Langstone Update 28/05/20
Langstone Update 28/05/20
2 changes...
GNU Radio Modules updated to use 48Khz sampling rate for the audio. This may make it more compatible with more of the cheap dongles.
(You will still need to change the dongle name in Lang_TX.py and Lang_RX.py if it is not 'hw:Card=device'
It will be interesting to see if it improves the Tx latency drift.
Change the way the update works to the same method used by Dave in the Portsdown A27 load.
Update commands without Portsdown are:-
cd Langstone
./stop
./update
cd .
./run
The additional cd . is needed because the update command now completely deletes and re-creates the Langstone Directory, this confuses the shell!.
Langstone .conf is preserved during the update so you settings should not be lost.
Colin G4EML
GNU Radio Modules updated to use 48Khz sampling rate for the audio. This may make it more compatible with more of the cheap dongles.
(You will still need to change the dongle name in Lang_TX.py and Lang_RX.py if it is not 'hw:Card=device'
It will be interesting to see if it improves the Tx latency drift.
Change the way the update works to the same method used by Dave in the Portsdown A27 load.
Update commands without Portsdown are:-
cd Langstone
./stop
./update
cd .
./run
The additional cd . is needed because the update command now completely deletes and re-creates the Langstone Directory, this confuses the shell!.
Langstone .conf is preserved during the update so you settings should not be lost.
Colin G4EML
Re: Langstone Update 28/05/20
Hi Colin,
That seems to work nicely and it does appear to improve the TX latency drift. I gave it a good waffle on the dummy load and I couldn't hear any latency additional to the expected few mS.
I might try out a different, but still cheap, sound card tomorrow.
All the best,
Martin - G4FKK
That seems to work nicely and it does appear to improve the TX latency drift. I gave it a good waffle on the dummy load and I couldn't hear any latency additional to the expected few mS.
I might try out a different, but still cheap, sound card tomorrow.
All the best,
Martin - G4FKK
Re: Langstone Update 28/05/20
Hmm - I may have spoken too soon - sorry. After a lot more tx testing, the latency did begin to build up but not as fast as previously and, so far at any rate, not such a long delay. I'll try again tomorrow.
73, Martin - G4FKK
73, Martin - G4FKK
Re: Langstone Update 28/05/20
I am not surprised the latency is still there. It will probably be different due to the change in clock, but the sound dongle is still using the same crystal so it’s unlikely it will change much.
I am still investigating how I can reset it before each transmission. From searching on the net it appears to be a common problem, there are a few suggestions but no definite cure.
Colin.
I am still investigating how I can reset it before each transmission. From searching on the net it appears to be a common problem, there are a few suggestions but no definite cure.
Colin.
Re: Langstone Update 28/05/20
I have made a further small change to the TX and RX files. This is now stopping and starting the flowgraphs when going from TX to RX instead of freezing them.
I think this will reset the internal buffers. (It seemed to do so on my testing) Latency will still increase during a transmission but should be reset when going back to receive.
Colin G4EML.
I think this will reset the internal buffers. (It seemed to do so on my testing) Latency will still increase during a transmission but should be reset when going back to receive.
Colin G4EML.
Re: Langstone Update 28/05/20
Hi Colin,
I updated to the latest version yesterday morning and spent a fair amount of time chasing es on 4m throughout the day. Your mod' appears to work as you expected and the TX latency gets reset between overs. I did notice that there is, intermittently, a delay between going from RX to TX and vice versa. On one occasion, going back to receive, the audio disappeared on TX and RX. Everything else appeared to be working normally - just no audio. No amount of 'fiddling about' (swapping modes, operating PTT, changing bands etc.), restored it. I then exited Langstone and used the Portsdown re-boot. The Pi shut down but didn't start up again. A power-cycle brought things back. There was still, as before, the occasional hesitation when using the PTT (either the GPI or the PTT touch button), but no further loss of audio. I wonder if anyone else is experiencing the same effects?
I don't think the TX/RX random delays are down, necessarily, to your latest mod' as I'd noticed that effect before, but maybe the loss of audio?
Hope this helps and let me know if it doesn't!
73, Martin - G4FKK
I updated to the latest version yesterday morning and spent a fair amount of time chasing es on 4m throughout the day. Your mod' appears to work as you expected and the TX latency gets reset between overs. I did notice that there is, intermittently, a delay between going from RX to TX and vice versa. On one occasion, going back to receive, the audio disappeared on TX and RX. Everything else appeared to be working normally - just no audio. No amount of 'fiddling about' (swapping modes, operating PTT, changing bands etc.), restored it. I then exited Langstone and used the Portsdown re-boot. The Pi shut down but didn't start up again. A power-cycle brought things back. There was still, as before, the occasional hesitation when using the PTT (either the GPI or the PTT touch button), but no further loss of audio. I wonder if anyone else is experiencing the same effects?
I don't think the TX/RX random delays are down, necessarily, to your latest mod' as I'd noticed that effect before, but maybe the loss of audio?
Hope this helps and let me know if it doesn't!
73, Martin - G4FKK
Re: Langstone Update 28/05/20
Hi Martin,
I haven’t noticed the delay or the loss of audio. But it sounds like you have probably been doing a lot more testing than me.
Because I am now stopping the receive code while transmitting and restarting it again when going back to receive it could be that it is uncovering a problem starting and stopping the audio dongle. If the pi wouldn’t start again I wonder if the audio dongle locked up.
Colin
I haven’t noticed the delay or the loss of audio. But it sounds like you have probably been doing a lot more testing than me.
Because I am now stopping the receive code while transmitting and restarting it again when going back to receive it could be that it is uncovering a problem starting and stopping the audio dongle. If the pi wouldn’t start again I wonder if the audio dongle locked up.
Colin
Re: Langstone Update 28/05/20
Hi Colin,
The complete loss of audio has only happened once so far and the delay in switching is intermittent so will, I guess, be difficult to pin down. I assumed the sound card had locked up as there was no audio in either direction. It was flashing its LED though which it seems only to do when it's 'active'.
I'll give it another thrashing later today and see if I can detect any pattern to the switching and also see if the audio packs up again.
Cheers,
Martin - G4FKK
The complete loss of audio has only happened once so far and the delay in switching is intermittent so will, I guess, be difficult to pin down. I assumed the sound card had locked up as there was no audio in either direction. It was flashing its LED though which it seems only to do when it's 'active'.
I'll give it another thrashing later today and see if I can detect any pattern to the switching and also see if the audio packs up again.
Cheers,
Martin - G4FKK
Re: Langstone Update 28/05/20
I have nearly completed the basic build - just the mouse to go and than the IO. Would it be possible to have 7 outputs for relays - I thought I would use one of the cheap Ebay 8 relay boards, they are opto-isolated and 3.3V driven which is fine for Langstone, but a waste to only use 5 of them.
Mike
Mike
Re: Langstone Update 28/05/20
Hi Mike,
Couldn't you use something like a 3 to 8 line decoder chip? (74HC138/238 maybe?). Colin has pre-coded the GPOs in binary.
73, Martin - G4FKK
Couldn't you use something like a 3 to 8 line decoder chip? (74HC138/238 maybe?). Colin has pre-coded the GPOs in binary.
73, Martin - G4FKK