Re: Adding external relay control two V2/V3
My version turned out like this
--
Gyula HA3HZ
?
|
I found in back to back tests that wsjt-x has a much lower CPU load and much less lag than JTDX.
I use JTDX on my home computer with my Hermes Lite 2 thanks to its support for TCI, but not not on a system is that is CPU limited.
--
73
??? Bob? KD8CGH
|
On Sat, Nov 2, 2024 at 03:16 PM, Dino wrote:
All I can say is that I’ve tried these recommendations and the lag still creeps up after a while and the pop up message returns.?
I'm glad that other people have opinions, because I think we should look for the possibility of acceleration in the radio.
If there is such a possibility.
?
--
Gyula HA3HZ
|
All I can say is that I’ve tried these recommendations and the lag still creeps up after a while and the pop up message returns.?
|
It is difficult to find a description for jtdx, and what is there was written 6 years ago.
Fortunately, the description already addressed the slow PC problem and included an automatic RX filter.
The next step is the LAG problem, which is only visible with jtdx, because when the number of stations in the audio track exceeds 30, there is little time for decoding.
How little time is left can be read in the description at
Out of the 15 second period, 12.64 seconds are the transmission time, leaving 2.36 seconds for decoding.
It is understandable that with a large number of stations this can only be done with a fast PC.
What can someone using the RPi do?
?
My experience shows that the Lag value is at most -0.55 seconds / 0 stations or -0.50 / 2 stations.
This is constantly changing.? So our device takes 1.8 seconds to operate.
There is not much we can do:
?
- set the decoding functions to reduce LAG. (some settings need to be changed, experiment)
- select option 3 from the Autoseq options and do not forget to set the Auto RX frequency filter accordingly.
- since there is little possibility to reduce the signal between the radio and the jtdx input, keep the IF and volume values ??low.
- another option is to reduce the receiver BW
- you need to learn to use the auto RX Filter, check from time to time that when you want to monitor the band, the filter is turned off. In the receive window, the activated state of the filter is gray.
- I think that the time between the end of the transmission and the start of the next period is 15 - 12.64 = 2.36 sec.
When I excluded the other stations with the RX filter, I saw the best Lag value of -0.55 / 0.
The remaining time is 2.36 - 0.55 = 1.81 seconds that our device uses, this is the time lost from decoding.
?
I am curious if this time could be shortened?
What does our device use the 1.81 seconds that it loses from decoding for?
Does anyone know the answer?
--
Gyula HA3HZ
|
Me I find the Arduino nano is more than enough for building a VFO for either DDS or 5351
plus? the needed encoder and display.
?
?
--
Allison
------------------
Please use the forum, offline and private will go to bit bucket.
|
In my list of future work is a replacment for the Si5351. It is tragic that all our radio designs depend on one proprietory chip.
On Thu, Oct 31, 2024, 12:54 AM Richard Neese via <n4cnr.ham= [email protected]> wrote: the unit I am making iof for 2m or 6m right now so I have to use the si5351
On Wed, Oct 30, 2024 at 10:32?AM Ricardo Suarez via <elu9da= [email protected]> wrote:
The Raspberry pi pico.?
Programmable i/o pins can generate frequencies from 0 to 30 mhz
without use SI5351 or AD98xx dds even in quadrature for tayloe
mixers....
Rick LU9DA
El 30/10/2024 a las 13:38, Richard
Neese via 别蝉肠谤颈产颈ó:
What is the preferred?board between these 2 for
building a vfo ?
|
VE6LX,
Your original posts got me to look at the K1EL keyer documentation - for me that was a great introduction to the inner workings of electronic keyers and what they can do to help with real-world hardware. But - I think that the RPI has the headroom to do the electronic keying task on-board.
What I would like to do is get a two channel oscilloscope and put one probe right on the terminal of a straight key and watch the key make contact ... on the other channel I'd like to see the output of the transmitter. And then have my fancy imaginary o'scope tell me the time difference between the two. Of course the time difference will include software delays due to polling, handling that TR switch you mentioned and who knows what else ... but we should be able to measure what the timeline really is, and decide if its a real problem or not. If it's a few milliseconds maybe padding lead-in time could be the answer.
toggle quoted message
Show quoted text
On Thu, Oct 31, 2024 at 12:32 PM, VE6LX wrote: Mike I believe this is precisely what I have been suggesting from my first
post. It is not the CW keyer per say rather the need to turn on the
transmitter X ms prior to the first CW character reaches the Transmitter input
port (Lead-In Timing) and on the Exit the (Tail Timer) time Y ms after the
last CW character prior to shutting off the transmitter
--
Mike KB2ML
|
Bob, it's in Farhan's 32-bit repository, here ...
toggle quoted message
Show quoted text
On Thu, Oct 31, 2024 at 01:18 PM, Bob Benedict, KD8CGH wrote:
Mike
What dev was the July 8th commit to? I don't see it in JJ's timeline.
