Noise Burst on initial transmit

User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Mon Mar 29, 2021 3:48 pm

It is extraordinarily hard for me to believe that any Windows sound interface would generate 90 or 200ms of delay. That is what would be required to fit the evidence. There must be something else wrong somewhere.

What happens if you use the motherboard sound interface instead of the Soundblaster card?

It doesn't seem like an RFI problem because RFI generally shows up as an echo.
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Noise Burst on initial transmit

Postby K1LSB » Mon Mar 29, 2021 5:25 pm

Scott,

Here's why I'm cursed with the audio latency that I deal with...

My AT-2035 mic feeds into a Behringer preamp (needed only to supply 48v to the mic). Not sure how much latency that preamp adds, shouldn't be much if any. The output is fed into Windows over USB.

I use EQ APO Windows-based software to do all of my audio processing (EQ'ing for both the mic and speakers, including several VST plug-ins in both streams). That adds some latency to the speaker and mic audio (no idea how much).

The mic audio is then fed to the radio over VAC1 using Windows DirectSound, buffer size of 2049 and sample rate of 96000. Not sure how much latency those settings add.

My manual buffer settings in VAC1 are 20/20 for RingBuffers and 40/40 for PortAudio - there's more latency there, not sure how much.

The process I used to arrive at the 90 ms value for RF Delay was to simply place the mic in front of the speaker and press the PTT while observing the peak-reading power meter on my AL-80B - any jump of the meter needle at key-down has to be a direct result of the sum of the total latencies in the speaker and mic streams. I kept gradually increasing the RF Delay value until the meter needle no longer moved upon PTT being pushed.

I have no ASIO drivers anywhere in the system so there is bound to be a fair amount of latency both coming and going (mic and speakers), given the rather immersive Windows-based audio processing for both streams.

You're probably wondering why I don't just use the embedded Thetis audio tools but I've never been able to get the mic audio results I'm looking for. Besides, in the big picture 90 milliseconds isn't enough for me to lose any sleep over (for comparison, Flex is roughly 120 milliseconds) . But that's how much I had to dial in to eliminate the burst at the beginning of my transmissions (which was brought to my attention by other hams who were listening to my transmissions).

Hope that helps,

Mark
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Mon Mar 29, 2021 7:14 pm

w-u-2-o wrote:It is extraordinarily hard for me to believe that any Windows sound interface would generate 90 or 200ms of delay. That is what would be required to fit the evidence. There must be something else wrong somewhere.

What happens if you use the motherboard sound interface instead of the Soundblaster card?

It doesn't seem like an RFI problem because RFI generally shows up as an echo.


My problem is exactly as K1LSB is describing. I will set and use the motherboard sound interface and try that. I also remember seeing this when I used a Behringer U-Phoria UMC204HD and using ASIO drivers. I have always seen the burst at the end of each transmission. I have always thought it was normal. It was only recently that someone said something about it. The folks that pointed it out are prone to pointing out every little tiny thing happening in everyone's audio. They have also been known to making things up. :D I'm sure they didn't give me any false issues because two people there played separate recordings of my transmission.

Using K1LSB's suggestion I have eliminated it by setting the Settings>General>Options>RF Delay to 200ms in Thetis.
I will still do what you suggested and use the motherboard sound interface. I will continue to search for other solutions or
diagnose the root cause.

I may put the Behringer back on and try that.
I'll be back
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Mon Mar 29, 2021 7:50 pm

Mark,

You can reduce your latency by quite a bit and probably get a better sounding solution by using better tools. You've got a good, ASIO capable interface, so the foundation is there!

Use Voicemeeter Potato to tie everything together so that you can use ASIO. Play with Voicemeeter and the Behringer control panel until you find the lowest Behringer buffer size that'll run cleanly. Then use this guide to interface Cantabile to hold your VSTs. Or, if you want to step up the game further, use Reaper DAW instead of Cantabile (that's what I do). Use a proper EQ plug-in, not APO. Do that and at the ASIO interface to Thetis you should be experiencing only a few milliseconds of latency unless your plug-ins do something crazy. Pare down your buffers in Thetis and you're latency should be under 40ms, under 30 if you work hard at it.

