NTRIP and Mobile Internet

[steveR]

I have been looking again at some of the less mainstream offerings such as Fieldbee and I am wondering how close the RTK connection broadcast needs to be based, to allow accurate location adjustments, when received over the Internet?

10km, 100km?? I assume you need the same satellites need to be connected to the GPS Receiver AND the RTK adjustment receiver base?

 

[Pheasant Surprise]

Distance between base and “rover” will degrade the positional accuracy. Typically 40 to 50 km is your practical limit to maintain a reasonable “inch” level of repeatable accuracy, although so called “long baselines” are possible +100km (or more) if you’re prepared to acknowledge that the accuracy will be worse.

Most mainstream receiver manufacturers will report their basic level of RTK (horizontal) positional accuracy somewhere around 8 to 11mm + 1ppm - that’s parts per million. So with some quick maths you can work out how much the accuracy degrades as you move away from the base.

 

[steveR]

Thank you, that is hugely useful!

I have been trawling around for a while trying to get a handle on the effects of distance between Base and rover (I like it!) as we have a free access base unit about 30 miles away as a crow fly's...

For "Drive by Human" 15-20cm accuracy is more than accurate enough!! Better than the Garmin Glo accuracy of about 1.5m, and waaay better than a Phone or Tablet GPS!!

 

[Pheasant Surprise]

Yeh 30 miles will be absolutely fine.

 

[Northern farmer]

Is this FieldBee a dual frequency device? 30 miles would probably be too much for a single frequency device.

 

[steveR]

Is this FieldBee a dual frequency device? 30 miles would probably be too much for a single frequency device.

Cannot do better than point you here...

https://fieldbee.com/wp-content/uploads/2019/02/FieldBee-Datasheet_en.v-1.6.pdf

 

[Briantee]

Residual errors grow as the baseline length increases, which limits practical single-frequency RTK to the order of a kilometer or two. A second or third frequency allows for the formation of linear combinations of phase measurements with longer effective wavelengths, and this enables ambiguity resolution with shorter convergence times and over longer baselines. Almost all commercial RTK systems are dual-frequency for this reason and allow for single baselines solutions on the order of 20 kilometers that can be resolved in minutes.

 

[Northern farmer]

Cannot do better than point you here...

https://fieldbee.com/wp-content/uploads/2019/02/FieldBee-Datasheet_en.v-1.6.pdf

I did google something similar but was not sure if I got the correct product variant. This one is L1 only, not good for long range.

 

[Pheasant Surprise]

For very similar-ish money to the FieldBee thing now you can get decent L1/L2 dual (even triple) frequency (and 4 global constellation) receivers.

I would strongly advise the OP to use a more or less industry standard dual-frequency setup. Something I missed in his OP when I said that 30 miles would be fine.

Having used an L1 only setup vs a dual-frequency, it’s dual frequency every time.

 

[steveR]

I did google something similar but was not sure if I got the correct product variant. This one is L1 only, not good for long range.

My query was more based on the use of "base" units that "broadcast" the data over the Internet and act as the RTK base but at effectively zero cost to the end user of the data. A mobile phone receives the data and is then acting as a mobile hotspot for the "roaming" unit to access and utilise.

The issue of the range is not down to a radio transmitter then, rather if the "base" unit is close enough to the mobile for a working system. ie. The "baseline length" ?

As a complete numpty/novice, it seemed to me that this would allow a low cost approach off just a single L1 unit. Could an L2 unit access 2 internet data feeds from differing sources? Presumably giving in effect, a triangulation?

The aim being to keep cost down by not requiring the purchase of a base unit... as per teh Fieldbee!

 

[Pheasant Surprise]

My query was more based on the use of "base" units that "broadcast" the data over the Internet and act as the RTK base but at effectively zero cost to the end user of the data. A mobile phone receives the data and is then acting as a mobile hotspot for the "roaming" unit to access and utilise.

The issue of the range is not down to a radio transmitter then, rather if the "base" unit is close enough to the mobile for a working system. ie. The "baseline length" ?

As a complete numpty/novice, it seemed to me that this would allow a low cost approach off just a single L1 unit. Could an L2 unit access 2 internet data feeds from differing sources? Presumably giving in effect, a triangulation?

The aim being to keep cost down by not requiring the purchase of a base unit... as per teh Fieldbee!

Pretty much every commercial (or paid for Ag or survey spec) base will be dual frequency GPS/GLONASS . In very simple terms a single frequency base will be play / hobbyist stuff from the last 2-3 years or something extremely ancient. So ancient I’m talking last millennium.

