[time-nuts] Novatel ProPak-V3 GPSDO

Bob Camp kb8tq at n1k.org
Sat Dec 26 14:51:32 EST 2015


Hi

Having spent the last year heading down this road, I may be able to save
you some time. 

To get the full time information out of the NovaTel stuff, you have a “stacked license” situation. 
You get this that is needed for that and then you buy something else. Net cost for the 
licenses is in the $3 to $15K range depending on just how precise you want to be. 

For under $300 with some shopping you can get L1 / L2 receivers that will get you going 
on precise timing. It takes some patience and some time digging. Since the whole gizmo 
fully licensed is cheaper than the L1 / unlicensed NovaTel board, bang for the buck is 
pretty good. 

With L1 only, there are several issues (like ionosphere) that you will have a very hard time
correcting for. Since the solution depends on the exact path to your location, L1/L2 is indeed
the right way to do this. Simply *having* that correction, puts you way ahead of any bells
and whistles there may be in newer software. 

It is interesting to note (as you dig through the various station records) that some of the gear
made back in the late 1990’s still is doing fine in L1/L2 service. The only thing that has made
them switch that gear out has been the mandate to prepare for 2020 and beyond. That 
suggests that the old firmware still does pretty well ….

There is nothing in the post processing “stuff” that is enabled by a fancy L1 receiver that is not present
on some of the cheap modules. I’d start out there ….

Bob

> On Dec 26, 2015, at 6:26 AM, Ole Petter Ronningen <opronningen at gmail.com> wrote:
> 
> Bob,
> 
> Perhaps my subject line should have been "Novatel ProPak-V3 and TAIPPP" - I
> got the unit primarily to play around with PPP and time transfer. I was
> unaware of the GPSDO-functionality when I got it, but I thought it was kind
> of neat and at least worth testing. Will report back if I get any useful
> results on that front.
> 
> But, as you rightly point out, the licenses to get L1/L2/L5 GPS+GLONASS and
> so on are *not* cheap - and I am not at all clear on exactly which (if any)
> firmware options are required to get PPP working properly in a timing
> context.
> 
> I am therefore exploring the possibility to get useful performance on
> L1-only - and there is at least a couple of papers[1] on that subject. It
> would seem that L1 code+carrier tracking, combined with SBAS (or,
> preferably, other precise corrections) in post processing will get the
> PPP-solutions a lot closer to L1/L2 than I would have expected!
> 
> I assume that this accuracy in position will directly transfer over to
> solving for time. If this is indeed the case, anyone with a cesium (or
> perhaps even a rubidium) and one of these receivers could compare their
> frequency-sources to e.g. USNO. It might take several days or weeks to get
> any meaningful precision, and it would be post-processing all the way, but
> still useful if the frequency source is stable enough. (And if not
> particularly useful, then at least interesting!)
> 
> It is quite possible that other (more reasonably priced) receivers will
> also produce useful precision in this setting. The referenced paper makes a
> point that the reported levels of accuracy depend on "high grade equipment"
> - but I am not clear on which properties distinguish "high grade" from "low
> grade" in this context - multipath is mentioned. I have a uBlox that will
> output raw measurements, and I may try to compare the two.
> 
> The receivers used in time transfer needs to be code+carrier tracking. It
> needs to be driven by the frequency source to be calibrated, and according
> to what scant information I have found, also needs a PPS input from the
> local clock. I am unsure about that last requirement, I do not fully
> understand the reasoning - but I *think* it is required to align "time" (as
> opposed to only frequency). It may be that lack of PPS input on the
> receiver can be compensated for with an external TI counter. It may also be
> that it is possible to "calibrate" only frequency, and leave the PPS to its
> own devices.
> 
> But we'll see how it progresses - away from the "lab" at the moment, so
> it's all based on reading papers and guessing. (Which is at least half the
> fun!)
> 
> BR,
> Ole
> 
> 1. Wanninger, L. Hesselbarth, A., "SBAS Based Single and Dual Frequency
> Precise Point Positioning" and "SBAS Orbit and Satellite Clock Corrections
> for Precise Point Positioning"
> 
> 
> On Fri, Dec 25, 2015 at 6:19 PM, Bob Camp <kb8tq at n1k.org> wrote:
> 
>> Hi
>> 
>> I think you will find the Novatel OEM cards inside a number of L1/L2
>> timing receiver designs. The
>> cost for the basic card is not crazy if bought surplus. Doing the *full*
>> set of firmware / software
>> upgrades to make them “work right’ is more expensive than some new
>> automobiles. If you
>> decide to go that way, it’s almost as cheap to buy the brand new V6 board
>> plus the software as
>> to upgrade an old one. The advantage is that the new boards have a bit
>> more processing
>> horsepower on them.
>> 
>> If you run them as “L1 only” devices, the Novatel cards do not have much
>> (if any) advantage
>> over the other boards out there. Since they fetch a premium price surplus,
>> their bang for the
>> buck factor on L1 is not very good. Something like a TBolt is a cheaper /
>> “at least as good”
>> alternative.
>> 
>> I do not know of any of the precision timing people who use the charge
>> pump output of the
>> board for their systems. They pretty much all team the board up with some
>> sort of cpu board
>> (single board computer) and let it do the heavy lifting and data
>> collection. The full implementation
>> then is the Novatel card, the CPU card, and a custom board for the GPSDO /
>> measurement
>> system.
>> 
>> This is by no means a knock on the Novatel cards. They are *very* good
>> devices. The gotcha
>> is the fact that they need a lot of firmware “unlocks” to do all the good
>> stuff they are capable of.
>> With the *full* set of unlocks the newest cards are (when they come out)
>> likely the best thing you
>> can buy for this sort of thing when teamed up with a capable computer.
>> 
>> Bob
>> 
>> 
>>> On Dec 25, 2015, at 6:35 AM, Ole Petter Ronningen <opronningen at gmail.com>
>> wrote:
>>> 
>>> Hello
>>> 
>>> I'm sure many of you have noticed quite a few surplus NOS Novatel OEM-V3
>>> L1-only units appearing on eBay lately. Reading the firmware manual, it
>>> seems the ProPak is capable of steering an external oscillator with the
>>> addition of a filtered chargepump circuit. Has anyone made any
>> experiments
>>> using this functionality?
>>> 
>>> Also, a couple of labs (APL and USNO) use ProPak-V3 receivers to
>>> participate in TAI, by the use of PPP. BIPM has a little information
>> about
>>> the ProPak (calibration procedures), but other than that I have not found
>>> much information about the configuration of the receiver in this
>> particular
>>> setting. Any clues would be appreciated.
>>> 
>>> (In case anyone is interested - the receiver can be software upgraded to
>>> also track L2 - the license is north of USD2K..)
>>> 
>>> Thanks,
>>> Ole
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at febo.com
>>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>>> and follow the instructions there.
>> 
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
>> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.



More information about the time-nuts mailing list