The other thing is that you really want to set up your mic and levels so that your mouth is within a couple of inches of the microphone. If you are not picking up the entire room then the mic won't pick up receive audio from the speakers much, either.

73,

Scott
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Noise Burst on initial transmit

Postby K1LSB » Tue Mar 30, 2021 12:04 am

Scott,

My preamp is a Behringer Tube UltraGain MIC500USB which is not ASIO capable (if it is I'm not aware of it). So if ASIO is an option someone will have to explain it to me because I'm clueless.

In fact, that's one of the reasons I went with EQ APO - that software loads its own low latency audio driver.

Given that I don't have ASIO capability I haven't seen a need for Voicemeeter.

With that said, let me repeat that I'm clueless about the function of ASIO so I wouldn't even know where to begin if I were to focus on a true low latency setup. But that's a subject for another thread, as I don't want to derail this one.

Thanks,

Mark
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Tue Mar 30, 2021 12:38 am

You need a UMC202HD interface, or equivalent.
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Noise Burst on initial transmit

Postby Tony EI7BMB » Tue Mar 30, 2021 8:22 am

Take a look at Motu M4 also
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Noise Burst on initial transmit

Postby K1LSB » Fri Apr 02, 2021 12:19 am

Well, I owe you one, Tony. I looked up some info on the Motu M4, liked what I saw so I did a bunch more research and learned that the M4 was just about the best value in that price range so I decided to get one. That's when I learned there aren't hardly any to be found anywhere. I finally found one at Markertek and snatched it up quick. It'll be here sometime next week.

So now I have a heckuva lot of learning to do 'cuz I don't know anything at all about audio interfaces. I'm gonna be depending on you guys to bail me out when I get myself painted into a corner.

Thanks again,

Mark
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Noise Burst on initial transmit

Postby Tony EI7BMB » Fri Apr 02, 2021 8:01 am

Good stuff Mark, I'd be happy to help out as I think our setups will be pretty similar.
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Fri Apr 02, 2021 4:16 pm

rbduck wrote:
w-u-2-o wrote:It is extraordinarily hard for me to believe that any Windows sound interface would generate 90 or 200ms of delay. That is what would be required to fit the evidence. There must be something else wrong somewhere.

What happens if you use the motherboard sound interface instead of the Soundblaster card?

It doesn't seem like an RFI problem because RFI generally shows up as an echo.


My problem is exactly as K1LSB is describing. I will set and use the motherboard sound interface and try that. I also remember seeing this when I used a Behringer U-Phoria UMC204HD and using ASIO drivers. I have always seen the burst at the end of each transmission. I have always thought it was normal. It was only recently that someone said something about it. The folks that pointed it out are prone to pointing out every little tiny thing happening in everyone's audio. They have also been known to making things up. :D I'm sure they didn't give me any false issues because two people there played separate recordings of my transmission.

Using K1LSB's suggestion I have eliminated it by setting the Settings>General>Options>RF Delay to 200ms in Thetis.
I will still do what you suggested and use the motherboard sound interface. I will continue to search for other solutions or
diagnose the root cause.

I may put the Behringer back on and try that.
I'll be back



Over the past couple of days I have tried several different setups. I have a clone drive so I booted up on that to make any changes. I removed the Sound Blaster and set it up wit nothing other than the integrated sound on the motherboard. I had the same resulting noise burst as I did with the SB card.

Next, I have some older computers around so I loaded Thetis 2.8.11 onto an HP 8200 ultra slim desktop with an Intel i7 2600 CPU. I set it up to use the integrated soundcard on the motherboard. It took some time to get it all configured. The noise burst was not quite as pronounced with that computer. It was there.