An L1 rover will happily get corrections from an L1/L2 base, but unfortunately (as for those of us that have tried and know) not the other way around. An L1 only “hobbyist” base is basically useless with a commercial spec L1/L2 rover.

 

[Briantee]

It's shocking what you can get now. 2 of these with this kind of accuracy were 40k not long ago L1/L2/L5. Now $3700 for complete set - even has built in sims for NTRIP. Incredible for a waterproof ready to run system.

https://emlid.com/multi-band-reach-rs2-is-here/

 

[Pheasant Surprise]

It's shocking what you can get now. 2 of these with this kind of accuracy were 40k not long ago L1/L2/L5. Now $3700 for complete set - even has built in sims for NTRIP. Incredible for a waterproof ready to run system.

https://emlid.com/multi-band-reach-rs2-is-here/

Early tests by some users around the globe are indicating excellent range from the built in LoRa radios, which have really minuscule power output compared to traditional UHF digital radios we’re used to....

https://twitter.com/emlid/status/1147169183370829826?s=21

As Emlid use a proprietary/closed protocol over LoRa, I’ve suggested to Emlid that they make a compatible inexpensive standalone LoRa radio with serial output that could be connected to a third party receiver to receive corrections direct from an RS2 base.

Otherwise as these have built-in cellular radios/SIM card, I suppose one could just use standard NTRIP and hook the base up to something free like SNIP as a caster.

 

[steveR]

Pretty much every commercial (or paid for Ag or survey spec) base will be dual frequency GPS/GLONASS . In very simple terms a single frequency base will be play / hobbyist stuff from the last 2-3 years or something extremely ancient. So ancient I’m talking last millennium.

An L1 rover will happily get corrections from an L1/L2 base, but unfortunately (as for those of us that have tried and know) not the other way around. An L1 only “hobbyist” base is basically useless with a commercial spec L1/L2 rover.

Thanks you for that, I am improving my knowledge by the day on this topic!

I will go back... and look again at any possible purchases more closely, and with a MUCH more informed viewpoint.

 

[steveR]

ndicating excellent range from the built in LoRa radios, which have really minuscule power output compared to traditional UHF digital radios we’re used to....


Otherwise as these have built-in cellular radios/SIM card, I suppose one could just use standard NTRIP and hook the base up to something free like SNIP as a caster.

This is what I was wondering in my OP, but using other's caster Base data. Hence the question on how far (what I now know is called the Baseline!) can be considered effective when used with a basic unit such as a Fieldbee or Reach RS. I'e read the info on the Reach2 and it looks great kit...But out of my league pricewise!

http://www.rtk2go.com/

 

[Briantee]

I think its important to know how accurate you want to be - both repeatability vs resolution (year to year vs track to track)

 

[steveR]

I think its important to know how accurate you want to be - both repeatability vs resolution (year to year vs track to track)

As I said earlier in the thread:

"For "Drive by Human" 15-20cm accuracy is more than accurate enough!! Better than the Garmin Glo accuracy of about 1.5m, and waaay better than a Phone or Tablet GPS!! "

With no realistic need for autosteer just yet, 10-20cm would be amazingly good! Year to Year is not required. It is more for accurate spraying and fert applications on grassland.

My foray into GPS guided farming started 14-15 years ago with the FarmerGPS software running on a Toughbook in W95 (I guess) I even have the interface hardware still. However, the accuracy was very dubious at the time for even something as non fussy as slug pelleting which is what I was experimenting with it for.

 

[Pheasant Surprise]

£1,400 or so for one of these....

.....or somewhere between £6K to £7K for one of these....hmmmmm

 

[steveR]

Nice kit!!

I have kept looking at the Emlid Reach site since I saw the post/link you put up a while back...

That was the RS+ which has dropped enough in price to buy the base and rover unit for 1500 quid, although no one seems to have them in UK stock at the moment...

 

[Northern farmer]

With no realistic need for autosteer just yet, 10-20cm would be amazingly good! Year to Year is not required. It is more for accurate spraying and fert applications on grassland.

EGNOS should give 20 to 30 cm pass-to-pass accuracy and is usually considered accurate enough for spraying and fertiliser spreading. No need for any base and the correction signal is free.

Considering Reach RS2, I wonder if their deliveries are in schedule. Have not seen owner reports at Ag forums and the Reach website still promotes pre-orders for delivery in mid June.