Motorboating on Tx audio

VK3KYY
Posts: 7475
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Motorboating on Tx audio

Post by VK3KYY » Thu Sep 17, 2020 11:33 pm

I've been getting increasing reports of motorboating on Tx audio

Some speculation was that this is caused by the backlight dimming system

To rule this out, can anyone who has motorboating on Tx on their GD77 try the version I posted to this github issue

https://github.com/rogerclarkmelbourne/ ... issues/728

Which may be downloadable from this link.

https://github.com/rogerclarkmelbourne/ ... imming.zip

User avatar
m1dyp
Posts: 601
Joined: Sat Nov 16, 2019 8:03 am
Location: Hertfordshire, U.K.
Contact:

Re: Motorboating on Tx audio

Post by m1dyp » Fri Sep 18, 2020 6:13 am

hi, i find this is usually down to cheap mics. i use a kenwood smc-34 and never have issues, however, baofeng ones are notorious for this.

VK3KYY
Posts: 7475
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Motorboating on Tx audio

Post by VK3KYY » Fri Sep 18, 2020 6:23 am

m1dyp wrote:
Fri Sep 18, 2020 6:13 am
hi, i find this is usually down to cheap mics. i use a kenwood smc-34 and never have issues, however, baofeng ones are notorious for this.
Yes. The external mics are usually a cause of this, but some radios seem to do it themselves.

I've done some testing this afternoon with VK7ZCR, who has a radio which makes this noise, and the version without the PWM dimming did not make any difference to the Tx audio for him. So I think its a red-herring...

He also tried the version from last Christmas, and the motorboating on his radio sounded the same.

So, at least in his case we can rule out the dimming of a recent change to the firmware.

I know this noise can be very loud on some radios, VK3MBW's radio seems to have particularly loud background noise, which sounds like he has large fans running in the shack, but he could have some other hardware fault on his radio.

If I get time a the weekend, I'll build a version, which does not actually sample the mic audio, and instead feeds the digital equivalance of silence into the AMBE encoder, and see whether VK7ZCR's radio is then silent or has mototboating still.

One think which I recall testing was the output from the AMBE codec in the GD77 when its fed with silence, because I don't think the audio it encodes is completely silent. This is probably because of the AMBE codec in the firmware, which Radioddity (TYT or whoever in China) wrote from the AMBE spec, so that their version is not copyrighted, as its not exactly AMBE ..

So I may need to also write a test version, which sends the 27 bytes of data, which MMDVMHost uses when it has to insert silence into the audio, because of short network outages etc.
And see how that sounds

User avatar
m1dyp
Posts: 601
Joined: Sat Nov 16, 2019 8:03 am
Location: Hertfordshire, U.K.
Contact:

Re: Motorboating on Tx audio

Post by m1dyp » Fri Sep 18, 2020 11:58 am

ah ok

User avatar
kd2lh
Posts: 312
Joined: Mon Dec 02, 2019 2:44 pm

Re: Motorboating on Tx audio

Post by kd2lh » Sat Sep 19, 2020 2:19 am

Interesting...

I spoke with someone a couple of days ago that said they were using a Motorola mobile XPR4550 through an area repeater (I was receiving them through a jumbospot and a Motorola handheld) and there was low level motor boating on the signal.

I don't think I've ever heard it on my equipment (GD77 or otherwise).

VK3KYY
Posts: 7475
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Motorboating on Tx audio

Post by VK3KYY » Sat Sep 19, 2020 2:59 am

kd2lh wrote:
Sat Sep 19, 2020 2:19 am
Interesting...

I spoke with someone a couple of days ago that said they were using a Motorola mobile XPR4550 through an area repeater (I was receiving them through a jumbospot and a Motorola handheld) and there was low level motor boating on the signal.

I don't think I've ever heard it on my equipment (GD77 or otherwise).
Interesting.

I would be good to track down what is causing this, so I can establish if there is a possible firmware fix for this.

Its also strange that people who have this problem, say the official firmware doesn't have it.

If its analogue noise on the mic, then its likely to be caused by supply voltage fluctuations, but I've no idea what we could turn off during Tx, apart from the PWM dimming (if that was the problem, which it doesn't seem to be)

The EEPROM and Flash memory can be put into low power modes, but I doubt they would cause this sort of noise, as we should not be accessing either the EEPROM or the Flash during transmission.

VK3KYY
Posts: 7475
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Motorboating on Tx audio

Post by VK3KYY » Sun Sep 20, 2020 4:38 am

kd2lh wrote:
Sat Sep 19, 2020 2:19 am
Interesting...

I spoke with someone a couple of days ago that said they were using a Motorola mobile XPR4550 through an area repeater (I was receiving them through a jumbospot and a Motorola handheld) and there was low level motor boating on the signal.

I don't think I've ever heard it on my equipment (GD77 or otherwise).
Mark

After your discussion this morning, I tried clearing the lower bit, and also multiple lower bits, to see if it resolved the motorboating problem, which occurs on some radios.

However it didn't fix the problem.

I can clear several of the least significant bits and the audio remains the same, and its not until I clear the lower 5 bits that the change becomes particularly noticeable, with what appears to be a hiss sound on the DMR audio, but there is still the motorboating sound, even though there is a hiss.

Its difficult to actually look at the sampled audio, because the sample rate is 8kHz, 16 bit, and I don't have a way to send data that fast to the PC, either via USB serial or by the serial wire debug (as the serial wire debug only runs at 4kHz)

The only way to can see to look at the samples, would be to attach a PC logic analyser to the I2S bus, but its not practical for me to do this, as it involves taking a radio completely apart and soldering tiny wires onto the I2S data and clock lines etc.


Its almost certainly a hardware problem on some radios, but its strange that apparently the official firmware does not suffer from this problem.

Also as discussed this morning, perhaps the official firmware does something like not aligning the sampled audio blocks with the DMR transmission blocks, so perhaps the noise signal would be in the middle of the 60mS sampled audio and this would possibly have a different effect on the way the codec encodes the audio.

Either way, I don't have a simple solution to this.

User avatar
kd2lh
Posts: 312
Joined: Mon Dec 02, 2019 2:44 pm

Re: Motorboating on Tx audio

Post by kd2lh » Mon Sep 21, 2020 1:08 am

Well, checking for dither was worth the test.

The hiss would reflect less dynamic range caused by eliminating the low order sample data.

It would be interesting to capture the analog data on the way into the DAC to see if the motorboating is in the analog audio.

VK3KYY
Posts: 7475
Joined: Sat Nov 16, 2019 3:25 am
Location: Melbourne, Australia

Re: Motorboating on Tx audio

Post by VK3KYY » Mon Sep 21, 2020 1:14 am

kd2lh wrote:
Mon Sep 21, 2020 1:08 am
It would be interesting to capture the analog data on the way into the DAC to see if the motorboating is in the analog audio.
Yes. It would, but I can't think of an easy way to capture it.

dl9sec
Posts: 29
Joined: Sun Jun 28, 2020 11:32 am

Re: New ongoing "Development" version has been created

Post by dl9sec » Wed Oct 07, 2020 11:56 am

Interesting...

with the last firmware before that one from today this disturbed audio (especially if the loudness is set to low) was gone.
With the update today, the disturbed audio came back.

Really strange.

73, Thorsten

Post Reply