I tried the Behringer and the noise burst was there (drat). I just tried it on my main computer. I was also unable to get low latency settings as I do with the SB card. I figured if I tried harder I might get them lower. Maybe another time.

I'm still digging. I may try different firmware if you think that may have an affect on it? My curiosity has no bounds. Im geting good results with the RD Delay setting as suggested earlier.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Fri Apr 02, 2021 4:28 pm

Unfortunately I'm totally out of ideas at this point.
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Fri Apr 02, 2021 6:35 pm

w-u-2-o wrote:Unfortunately I'm totally out of ideas at this point.


Scott, I appreciate the ideas you have brought forward. You do an awesome job keeping the forum going and answering so many questions so well. I like the way you are able to memorize where folks need to go for information that would help in solving whatever issue they have.
Sometimes I might welcome being spoon fed answers , but mostly I would like to dig it out myself. With help sometimes of course.

I'm taking a break from this right now. I will continue to explore more for answer. I will post it here.
Happy good Friday and Happy Easter!
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Fri Apr 02, 2021 7:08 pm

Thanks for the kind words, Ruben :)
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Noise Burst on initial transmit

Postby K1LSB » Fri Apr 02, 2021 7:26 pm

Ruben,

Assuming that the burst is coming from the mic picking up sound still coming from the speakers after PTT is pressed, there are two possible sources of latency - the mic stream and the speaker stream. Minimizing both is what's desirable. In my case, the receive audio latency is minimized when the speakers are plugged directly into the Anan. Do you have the ability to drive your speakers from the Anan and see if that reduces the length of the burst?

On a side note, there's always the option of simply wearing headphones, so there's no audio coming from any speakers for the mic to pick up. Many hams use headphones exclusively in the shack (I rarely do so that's not a viable solution for me).

Mark
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Fri Apr 02, 2021 11:16 pm

K1LSB wrote:Ruben,

Assuming that the burst is coming from the mic picking up sound still coming from the speakers after PTT is pressed, there are two possible sources of latency - the mic stream and the speaker stream. Minimizing both is what's desirable. In my case, the receive audio latency is minimized when the speakers are plugged directly into the Anan. Do you have the ability to drive your speakers from the Anan and see if that reduces the length of the burst?


Mark,....Ya know. DUH! That's not a bad idea. Plus it wouldn't be difficult to try! It may be a day or two before I have a chance.(drat..again)

K1LSB wrote:On a side note, there's always the option of simply wearing headphones, so there's no audio coming from any speakers for the mic to pick up. Many hams use headphones exclusively in the shack (I rarely do so that's not a viable solution for me).

Mark


Yes I have a set of headphones with a mic and it works pretty well. I've had good reports. I hate not using the full fidelity of my Audio-Technica mic. I know it doesn't all translate through on SSB. What would help me most is a good set of hearing aids. :lol: No seriously, if I monitor more then talk I use my headphones. When I start talking off they come. I will report back with my results. Thanks again.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
wb8yst
Posts: 1
Joined: Thu Jan 21, 2021 3:21 am

Re: Noise Burst on initial transmit

Postby wb8yst » Sun Apr 04, 2021 2:26 pm

On my 100D setup>audio check marked "allow (PTT/Space/Mox) to override/bypass VAC for phone" to get rid of the "noise Burst on initial transmit"; the burst isn't there if the radio is squelched with or w/o these boxes checked.

However, using afsk WSJT-X while still in usb the burst comes back (no matter how the boxes are checked) only if the 100D isn't squelched.

wb8yst
100D
Protocol 1 ver 6.0
Thetis 2.8.11

My apologies if this has been covered.
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Sun Apr 04, 2021 6:02 pm

wb8yst wrote:On my 100D setup>audio check marked "allow (PTT/Space/Mox) to override/bypass VAC for phone" to get rid of the "noise Burst on initial transmit"; the burst isn't there if the radio is squelched with or w/o these boxes checked.


