Добро пожаловать, Гость
Логин: Пароль: Запомнить меня
*****PLEASE, CHECK ALL PREVIOUS TOPICS, BEFORE CREATING A NEW ONE! DUPLICATED TOPICS WILL BE DELETED!*****
  • Страница:
  • 1
  • 2

ТЕМА: RX AGC ver. 1.3.0 and 1.3.1

RX AGC ver. 1.3.0 and 1.3.1 13 Окт 2019 18:16 #1

  • DL5KH
  • DL5KH аватар
The RX AGC works not good.
AGC is pumping.
Rise time is extremely short. :evil:
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 14 Окт 2019 07:57 #2

  • Rome
  • Rome аватар
  • Вне сайта
  • Quod Licet Jovi Non Licet Bovi
  • Сообщений: 534
  • Спасибо получено: 220
We haven't changed anything in AGC operation neither for RX or TX.
If you operate in SSB, use Long preset.
Роман, Roman
Expert Electronics
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 17 Окт 2019 20:55 #3

  • DL5KH
  • DL5KH аватар
Realy no changes ?
1.2.0 final:

AGC1_2_0.jpg


1.3.0 final:

AGC1_3_0.jpg



I don`t like the "hang" feature
The "attack" & "release" time is definitive not the same in real life.
Последнее редактирование: 17 Окт 2019 21:00 от DL5KH.
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 18 Окт 2019 09:17 #4

  • g0cgl
  • g0cgl аватар
  • Вне сайта
  • Erik EI4KF / G0CGL
  • Сообщений: 659
  • Спасибо получено: 261
I have to say, this is still the most major fault of 1.3 and is very different to 1.2

Expert Electronics, please ignore your programmer just for an hour or so and make some tests. AGC pumping is prevalent in 1.3 whereas it was absent in 1.2

Try a digital mode for example. You will see AGC kicks in and quickly lowers the gain. But as the signal(s) decay, the AGC relaxes and allows the gain to go up again. The result is that, between signals, the volume swells. This fault is exacerbated by not being able to set AGC Threshold any longer. So, in addition to the problem, AGC is also being actioned on noise when it should be set above that threshold so that it actions only on signals. The combined result is not good. You can even see it in the meter - use DIGU as an example and on TX the AGC will over-shoot to well above 0dB before settling at 0dB. This is real over-shoot and not just a characteristic of the meter as evidenced by listening to the receiver. On other modes, the meter is not quick enough for a visual representation but poor AGC action is still there. Then load 1.2 and note the difference.

Erik EI4KF.
Последнее редактирование: 18 Окт 2019 09:25 от g0cgl.
Администратор запретил публиковать записи гостям.
Спасибо сказали: G4KVT

RX AGC ver. 1.3.0 and 1.3.1 18 Окт 2019 11:18 #5

  • F1DJY
  • F1DJY аватар
  • Вне сайта
  • Сообщений: 33
  • Спасибо получено: 6
I confirm what you say erik, and it is very unfortunate.
The analysis you make is very accurate and the pumping effect is present in certain conditions and it is unpleasant, even in SSB on "long"
- Did you see what I reported about "transmit audio adjustments":
Jean-Marie F1DJY
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 18 Окт 2019 11:52 #6

  • KV5J
  • KV5J аватар
  • Вне сайта
  • Сообщений: 24
  • Спасибо получено: 3
As a SSB contester. Having the agc threshold was very important go me. I need to control when the agc activates. When I'm working a pileup the last thing a want is everyones audio to be equal. I most important thing is rates. I want the strongest stations to be strong. If all the stations are the same or near the same audio that creates mush. Other radios suffer the same problem. Those that have added an agc threshold adjustment are better contesting radios.

Keith, KV5J
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 19 Окт 2019 06:32 #7

  • Gian
  • Gian аватар
  • Сейчас на сайте
  • HB9FAR
  • Сообщений: 36
  • Спасибо получено: 5
Confirming your analysis Erik ! On 2 mt with the mb1 the only way to work weak signals is switching off the ALC..
HB9FAR
Администратор запретил публиковать записи гостям.
Спасибо сказали: G4KVT

RX AGC ver. 1.3.0 and 1.3.1 19 Окт 2019 12:49 #8

  • g0cgl
  • g0cgl аватар
  • Вне сайта
  • Erik EI4KF / G0CGL
  • Сообщений: 659
  • Спасибо получено: 261
Hello Gian - yes switching off AGC is the "workaround" I describe in the UMA. But it is not a good solution. You need to be careful for the receiver and your ears in the presence of a sudden strong signal or static crash. If 1.3 has come to having to do this, I would rather do as I am now and that is staying with 1.2 until such time as any remedial work is done by EE on the issues.

