Re: [gpsd-users] using gpsd on Cubieboard with Garmin 16x-HVS and app written in Qt

[Tom Fanning]

So long as it outputs the NMEA-compliant sentences that gpsd requires, you're good. I suggest you buy one and try it. Return it for a refund if it is not NMEA compliant!

I can't comment on hardware choices, or the urban canyon, since I have no experience.

With regards voltages, please remember that over multi-metre distances, the voltage you measure at the far end of a cable will be lower than the voltage you put in the cable. This is called voltage drop, and is due to the resistance of the wire (it is not a perfect conductor). Physically thicker cables are better, copper cables are better. Don't buy cheap. 

On that basis, gut feeling, I would not expect 3.3V input to a 9m cable to be above 3.0V (the RS-232 limit) when it reaches the far end of the cable. For your application, buy an industrial serial adaptor.

Good luck with the project!

On 21 March 2017 at 16:48, Marek Floriańczyk <address@hidden> wrote:

Dnia wtorek, 21 marca 2017 16:24:16 CET piszesz:
> That Garmin has "true RS-232 output" levels
> ref http://static.garmincdn.com/pumac/GPS16x_TechnicalSpecifications.pdf
> page 5 section 1.5.5
> whatever that means - RS-232 space (0) is +3V to +15V, mark (1) is -3V to
> -15V

Thanks for answer.
There is a new GPS from Garmin:
https://buy.garmin.com/en-US/US/p/100686#specs
http://static.garmin.com/pumac/GPS_19x_HVS_Tech_Specs_EN.pdf

with 9m serial cable and 1.5.2.3 section says it handles CMOS serial output
levels 0 Vdc to 3.3 Vdc - looks like my case.
and also needs separate Input Voltage 8-33 Vdc

So would You say this is a better choice ? I can't see it on GPSD compatible
list
http://www.catb.org/gpsd/hardware.html

Generally I need device to put it on the roof of the bus, so it needs to be
waterproof and temp ranges from -10 to +50 Celsius.
However Garmin needs clear sky, and in urban environment there are tall
buildings.

Could You recommend some device ? ;)

Best Regards
Marek

>
> RS-232 on proper cables is good for 75 feet at 9600 baud, and empirically
> more.
> http://www.marcspages.co.uk/tech/long232.htm
>
> Voltage drop on good quality cables over 5m shouldn't be too bad. Might be
> worth measuring what actual signalling voltages that Garmin actually uses,
> then run the numbers, check the signal is still > 3V and < -3V (i.e. within
> RS-232 spec) by the time it has traversed your cables.
>
> If you need to get commands up to the GPS, you might struggle if your UART
> only signals at 3.3V. You might well drop more than 0.3V over 5m, and the
> Garmin might not tolerate running out of spec.
>
> Good quality cables with twisted pairs and proper shielding should stop any
> RFI problems.
>
> Might just be easier to buy it and try...
>
>
>
> On 21 March 2017 at 12:42, Marek Floriańczyk <address@hidden>
>
> wrote:
> > Hi all,
> >
> > I need some answer/guidance if anyone is familiar with similar setup:
> > Microcomputer Cubieboard2 or Orange Pi plus
> > Garmin 16xHVS or some new model
> > at least 5m long cable between computer and the receiver.
> > Qt app using gpsd as a position source.
> >
> > Will this work reliably if I connect PORT1 of the Garmin dev to 3.3 V UART
> > on
> > microcomputer with cable length at least 5m ?
> > Or should I use usb to rs232 adapter ?
> > Would you recommend some newer/different model of the GPS dev to work
> > with?
> > This setup is for tracking a bus in an urban environment.
> >
> > Best Regards
> > Marek