In my case the burst is there no matter how or what combination those four items are selected.


wb8yst wrote:However, using afsk WSJT-X while still in usb the burst comes back (no matter how the boxes are checked) only if the 100D isn't squelched.

wb8yst
100D
Protocol 1 ver 6.0
Thetis 2.8.11

My apologies if this has been covered.


Thank you for your information. I don't think this has been mentioned. I'm still going to plug the headphone out into the audio amp to see if that
changes anything. I thought it was going to be simple. I can't get to it until after Easter.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Mon Apr 19, 2021 11:29 am

@rbduck

Revisiting this problem of the noise burst: have you tried upgrading to the latest P2 firmware? See this topic:

viewtopic.php?f=32&t=3192
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Mon Apr 19, 2021 2:23 pm

w-u-2-o wrote:@rbduck

Revisiting this problem of the noise burst: have you tried upgrading to the latest P2 firmware? See this topic:

viewtopic.php?f=32&t=3192


Yes, this is the version I have been running. I have also tried every other version including Protocol 1. There is no difference at all between
them all. I have intentions in trying the motherboard sound circuitry. I want to pull the Sound Blaster card completely when I do. I haven't had the time to do that. (((((((I read over in another topic where someone is having a similar issue when running VOX? The problem doesn't occur at all when I run VOX.))))))) oops(( I didn't read closely enough. I think he was actually referring to this topic))
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Mon Apr 19, 2021 2:29 pm

Ruben,

There's another thread dealing with this same problem. It's been suggested there that

a) this problem is related to the use of DEXP

and

b) the RF delay can be set to around 35 or 45ms to work around (I would not say "solve") the problem.

Does any of this apply to your problem?
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Mon Apr 19, 2021 9:08 pm

w-u-2-o wrote:Ruben,

There's another thread dealing with this same problem. It's been suggested there that

a) this problem is related to the use of DEXP

and

b) the RF delay can be set to around 35 or 45ms to work around (I would not say "solve") the problem.

Does any of this apply to your problem?


I want to make very sure of what I am saying here. I have been reading the topic "Downward Expander 2.8.11". I too was curious if this is similar to what I have been discussing.

a) This seems to have nothing to do with DEXP. My noise burst only occurs when using MOX. The result is the same whether I push my PTT, Click the MOX button with the mouse or pressing the spacebar. I can detect no noise burst using VOX at all. That is with RF Delay set at 0ms.


b) Until I can find the root cause and fix it, I have been setting the RF Delay to 200ms. Anything less that that isn't quite enough to mask the burst. DEXP being on or off had no effect.

Hopefully later this week I can crack the PC, remove the sound card and remove all the SB software. I may try it a step at a time by trying it after
I remove parts and software. Actually, I have already installed another SSD drive with a fresh copy of Windows 10 with no software loaded.
It is easy enough to boot into that. I can install a fresh copy of Thetis and start from the ground floor. I loaded Windows on that drive with
the intent of troubleshooting this problem.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Tue Apr 20, 2021 4:55 pm

Ok I got up early this morning. Drank an energy drink and made some tests. On a drive devoid of anything other than Windows 10 pro I loaded Thetis 2.8.11. Just out of curiosity I loaded the 32 bit version. I plugged the microphone into the motherboard soundcard input and my amp is running out of the soundcard output. I removed the Sound Blaster card. Adjusted my sound at my normal listening levels and triggered the MOX. The sound bump was there. Exactly the same as when running my normal setup. I also tried triggering the MOX with both the mouse and spacebar on the keyboard. Identical results.

Plus I set up and ran VOX. No bump at all. I set the RF Delay to 70ms. The bump was there ,but not as pronounced. At 200ms it was gone.

Unless there is something particular to the motherboard, I believe that what I have is a more behavioral issue than an equipment or software issue. I'm in my late 60's and have diminished hearing. I would guess I'm listening at a louder level than most. I did lower the volume to a point I think the average person would use. When I triggered the transmit that burst was barely noticeable on the screen. I going to make some contacts
and see if they notice anything.

