454 MHz National Grid stuff

n1zyy

n1zyy

Well-known member
After (erroneously, it seems) thinking one of them was possibly contributing to intermod, I spent a bit tonight looking at the 454 MHz National Grid Passport sites in SDRTrunk. This post is just going to be a list of what I'm seeing. Maybe that's interesting to someone?

https://www.snewiki.com/wiki/index.php/National_Grid_Passport_LTR is the wiki link.

I remain mildly intrigued by this system because I have never heard voice traffic on it, and the last confirmation of any of the talkgroups was more than a decade ago. Is it still being used, and they're just not chatty? I'm curious.

Also, we can do tables now? That's cool!

Each of these channels seems to have regular (once a second or so) data calls to a talkgroup, which seems to always be the same at a given site. I've noted that talkgroup on each frequency, though I doubt it's actually fixed to a single frequency. I don't understand the meaning of the data calls right now -- are they important telemetry, or routine control channel signaling?

I also don't understand channel numbers. Every frequency, except for one, seems to usually identify itself as being channel 1825, but sometimes it will jump around. (There's also an indication of a "Free" channel, which I'm guessing is a signal of where the next call should go, like LTR. It's not clear what these numbers mean, though -- they're too big to be regular LCNs; are they maybe tied to frequency like NXDN or something? So much to learn!)

I have bolded where the site number I observed doesn't match what is on the wiki. I would not go updating the wiki based just on what one sdrtrunk run says, though.

FrequencyNotesWiki says...
454.05000Site 015, Channel 1825. TG 21909.Site 15, Norwell
454.05625Site 003, Channel 1825. TG 21909.Site 03, Georgetown?
454.06875Site 005, Channel 1825. TG 30101.Site 05, Ayer
454.23125Site 002, Channel 1825. TG 38293.Site 02, Merrimack Valley
454.26875Site 012, Channel 1825. TG 21909.Site 12, Marlboro
454.29375Site 004, Channel 1825. TG 21909.Site 04, Unknown
454.30625Site 005, Channel 1825. TG 29973.Site 05, Ayer
454.33750Site 005, Channel 1825. TG 1301.Site 06, Danvers
454.35000Site 015, Channel 1825. TG 38165.Site 15, Norwell
454.35625Site 002, Channel 1825. TG 5397.Site 02, Merrimack Valley
454.39375Site 004, Channel 1825. TG 21781.Site 04, Unknown
454.40625Site 012, Channel 1825. TG 29973.Site 12, Marlboro
454.55000Site 004, Channel 1825. TG 5397.Not listed anywhere
454.55625Site 006, Channel 1827. TG 30101.Site 06, Danvers
454.61875Site 012, Channel 1825. TG 5397.Site 07, Malden
454.65625Site 002, Channel 1825. TG 29973.Site 02, Merrimack Valley
 
Boy, this brings back memories. Does anyone remember Trunker?

I modified the Trunker program around 2004 to decode LTR, Passport LTR, Multi-Net LTR, and LTR-Net. It ran in DOS and eventually I stopped using it due to later versions of Windows. I originally scoped out the National Grid system years ago.

One repeater at a site is the collect or registration repeater. There is no voice traffic on this repeater unless no other repeaters are free. Usually another repeater is used for the home talkgroups and a third repeater is used for roaming talkgroups. Radios roaming on the system will register on the collect repeater and get assigned a repeater on the system. The neighbor information lists the collect repeater of the neighboring sites as the radio is programmed with their home site repeaters.

It is possible you saw a unit homed on Site 5 that roamed and talked on Site 6.

I would have to setup and run sdrtrunk to see what is going on. I do not think I have any computers left where I can run the ltrdump program.

