Today I have some good news to share with everyone confirming the following

---------------
AOR says.

Any and all upgrades, updates, removal of "bugs", reducing frequency drift, etc., will ALL be done exclusively via firmware,

---------------

So now we need feedback on the firmware updates please with as much details as possible. There will still be some issues I am sure, but let's tick off the ones that have been sorted out so they do not have to be revisited.

 

[woodpecker]

---------------
AOR says.

Any and all upgrades, updates, removal of "bugs", reducing frequency drift, etc., will ALL be done exclusively via firmware,

---------------

The HF frequency error, at the lowest temp I could get the radio to (about 5 degrees C) was -400Hz which is 72ppm, they've had one go at fixing the drift fault in firmware, improved the UHF somewhat although its still poor.

I don't believe its possible to fix drift errors this large in firmware, there must be an issue with the TCXO or they did not use a TCXO or there is something wrong with hardware design itself.

 

[TMac20]

Thanks Woodpecker, can anyone else replicate this and please report using the exact same methods.

 

[c0ne]

Thanks Woodpecker, can anyone else replicate this and please report using the exact same methods.

I don't have such expensive radio analyser, but I did a different test. Here we have a POCSAG pager signal that broadcasts the same message ever 30 seconds. It has a signal bandwidth of 6Khz and is located at 456.4500Mhz

Before the firmware upgrade, I could only make a discriminator recording that decoded fine when the radio was cold. eq. just turned on.. 20 minutes later the recordings were allready corrupted and no longer useful.

With the latest firmware, the recordings also seem to work when the radio is warm. They reduced the drift to a certain level that is seems workable for me. needless to say, any further improvement of this is welcome.

Next, I would like to see AOR address the DMR decoding issue(s). I know from a AOR reply they did not have live networks and signal to test on, so any idea on how we could help AOR address this is welcome. Maybe its also a good idea if AOR listens a bit more to woodpecker his suggestion regarding the use of a 6Khz IF filter, even if its just a Beta firmware so we can test it.

 

[woodpecker]

UHF is better, so far I have noted a drift of -1400Hz to + 1500Hz, its better than it was but not very good for a modern receiver.

If trying to use this on narrow channels like COSPAS with 3kHz spacing, when trying to listen for a beacon on for example channel 3, a user doing a rescue on a cold mountain would be tuned half way between channels 2 and 3, whereas another rescuer doing a rescue on a hot beach could be half way between channels 3 and 4 when tuned to channel 3, only at some fixed temperature in between would the DV10 actually be on the correct frequency for a cospas channel rather than somewhere in the middle between channels.

It doesn't seem possible to use a radio with this inaccuracy for any life saving or professional function, its not really accurate enough for hobby listening.

 

[F5HPE]

...If trying to use this on narrow channels like COSPAS with 3kHz spacing a user on a cold mountain could be half way between channels 2 and 3 whereas someone on a hot beach could be half way between channels 3 and 4, only at some fixed temperature in between would you actually be on the correct frequency for a channel....

Hey woodpecker
If you have planned to spend a few days of summer vacation in the country of Football World Cup 2018 Champions , do you think to make a hook by Paris?
If so, I invite you to share several beers at the feet of the Effeil Tower.
And I will give you training on COSPAS, PLB and SAR.
Be reassured you will not have to pay anything.
It will be my pleasure to share this knowledge with you.
Cheers

 

[EricCottrell]

The HF frequency error, at the lowest temp I could get the radio to (about 5 degrees C) was -400Hz which is 72ppm, they've had one go at fixing the drift fault in firmware, improved the UHF somewhat although its still poor.

I don't believe its possible to fix drift errors this large in firmware, there must be an issue with the TCXO or they did not use a TCXO or there is something wrong with hardware design itself.

Hello,

Was the HF frequency below 18 MHz? The AR-Alpha and AR-DV1 use Direct Conversion below 18 MHz. I assume the AR-DV10 does the same. I wonder if it is the sample clock used for the ADC that is drifting. It would be interesting to compare the frequency errors below and above 18 MHz.