I'm not finished yet.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Noise Burst on initial transmit

Postby K1LSB » Tue Apr 20, 2021 8:35 pm

Ruben,

Your observations reinforce my firm belief that the mic is picking up audio from the speakers. I also have poor hearing so I turn up my speaker volume, which of course means there's more audio coming out of the speakers for the mic to pick up, which exacerbates the burst on transmit.

I noticed the exact same problem with my Flex-6300 running SmartSDR a couple of years ago, so I asked a local ham friend to try an experiment on his Flex 6400M running DogPark (a Mac interface for the Flex 6000 series radios) and he saw the same phenomenon - the mic is picking up latent audio coming from the speakers. So the problem exists in Anan and Flex radios, it's just a matter of degree, based on how loud the speakers are turned up, how sensitive your mic is and how close the mic is to the speakers.

The problem is caused by latency in the audio that's coming out of the speakers. If the speakers were somehow instantly muted on PTT then the problem would disappear (and as I've noted earlier, if you silence the speakers and use headphones the problem also disappears).

The latency can possibly be reduced, but it's unlikely that it can be eliminated without modifying the software to instantly mute the speakers on PTT regardless of any audio that may still be in the output buffer....so the best software solution would be to take the following actions whenever the PTT is pushed: 1) mute the speakers, and 2) clear the buffer.

Are there any programmers out there who could possibly make those 2 hopefully simple changes? Just asking..

Mark
AB2EZ
Posts: 99
Joined: Sun Apr 09, 2017 2:29 pm
Location: Princeton, NJ

Re: Noise Burst on initial transmit

Postby AB2EZ » Wed Apr 21, 2021 3:10 am

Why not place a relay between the microphone and the preamplifier/sound card, so that the microphone is disconnected when receiving? Use the PTT output of the ANAN (the same output as is used to key an external RF amplifier on) to activate the relay between the microphone and the preamplifier sound card... to connect the microphone to the preamplifier/sound card on transmit.

Stu
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Noise Burst on initial transmit

Postby K1LSB » Wed Apr 21, 2021 9:09 am

Stu,

Good thinking, but putting a disconnect relay on the mic won't help because the problem is caused by delayed audio still coming out of the speakers after the PTT is pressed.

The solution would be for a relay to disconnect the speakers on PTT.

Mark
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Wed Apr 21, 2021 1:26 pm

AB2EZ wrote:Why not place a relay between the microphone and the preamplifier/sound card, so that the microphone is disconnected when receiving? Use the PTT output of the ANAN (the same output as is used to key an external RF amplifier on) to activate the relay between the microphone and the preamplifier sound card... to connect the microphone to the preamplifier/sound card on transmit.

Stu


That is a great idea! The best way for me to accomplish this is to place a switch or relay between my PC and audio amplifier. When I pressed the PTT it would kill the audio stream to the amp. No sound No problem. This would be easy enough to accomplish. Thank you for the idea.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Wed Apr 21, 2021 2:46 pm

Has anyone actually bothered to prove that it is a late shut off of RX audio that is causing the problem?

It could easily be the audio that is already in the TX audio buffer pipeline emptying out first that could be the problem.

Or it could be a combination of both factors.

I have finally been able to reproduce the problem here. I realized that the way I set up my radio for phone (SSB, AM) is to adjust the S-ATT and the AGC gain in order to get a quiet RX signal. I also run NR2 all of the time. Together this makes for a nearly FM-like experience. And while I use some big, powerful speakers (JBL powered studio monitors), I don't run a lot of volume. So I was not seeing the problem.

However, when I turn off NR2, turn down my S-ATT to zero, and jack up my AGC gain so that I get a ridiculous noise floor, I can duplicate the problem. And of course it is worse with 60ms of VOX look-ahead turned on. That said, I do not get ridiculous 500W spikes as my mic gain is set for a 2" distance from mouth to mic. I have to literally pick my microphone up off of the desk and hold it next to a speaker to get that to happen.


