OP25 CC timeout crazyness

[zapman987]

1555872541.132048 control channel timeout
1555872541.131991 set control channel: 858.987500
1555872542.223717 control channel timeout
(Truncated)
1555872547.685944 control channel timeout
1555872547.685887 set control channel: 859.437500
1555872548.779339 control channel timeout
(truncated)
1555872554.239552 control channel timeout
1555872554.239501 set control channel: 860.437500

What the heck is going on here? The site changed CC this past friday, and Ive been seeing this after I moved it to the new CC. Do I need to move it back, or are they just having system errors?

 

[boatbod]

1555872541.132048 control channel timeout
1555872541.131991 set control channel: 858.987500
1555872542.223717 control channel timeout
(Truncated)
1555872547.685944 control channel timeout
1555872547.685887 set control channel: 859.437500
1555872548.779339 control channel timeout
(truncated)
1555872554.239552 control channel timeout
1555872554.239501 set control channel: 860.437500

What the heck is going on here? The site changed CC this past friday, and Ive been seeing this after I moved it to the new CC. Do I need to move it back, or are they just having system errors?

What's going on is that op25 is not seeing the control channel. Suggest you enable the fft plot (Press key #1 from terminal) to see if there is indeed any control channel active on the tuned frequency.

 

[zapman987]

Ya its set. I dont even see it blink when it does it (have been watching). Its doing this intermittent, not constant, and since the whole system changed CCs, I think maybe its searching to see if the old primary is online maybe? I dunno. Its working fine otherwise (minus my trunk issue from my other post), just saw this and figured to ask on it.

 

[boatbod]

Ya its set. I dont even see it blink when it does it (have been watching). Its doing this intermittent, not constant, and since the whole system changed CCs, I think maybe its searching to see if the old primary is online maybe? I dunno. Its working fine otherwise (minus my trunk issue from my other post), just saw this and figured to ask on it.

Do you have two control channels listed in trunk.tsv? Depending how it's configured it may try to cycle between them.

 

[zapman987]

Do you have two control channels listed in trunk.tsv? Depending how it's configured it may try to cycle between them.

Nope. Only 1. See my trunk file from the other post.

 

[boatbod]

Nope. Only 1. See my trunk file from the other post.

What does the mixer plot (#5) look like? If it's not well centered it's possible that it could loose lock.
ETA: from your other thread I see that it's giving you tuning errors:

p25_framer::rx_sym() tuning error +1200

I recommend you tweak your ppm (-q) to center the mixer plot and eliminate the +1200.
As an aside, are you using a tcxo dongle?

 

[zapman987]

Sorry didnt realize a reply on this thread. Its nicely well centered. I quit seeing the errors after I posted this so I think it was on the system side.

Ive tried messing with the -q and never really understood it. Right now q is off as an option and working fine with the manual adjustment.

Na this is a RTLSDR v3 silver

 

[boatbod]

Right now q is off as an option and working fine with the manual adjustment.

That right there is the problem. Manual correction comprises a fixed offset regardless of tuned frequency.

PPM correction is an offset derived from the tuned frequency; the acronym stands for "parts per million". In other words, a 1ppm correction corresponds to 1Hz increment per 1Mhz of tuned signal. At typical trunked frequencies - say 800Mhz - a 1ppm correction is an 800hz offset.

When you make fixed adjustments they may work fine on the control channel, but tuning to a voice channel some number of Mhz away may not work so reliably. The general rule of thumb is to adjust out the coarse tuning errors using ppm correction (-q) and then dial in the final +/- 300 using fine tuning. This is of course less of an issue with modern temp compensated crystal oscillators (tcxo) as they drift far less than the old uncompensated dongles, but the theory still applies all the same.

 

[zapman987]

If I understand this as a whole (funny as I understand all the individual words perfectly), I need to do q 1.5. Which does nothing it seems. Still need to do a manual 1200.

 

[boatbod]

If I understand this as a whole (funny as I understand all the individual words perfectly), I need to do q 1.5. Which does nothing it seems. Still need to do a manual 1200.

I usually specify ppm correction as whole integers. Try "-q 2" and see if that helps. You can also do the negative version if you need to correct the other way. e.g. "-q -2"

 

[zapman987]

Ya positive 2 is what I had used before i removed it (didnt seem to do anything) Will try -2 and report back.

 

[boatbod]

Ya positive 2 is what I had used before i removed it (didnt seem to do anything) Will try -2 and report back.

Watch the mixer plot (#5) and see which way the signal peak moves. Ideally you want it centered on the central tuning line.

 

[zapman987]

plot 5 is dead centered. Peaking at -20, with floor at -80. Not really moving sideways at all, moves up and down a little. Mixer balance is under 100, sometimes hitting 0. Is this something decent.

Messed around with q again and -1 seems to be good for picking up a signal with minimal adjustment. Plot 5 is still the same numbers. Im still checking this setup though as I dont think I'm getting audio here (and was on the old setup). But this might be a red herring with all my post crash stuff per my other thread.

 

[boatbod]

Mixer balance values around 20 or so are generally desirable. That said, the measurement is easily confused if there are nearby signals which make the plot lopsided. You can also look at the datascope (#4) which should have nice well defined and vertically centered eyes. If the plot is offset significantly up or down that would indicate a frequency offset.