Erik.
Администратор запретил публиковать записи гостям.
Спасибо сказали: Gian, G4KVT

RX AGC ver. 1.3.0 and 1.3.1 19 Окт 2019 12:55 #9

  • Gian
  • Gian аватар
  • Сейчас на сайте
  • HB9FAR
  • Сообщений: 36
  • Спасибо получено: 5
82F70F6D-A202-46B0-931B-63B4A2A65DCB.jpeg

..effects of agc pumping on msk144
HB9FAR
Администратор запретил публиковать записи гостям.
Спасибо сказали: DL5KH, G4KVT

RX AGC ver. 1.3.0 and 1.3.1 19 Окт 2019 23:33 #10

  • DK3UD
  • DK3UD аватар
  • Вне сайта
  • Сообщений: 24
  • Спасибо получено: 8
Hi everybody,

I think, some people are mixing up AGC for receiver, AGC for transmitter and ALC!
There seems to be some confusion.

In version 1.3.1 Beta 1 (for MB1) AGC for receiver is working well. The AGC-time-settings
for "Attack", "Decay" and "Hang" can be found in "Options / Device / DSP / AGC".

The AGC-level-control or AGC-threshold can be adjusted with the RF-Gain
knob. And this is working very well with me. Just try it!

Best 73
Manfred
DK3UD
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 20 Окт 2019 07:25 #11

  • g0cgl
  • g0cgl аватар
  • Вне сайта
  • Erik EI4KF / G0CGL
  • Сообщений: 659
  • Спасибо получено: 261
No there is no confusion at all. If you do not make the comparison between 1.2 and 1.3 then you are the one to be "confused". ALC is a limiter control whilst AGC is a gain control. Nobody that I have seen has made any remark about a limiter problem.

"Pumping", as it is colloqually known, is a AGC fault caused, in this case, by some error in the software programming and which is noted by receive faults. There is an over-arching problem with 1.3 that also affects transmit. Let those who have these serious problems make their case for EE to investigate, if they so desire, to help the customer base. It does not seem like too much to ask because, as Keith KV5J so rightly says, currently 1.3 is not as good as it could be for serious contest working. It should not be too hard to isolate and correct so we can all be as happy as you are. After all, EE diligently worked with users to improve the DSP in 1.3 and yes it took some hours of testing, but was done. This can be too.

Erik.
Последнее редактирование: 20 Окт 2019 08:18 от g0cgl.
Администратор запретил публиковать записи гостям.
Спасибо сказали: KV5J

RX AGC ver. 1.3.0 and 1.3.1 20 Окт 2019 08:57 #12

  • iw7dmh
  • iw7dmh аватар
  • Вне сайта
  • Сообщений: 87
  • Спасибо получено: 36
Hello,

during the last two weeks, following the T30GC, A82X and 5K0K expeditions, I often found myself having to deactivate the AGC to avoid the pumping effect described by Erik. I had not linked this issue to the new version of ESSDR mostly because I used the 1.2 for a few months and usually I don't use the PRO2 on SSB and DIG modes.
I also noticed the threshold control is missing in the 1.3 and in my opinion it is a serious lack for a SDR receiver.
I think the AGC threshold adjustment should be a control to be placed next to the RF gain control. In my previous experience with the Flex 6000 the only AGC-T control was enough to solve all the AGC problems without having to set the attack, decay and release timeouts. Actually they were totally missing in SmartSDR.
Anyway I think it is one of the drawbacks that Expert Engineers have to solve after the porting from Qt4 to Qt5. Even if it seems nothing has been changed, the effort for the code porting should has been great. It just takes time to fix some things.
It is a pity in the 1.3 roadmap there are no provisions to work on the AGC feature, but Expert designers should reconsider it and find a slot in their timesheet to solve it definitively.

73', iw7dmh
Последнее редактирование: 20 Окт 2019 08:58 от iw7dmh.
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 21 Окт 2019 08:16 #13

  • Rome
  • Rome аватар
  • Вне сайта
  • Quod Licet Jovi Non Licet Bovi
  • Сообщений: 534
  • Спасибо получено: 220
Hi, everyone.
For correct AGC operation, we recommend setting identical Hangtime and Decay parameters.
If you want to achieve leveled amplitude than Hangtime should be higher than decay.
Роман, Roman
Expert Electronics
Администратор запретил публиковать записи гостям.

RX AGC ver. 1.3.0 and 1.3.1 21 Окт 2019 11:05 #14

  • g0cgl
  • g0cgl аватар
  • Вне сайта
  • Erik EI4KF / G0CGL
  • Сообщений: 659
  • Спасибо получено: 261
:wall:
Последнее редактирование: 21 Окт 2019 11:07 от g0cgl.
Администратор запретил публиковать записи гостям.
Спасибо сказали: digitoils, J28AA, DL5KH, KV5J

RX AGC ver. 1.3.0 and 1.3.1 27 Окт 2019 17:33 #15

  • DL5KH
  • DL5KH аватар
I have done a few tests.
The rise time of the gain reduction is too fast.
This leads to a pumping effect with static QRM.
This error affects the RX as well as TX AGC.

The AGC from 1.2.0 is mutch better.
Администратор запретил публиковать записи гостям.
  • Страница:
  • 1
  • 2