So what we really need here are some careful measurements with an oscilloscope. Unfortunately I can't do that because my station is run remotely. I have no way to trigger on PTT on a scope and have the other scope channel look at the receive audio coming out of my speakers two floors above where the radio hardware is located. And that will only tell us if the audio persists past the assertion of PTT. Then we have to find a way to test how much "garbage audio" is in the TX pipeline (I have some ideas, but one test at a time).
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Wed Apr 21, 2021 7:38 pm

w-u-2-o wrote:However, when I turn off NR2, turn down my S-ATT to zero, and jack up my AGC gain so that I get a ridiculous noise floor, I can duplicate the problem. And of course it is worse with 60ms of VOX look-ahead turned on. That said, I do not get ridiculous 500W spikes as my mic gain is set for a 2" distance from mouth to mic. I have to literally pick my microphone up off of the desk and hold it next to a speaker to get that to happen.


I use NR2 nearly all the time. I have S-ATT set at 15. I set AGC on 80 and make minor adjustments when necessary. I am normally on Am or SSB. Your comments made me remember. The day that people were complaining about the noise burst on the transmission my noise floor was rather high. There were some fellows in the QSO that were very weak and I had to make adjustments to hear them. I normally do not have to make those adjustments. The Thetis software allows for very quiet listening. Also I may be running the mic gain too hot for the microphone. When speaking I do have the microphone very close to my mouth. This is why I say the issue, at least for me, could be behavioral.
That being said. I have not had anyone mention the noise burst before this time. Believe that the fellows I QSO with will mention any abnormality on an audio signal. As you have mentioned, I try to set it up to have almost like FM communication. That hissing noise drives me batty. I already have barely bearable tinnitus.

Unlike the other person I have never had that burst cause a 500 watt burst of power. I watched it yesterday and I did notice a 10 watt jump before settling to 0. I am running an Ameritron AL-80b. I will try some different combinations of settings. That with RF Delay set to the default of 10ms.
Thank you.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Noise Burst on initial transmit

Postby rbduck » Sat Apr 24, 2021 1:08 pm

I believe I have it complete. I have my audio set up where when I push PTT there is no noise burst without adding RF Delay. I have S-ATT at 15 and AGC GAIN at 80. This adjustment varies depending on propagation. I use NR instead of NR2. That noise that NR2 makes is just too distracting.
I try to adjust that out to no avail. I have a tendency to turn the volume up when not necessary. I am training myself not to do this. It has to with the tinnitus. I also find the sound is better if I don't run the ANF enabled. I do enable the receive Equalizer. I can contour the sound to my hearing and knock some frequencies down that feed the noise burst. I have made contacts and I'm told there is no noticeable noise burst at the beginning of my transmissions. I will probable continue to make minor adjustments.

Until Scott mentioned this, I don't know why I didn't think of it earlier. I know we all like quiet when there is no one speaking. That hiss drives me up a wall. I love the ability to reduce noise in Thetis. There is almost no noise adjustment that you can't make to eliminate unwanted noise and make voices appear natural to the ear. I hope this will help someone in the future that might notice a noise burst. Look at your basic noise adjustments FIRST.

Thank you Scott for all your help. And thanks to all that participated in the discussion.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Noise Burst on initial transmit

Postby w-u-2-o » Sat Apr 24, 2021 5:01 pm

Ruben--glad you got it sorted. I recommend you only add attenuation until you see the SNR of the largest signals on the band start to drop. S-ATT mode makes that easy.

However, what do you mean by "the noise NR2 makes"? While all NR algorithms change the temporal and spectral characteristics of the audio, they don't create new noise.

Meanwhile, it seems pretty clear now that the RX and TX audio pipelines are not being managed appropriately.

Return to “Thetis”