Page 2 of 5
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 11:15 am
by G4EML
F1CXG wrote: ↑Mon Mar 02, 2020 8:35 am
Thanks for this version Roger,
Version tested on my GD77 which squeaks, it still remains a small weak squeal, but I think it's independent of the software, from the beginning, an operation with microphone / speaker is normal.
(It looks like a "larsen" between the LF amplifier and the internal speaker which are very close.)
The squeaking is a know design error with some radios. It can be fixed but needs a component change. See
http://members.optuszoo.com.au/jason.re ... 77mods.htm for details.
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 12:36 pm
by F1CXG
Thank you Colin, I know the modifications to be made, but when opening one of my GD77 for another problem, I saw the size of the SMD around the amplifier and I gave up the idea of making the modification
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 12:42 pm
by G4EML
Yes, the components are very small. Luckily neither of my GD77s has the problem.
73 Colin.
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 2:37 pm
by KU4ZD
I am noticing that with the filter set to 'none', after I transmit the cc changes from the setting in the code plug to 0 and the net time I transmit it transmits on cc0. This happens on the hot and digital simplex. The result is that I am sending out the wrong cc. It does not happen if I have the filter set to CC. If the filter is set to CC all functions correctly. I confirmed the CC actually does change by going to a simplex frequency on another HT and testing.
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 3:44 pm
by G4EML
That is because with the filter set to none the radio is constantly trying to detect the correct colour code. Sometimes it gets it wrong especially if the signals are weak or have a lot of noise. For that reason I would not recommend leaving the radio set to none in normal use. This feature is really intended for occasional use when the signal CC is unknown.
Maybe we actually need to go as far as disabling transmission.
Colin.
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 3:45 pm
by KC7RBW
KU4ZD wrote: ↑Mon Mar 02, 2020 2:37 pm
I am noticing that with the filter set to 'none', after I transmit the cc changes from the setting in the code plug to 0 and the net time I transmit it transmits on cc0. This happens on the hot and digital simplex. The result is that I am sending out the wrong cc. It does not happen if I have the filter set to CC. If the filter is set to CC all functions correctly. I confirmed the CC actually does change by going to a simplex frequency on another HT and testing.
This is a bug I saw on the NA7Q firmware, but hadn't seen (yet) on Roger's build.
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 3:59 pm
by G4EML
I have seen it occasionally on the main firmware. It is something we may be able to improve in the longer term.
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 4:01 pm
by KU4ZD
KC7RBW wrote: ↑Mon Mar 02, 2020 3:45 pm
This is a bug I saw on the NA7Q firmware, but hadn't seen (yet) on Roger's build.
I have not tried NA7Q's version yet, but this is on Roger's current version here
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 4:01 pm
by EB3AM
G4EML wrote: ↑Mon Mar 02, 2020 3:44 pm
That is because with the filter set to none the radio is constantly trying to detect the correct colour code. Sometimes it gets it wrong especially if the signals are weak or have a lot of noise. For that reason I would not recommend leaving the radio set to none in normal use. This feature is really intended for occasional use when the signal CC is unknown.
Maybe we actually need to go as far as disabling transmission.
Colin.
Ok, but you could try to implement that if the software is not clear what color to use, use that which is defined in the channel's memory. This will prevent you from transmitting in the wrong color.
Re: Experimental versions to fix potential speaker hiss
Posted: Mon Mar 02, 2020 8:18 pm
by VK3KYY
The histogram table is probably being reset by the Tx system.
It should be easy to fix