73 Eric

 

[hamradionl]

I owned a DV1 but like share some opinion about these DV10 matters, i be engineer for 45 years using solder iron and iam one of those older generation who build all kind of radios between 1.0MHz to 24GHz. I know freq drifting is caused by faulty hardware, so i be open to any information or prove that firmware udates is able to solve very large freq drifting.

 

[woodpecker]

Hello,

Was the HF frequency below 18 MHz? The AR-Alpha and AR-DV1 use Direct Conversion below 18 MHz. I assume the AR-DV10 does the same. I wonder if it is the sample clock used for the ADC that is drifting. It would be interesting to compare the frequency errors below and above 18 MHz.

73 Eric

Hi Eric,

Testing at 25MHz.

Radio cooled for 1 hour in 5 deg C fridge - probably not long enough to get it cold enough and way above the -10 deg C spec limit.

Ambient air temperature 25 deg C

At power on - 282Hz off frequency 11.3ppm
After 5 minutes - 240Hz off frequency
After 10 minutes - 180Hz off frequency
After 15 minutes - 130Hz off frequency
After 20 minutes - 105Hz off frequency
After 25 minutes - 70Hz off frequency
After 30 minutes - 60Hz off frequency
After 35 minutes - 50Hz off frequency
After 40 minutes - 40Hz off frequency

Conclusion at 25MHz - still has a large frequency drift, way out of spec, if cooled to -10 spec limit it would clearly be even worse.

AOR need to test it properly in an environmental chamber and record frequency error vs temperature, even basic tests show its way short of meeting spec.

 

[EricCottrell]

Hi Eric,

Testing at 25MHz.

Radio cooled for 1 hour in 5 deg C fridge - probably not long enough to get it cold enough and way above the -10 deg C spec limit.

Ambient air temperature 25 deg C

At power on - 282Hz off frequency 11.3ppm
After 5 minutes - 240Hz off frequency
After 10 minutes - 180Hz off frequency
After 15 minutes - 130Hz off frequency
After 20 minutes - 105Hz off frequency
After 25 minutes - 70Hz off frequency
After 30 minutes - 60Hz off frequency
After 35 minutes - 50Hz off frequency
After 40 minutes - 40Hz off frequency

Conclusion at 25MHz - still has a large frequency drift, way out of spec, if cooled to -10 spec limit it would clearly be even worse.

AOR need to test it properly in an environmental chamber and record frequency error vs temperature, even basic tests show its way short of meeting spec.

Hello,

So the frequency error at 25 MHz was slightly less then 5 MHz at 5 degrees C?

I noticed in the brochure that the AR-DV10 is not using direct conversion like the AR-DV1, but is converting to a 47.25 MHz IF and digitizing it. There are two signal paths shown with a dividing point of 76 MHz.

73 Eric

 

[KevinC]

Hello All

Today I have some good news to share with everyone confirming the following

---------------
AOR says.

Any and all upgrades, updates, removal of "bugs", reducing frequency drift, etc., will ALL be done exclusively via firmware,

---------------

So now we need feedback on the firmware updates please with as much details as possible. There will still be some issues I am sure, but let's tick off the ones that have been sorted out so they do not have to be revisited.

Just curious, where did you get this information?

 

[TMac20]

Hello Kevin - it was a response that was direct from AOR.

I will open another thread shortly just for technical results for people to post to (me included). Hopefully everyone can keep it simple and just post on topic without badmouthing anyone - it's always more pleasant.

Nice to see you still on here after all these years - it's been a long time now hope all is well.

 

[woodpecker]

DV10 Useless for CW/SSB on 70cm or above

Although the 1807A firmware made some improvement in frequency drift the DV10 is unusable for CW or SSB on 70cm or above. I don't think there is any hope of a firmware fix for this.

The video below shows a comparison of the DV10 vs the AOR DV1 trying to listen to the GB3UHF 70cm beacon. Due to the frequency dirft and error, when the DV10 is tuned to the exact frequency of 432.430MHz it appears the beacon cannot be heard, howvever if the DV10 is tuned higher by 1.2-1.3kHz the beacon can be heard. It means when trying to tune into a narrowband signal you won't know what frequency to set the DV10 to. On 23cm (1296MHz) the error is approx three times as a bad.

https://www.youtube.com/watch?v=NMP3W1HKqC4