The data frame I documented is:
DCC (same as area) 2 bits
GoTo 11 bits 1 to 1791 is frequency, 1792 or 1793 and Type = 1 is band plan and neighbor info, 2047 is End of Transmission
Site ID 7 bits
Group 16 bits 65535 is CW ID
Type 4 bits 0 = voice, 2 or 4 = ?, 9 = data?, 1 = ack, 3 = home channel assignment, 6 = Unit ID, 11 = Register
Free 11 bits 1 to 1791 is frequency
Check 8 bits

The talk groups are made up of home_site-group

The system is setup with a band plan and uses direct frequency assignment for GoTo and Free. The normal UHF bandplan uses 12.5 kHz steps, so this system may be using a bandplan that sdrtrunk does not know about.

Bandplans I documented
0 800 MHz
1 900 MHz
5 450 MHz (12.5 kHz step)
6 460 MHz (12.5 kHz step)
7 470 MHz (12.5 kHz step)
8 480 MHz (12.5 kHz step)
9 490 MHz (12.5 kHz step)
11 400 MHz
13 216 MHz
 
This information is fantastic. I had come across an Aeroflex doc about Passport describing some stuff (like DFA), and then a RadioReference thread from 2004-2008 (where much of the information was posted by you). I suspect the SDRTrunk Passport decoder is incomplete. I think that also explains why everything is saying "channel 1825".

For example, I have a lot of stuff like this:

20251031 173710,PASSED,DCC:1 UNKNOWN SITE:005 CHAN:1825/0 FREE:739/0 TYP:01 TG:13717 MSG:10101100001111001000010000101001101011001010100010101110001111001110
20251031 173711,PASSED,DCC:1 UNKNOWN SITE:005 CHAN:1825/0 FREE:742/0 TYP:01 TG:21909 MSG:10101100001111001000010000101010101011001010100010101110011000110100
Click to expand...

Between your post here and some of the older RR stuff, I think these break down like:

101011000 01 11100100001 0000101 0011010110010101 0001 01011100011 11001110
101011000 01 11100100001 0000101 0101010110010101 0001 01011100110 00110100

Skip the first 9 bits as sync. (Is this right?)
  • DCC: 01 = DCC 1, which matches what is decoded
  • GOTO: 11100100001 = 1825, but this is Type 1 (more following)
  • SITE: 0000101 = Site 5
  • GROUP: 0011010110010101 = 13717; 0101010110010101 = 21909 as talkgroups?
  • TYPE: 0001 = Type 1
  • FREE: 01011100011 = 739; 01011100110 = 742 as free channels
  • CHECK: Last 8 = check digits
SDRTrunk is reporting "UNKNOWN" as the type. But it seems like this is IDLE, ACK, or UNKEY. UNKEY would show a channel of 2047. This does follow a data call, so it could be an ACK, but based on how regularly this is coming across, I think it's an IDLE.

The 2008 (!) RadioReference post by inigo88 suggests that in this case, the GOTO should be either 1792 or 1793. I am wondering if 1825 also has some special meaning.

The 16 bits for group are then interpreted as two hexadecimal groups, representing neighbor and then current system parameters.
In the first message, that's 35 and 95; in the second message, 55 and 95.

The second digit indicates bandplan, so these are bandplan 5. I don't know what "3" for the neighboring modulation means. 9 is apparently 1=FM + 8 = this is the registering channel for the current site that.

Since some channels are spaced 6.25 kHz apart, I took a wild guess that the band plan might be LCN * 0.00625 MHz. If that's true:
739 = 450 + (0.00625) = 454.61875
742 = 450 + (0.00625) = 454.6375

This is coming from the 454.06875 MHz system in Ayer. That suggests it's suggesting either Malden or Marlboro (both have 454.61875) and Gloucester (the only 454.6375) as some of its neighbors, and that those are the registering channels?