--
73
Bob? KD8CGH
--
Mike KB2ML
|
I have had my Sbitx for about 6 months. Original owner. I upgraded to 64 bit firmware? including the tool box app. Have made numerous ft8 contacts but have too many radios so I am going to sell. Asking $350 which includes domestic priority mail? shipping. (Gigabits asks $429). In addition to radio and microphone I am including 6 amp power supply. Glad to share photos if interested
|
Mike
What dev was the July 8th commit to? I don't see it in JJ's timeline.
--
73
??? Bob? KD8CGH
|
Mike I believe this is precisely what I have been suggesting from my first post. It is not the CW keyer per say rather the need to turn on the transmitter X ms prior to the first CW character reaches the Transmitter input port (Lead-In Timing) and on the Exit the (Tail Timer) time Y ms after the last CW character prior to shutting off the transmitter and another short time Z ms to create a signal to turn on the receiver.
This technology really goes back a ways in history as the "collective we" have forgotten how spoiled we (collective we) have become with the advent of transistorized radios. Now in the Age of the computer or "Software Defined Radio" all perviously insignificant timing events now have to be coded into the operational software. The physical amount of time may be very short but remember in the Computer Driven World even the smallest amount of dead air time has to be part of the Operational Software Code.
|
Re: sbitx, turn off TX audio.
I thought CW mode showed the Side Tome too. Not at the radio to test that.
toggle quoted message
Show quoted text
Okay, I tried it and that does work well.? Ran with the default 30 and 0 then switched to JS8Call.? In the sbitx app you have to be in FT8 mode to see the Sidetone menu.
Thanks
|
Re: sbitx, turn off TX audio.
Okay, I tried it and that does work well.? Ran with the default 30 and 0 then switched to JS8Call.? In the sbitx app you have to be in FT8 mode to see the Sidetone menu.
Thanks
|
Re: sbitx, turn off TX audio.
On Thu, Oct 31, 2024 at 08:52 AM, Kevin wrote:
I was using mine at a campsite with quiet hours and every time that I muted the volume through the OS it would unmute during transmit.? I muffled the speaker slots with a towel and quickly muted every time JS8Call transmitted.?
I could have opened it up and unplugged the speaker but it was dark, windy as heck, and I didn't want to accidentally drop something.? I'll probably wire a speaker kill switch but for now I've found that just plugging in headphones does the trick.
?
?
Just set the sidetone to 0.
?
?
|
Re: sbitx, turn off TX audio.
I was using mine at a campsite with quiet hours and every time that I muted the volume through the OS it would unmute during transmit.? I muffled the speaker slots with a towel and quickly muted every time JS8Call transmitted.?
I could have opened it up and unplugged the speaker but it was dark, windy as heck, and I didn't want to accidentally drop something.? I'll probably wire a speaker kill switch but for now I've found that just plugging in headphones does the trick.
|
Re: sBITX 64Bit Pi Image Available
#sBitx
I'm glad that I asked.? Thank you.? I'll refrain from switching.
|
Re: NBFM (Was: AM on sBITX?)
I/Q is one way. Here is a simpler way:
1. The signal is essentially AC.
2. Mark the exact time at which the signal transits from positive to negative.
3. This measures the time period of every crest and trough of the signal. 4. When the signal frequency is low, the time periods are lengthier. When the signal frequency is higher, the time period is smaller. 5. The time period is our demodulated signal. - f
The frequency differentiation method would be able to approximate the phase difference by using consecutive I samples. But the demod product would probably be pretty crappy at best.
I think that the transform method should at least be able to simulate a phased Q sample which may be a bit better as a starting point. I messed with this a while ago when I was offloading a synthetic IQ stream to SDRAngel.
it almost worked, but I lost interest in the challenge..?
|
Re: NBFM (Was: AM on sBITX?)
Hi Rafael,
?
Thank you for the links!
?
I have had a setback with my DE, so I cannot try this until I repair the low-power problems.
?
73
Evan
AC9TU
|
Re: NBFM (Was: AM on sBITX?)
Hi Evan,
I maintain this fork of it (just compilation fixes):
and packages for arm64 (run on any 64 bit image):
- Rafael
toggle quoted message
Show quoted text
On 10/31/24 10:00 AM, Evan Hand wrote: Hi Rafael,
Thank you for the feedback.? I will look into it.
73
Evan
AC9TU
|
HA3HZ
Bob Benedict, KD8CGH
Dino
VE6LX
Kevin