P25 below 146MHz on a K split Astro Spectra?

[Napalm]

As above really!

Will a K split Astro Spectra do P25 below 146MHz.

I've programmed in a spot freq (144.675) and the radio transmits what sounds like P25, but I'm not getting a decode on either my EFJ 5100 or my Astro Saber 3...

Are there any hacks I can do?

[n7maq]

As above really!

Will a K split Astro Spectra do P25 below 146MHz.

I've programmed in a spot freq (144.675) and the radio transmits what sounds like P25, but I'm not getting a decode on either my EFJ 5100 or my Astro Saber 3...

Are there any hacks I can do?

Some radios will do it, and most will not. I have heard that the low, and mid powers units are more likely to work. I have also heard that if you have newer firmware it helps, but I have not been able to confirm that one. I have a railroad Spectra that works IMBE below 146.00.

I have a Quantar on 145.490, so I feel your pain..


Jim

[Napalm]

Dang it.

Well the specs are:
Model: T04KKH9PW9AN
Host: 11.10.00.00
DSP: N08.02.06
Flash: 599108-145E00-0

I've had word there's a hack which involves deviation and what not.. but with no test gear I guess I'm SOL

[JohnG]

I've tried this on both the 146-162 mid-power Astro Spectra and Astro Spectra Plus. I have found that both of these radios will receive IMBE just fine below 146.000. They will transmit and receive analog just fine below 146.000. I have not been able to get them to transmit IMBE successfully below 146.000. If I recally correctly, the radio transmits an unmodulated carrier when IMBE TX is selected.

[Napalm]

Well my K split transmits IMBE/P25 fine below 146, but nothing will decode it... I presume the deviation is out of whack below 146? Or something isn't JUST right.

[jmr061]

Yeah I would say something is up. Generally they will receive out of band but not TX. I may be selling some range 1 parts from a mid power spectra (136-162). Keep you eyes open for that.

Jason

[Tom in D.C.]

After reading these posts, curiosity got the best of me so I set up
a test on 144.35 simplex P25 using a K split XTS5K and a VHF Thales P25.
(Note that these are portables and not the mobiles covered above.)
The radios talked to each other just fine. The XTS is v11 Host and
DSP and the Thales is a 5.13 version.

[mancow]

Maybe I'm missing something here but I think the issue is that the mobiles are meant to cut off at 146. The XTS series is a full 136-174 factory enabled bandsplit radio so it will perform correct across that entire range. If I'm tired and talking out my arse then please ignore me.

After reading these posts, curiosity got the best of me so I set up
a test on 144.35 simplex P25 using a K split XTS5K and a VHF Thales P25.
(Note that these are portables and not the mobiles covered above.)
The radios talked to each other just fine. The XTS is v11 Host and
DSP and the Thales is a 5.13 version.

[Tom in D.C.]

Casey,

You're right, of course. I was, as I said, just curious to see if/that the
portables would work on the low end.

[mancow]

Ok, I wasn't sure if I was reading it right. I guess the issue is getting that kind of bandwidth out of a high powr PA. That's what I was told regarding the EFJ mobiles anyway . The low power unit is full band capable but the high power stops around 146 or 148.

Casey,

You're right, of course. I was, as I said, just curious to see if/that the
portables would work on the low end.

[Astro Spectra]

While the VCO locks a ways out past the band limits the radio is unable to interpolate the proper high and low deviation digi pot settings to get the correct modulation from the preset values stored for each test frequency and the modulation fidelity suffers.

It is possible to directly observe this effect if you look at the modulation on a service monitor with the 'scope connected to display the deviation (+/- kHz vert time horz) in single sweep one shot mode. When the radio is operating correctly the trace for each P25 symbol is flat and the symbols can be distinguished as +/- 600Hz and +/- 1.8 kHz. It is a bit of an effort to get a good display as the signal looks random.

When the modulation goes wrong the deviations expand or contract and the symbol elements often slope or get wiggly… it easier to show than to explain. Even without a proper service monitor you can se this effect using by feeding the FM discriminator output of a good receiver into a scope, then comparing known good signals with the UUT. You can fiddle the tuning values to get the radio to work on a specific out of band frequency but then it will be wrong everywhere else! The same problem exists with analog FM, you’ll find the deviation either falls off or increases out of band as the radio is no longer able to set the correct modulation signal amplitude for the VCO.

The traditional Motorola FM system is open loop in that it relies on the stored tuning values and can’t tell how much deviation is actually being produced. The VCO sensitivity (+/- mV of modulation in vs +/- kHz deviation out) varies from unit to unit, varies over the tuning range, and varies with age. This is why tuning values are critical and need to be set up properly for each radio and adjusted periodically during the radio’s service life if good P25 performance is to be maintained.

While the PA performance will drop, this is mainly in the form of roll off in output at the high end because of the output LPF.

By the way Aeroflex have a whizzy P25 test set that has a special feature to look at the RMS symbol error which is a far more sophisticated measure of modulation fidelity. It compares what the transmitter under test is producing against what it should be producing by demodulating the transmitted data then producing internally a perfect reference signal.

[Napalm]

So - to sum up.

I'm SOL?

[motorola_otaku]

No and yes.

The hardware here isn't your issue, it's the codeplug. What the above poster is referring to is TX Deviation Compensation, which is an adjustable setting under the RSS Service menu and is one of the things you set when yo do a full alignment. The codeplug was set up from the factory for 146-174 MHz, ergo, it only contains softpot programming to regulate that range. Obviously you know the radio works south of 146, because it transmits analog (probably at perfect 5 kHz deviation) and receives IMBE there. IMBE RX and analog TX (unless you're using DPL) are not regulated by deviation compensation, however; digital TX is. If you had a codeplug with a TX Deviation Compensation tuning softpot at, say, 144.025 MHz, you could properly set it and have full IMBE TX capability. Old Astro Lab B03.04 has this capability; you can add, change, and remove tuning softpot frequencies in VSELP Astro Saber codeplugs. Obviously this does you no good because old Astro Lab no speaka da IMBE.. or Astro Spectra, for that matter. So to summarize, it can be done but the tools and methods to do so have not and probably will not be publicly disclosed. ;p

[Napalm]

Thanks mate.

I'll just keep an eye out for another J splitter.

[d119]

Er, What if you just dumped a J-split s-record into it? Wouldn't that give you the necessary softpots? Granted you'd need to do a complete realignment, but you know...

[n7maq]

Er, What if you just dumped a J-split s-record into it? Wouldn't that give you the necessary softpots? Granted you'd need to do a complete realignment, but you know...

That will not work on VHF.

Jim

[motorola_otaku]

Er, What if you just dumped a J-split s-record into it? Wouldn't that give you the necessary softpots? Granted you'd need to do a complete realignment, but you know...

That will not work on VHF.

I've never tried, but I think Seth (sglass) said he'd done it and got IMBE TX below 146.

In any event, I see no reason why it wouldn't work if you also swapped VCOs and preselectors with a 136-162 split analog Spectra. The VHF Astro RF board covers both ranges.

[Napalm]

Ugh. Be easier to buy a J-split to be honest. I don't have test gear to do that myself... but the info might be useful to someone else.