But then I'm confused by this one:

20251031 173712,PASSED,DCC:1 ** DATA TG:30101 SITE:005 CHAN:1825/0 FREE:744/0 MSG:10101100001111001000010000101011101011001010110010101110100001110010
Click to expand...
That breaks down to:

101011000 01 11100100001 0000101 0111010110010101 1001 01011101000 01110010

Type 9 suggests this is a data call, as it represents, and that it's going to talkgroup 30101. There's a lot of data calls to that TG.

The GOTO is still 1825. Normally I would take that to mean that the data call is taking place on talkgroup 1825. Based on what you've written, I suspect this has some other meaning. If I try to take it as a frequency anyway, it would be 461.40625 which doesn't map to anything on the wiki. A free channel of 744 mostly makes sense, except that I believe I'm listening to the Ayer site and 744 is 454.65, which is seemingly in Norwell per the wiki.

I also haven't seen any documentation anywhere describing 6.25 kHz spacing, but the LCNs mostly make sense with that.
 
I finished my last post by saying "but the LCNs mostly make sense with that." The emphasis was on "mostly." I think I finally figured it out tonight.

Here's an example from Site 5 (Ayer, MA) on 454.06875 when it appears the CWID was being broadcast at the site:

20251103 131345,PASSED,DCC:1 CALL TG:65535 SITE:005 CHAN:0726/0 FREE:176/0 MSG:10101100001010110101100000101111111111111111100000001011000011011000
20251103 131345,PASSED,DCC:1 CALL TG:65535 SITE:005 CHAN:0726/0 FREE:745/0 MSG:10101100001010110101100000101111111111111111100000101110100110001100
20251103 131345,PASSED,DCC:1 CALL TG:65535 SITE:005 CHAN:0726/0 FREE:176/0 MSG:10101100001010110101100000101111111111111111100000001011000011011000
20251103 131346,PASSED,DCC:1 CALL TG:65535 SITE:005 CHAN:0726/0 FREE:745/0 MSG:10101100001010110101100000101111111111111111100000101110100110001100
Click to expand...

This seems straightforward: it is showing that talkgroup 65535 is active on channel 726, and 176 and 745 are the free repeaters here at site 5.

The wiki lists 452.2, 454.06875, and 454.30625 as the three repeaters at this site. Indeed, I have SDRTrunk listening on those 3 and all are showing Site 5. I had tried using 450 + (0.00625 * channel), and that got me frequencies in the right ballpark, but they didn't really make sense. The frequencies were often either used at another site, or not used anywhere, and they never lined up with something else I was hearing (e.g., a new frequency that's active but not on the wiki).

I think I maybe finally made sense (mostly) of it. Please note that despite the frequency ranges, I am confident I am not sharing any (valid) repeater input frequencies:
  • 450 + (176 x 0.0125) = 452.2; this matches up to the site correctly
  • 450 + (726 x 0.0125) = 459.075; should seemingly be 454.06875
  • 450 + (745 x 0.0125) = 459.3125; should seemingly be 454.30625
So I think what happens is: "normal" frequencies are modeled as-is. Splinter channels are listed as +5.00625 of reality. So, if the result looks like it should be an input, subtract 5 MHz and 6.25 kHz. It would be interesting to monitor one of the sites with 460/470 MHz frequencies and see if they also follow this rule.

I am also curious if this is a unique behavior of the National Grid system (or a newer Passport feature), or if it's actually always been like this but most sites aren't using 6.25 kHz splinter channels so this never came up before? In theory, at least for systems under Part 90, the half of the range of channel numbers that maps to input frequencies are wasted otherwise, because it would never be a legal place for a repeater output.

I still do not understand a lot of the other data being broadcast in normal operations, such as all the sites usually referencing channel 1825. There are also lots of data calls and I don't know if they are actually used for messaging (SCADA sort of stuff?), or if Type=3, Channel=1825 has some special meaning.
 
Hmm, 1825 is 1793 ("normal idle") + 32. It's the difference between
11100000001 = 1793 and
11100100001 = 1825

That has to mean something, right?
 
You must log in or register to reply here.
Back
Top