Transmit power and range

[stephenmelody]

All I did was unplug it and plug it in again in another room... I'll have to check but the will to live walked out a few minutes ago and I went to the fridge to get a beer...!!

[Admin]

I see where youre coming from 

[stephenmelody]

I'm away from Friday, I may have some time tomorrow to test further, but I'm not sure as I need to pack and depart at 4am... Otherwise I'll test again when I'm at home.

[rodent0_2]

A thought, can the ARF board be remote mounted with the antenna and the control cable run back to the Pi, no coax needed, we could also do with decent antennas especially with the flex wigs as we don't won't stray bits n pieces going through the prop.

Nick

[Moffrestorer]

Yes, the ARF in PAW is using the "high power" (sic) bit of the band. What we need and what we can get are unfortunately likely to be poles apart once metalwork gets in the way. The problem with bolting on amps (transmitter or receiver) is the need for T/R switching so that both can use the single antenna. All is possible, of course but the technology isn't commonplace (i.e. cheap). The ARF already has a PA-cum-RF amplifier that includes the T/R switching fabric. (Edit - I see that Jerry has covered some of these points whilst I was doing my one fingered typist trick.)

I'm working up to an experiment. I shall place PAW #1 in the visual control room at my local airfield, then trundle off with PAW #2. In the aircraft I shall have a) the standard whip antenna lurking on the coaming, b) a dipole taped to the canopy, and c) a quarter wave ground plane antenna protruding through a convenient hole in the underside of my aircraft's all metal fuselage (don't ask...!). I'll then do some tests, flying from and to the airfield with the different antenna configurations. All I need is for it to stop raining... should be an interesting test.

I'm not sure what alternatives to the ARF Lee may be considering. If these do not achieve any improvement in transmit power, I wonder if a narrow band ( or even broad band) power amplifier connected to the ARF TX output, might be the way to go (dependent also on TWF's experiment described above).

My thoughts are;
1) Existing ARF antenna retained and used for ARF receive (and low power TX) as at present, via the end feed SMA connector.

2) ARF pig-tail coax connected to ARF output U/FL connector (effectively "T" ee'd off (1) above), and connecting to power amplifier input, via cable's SMA female end.

3) Amplifier output connected to additional antenna (for high power transmit).

4) Separate (high power i.e. 450 mw) antenna and existing ARF antenna gets around Tx/Rx switching referred to by TWF.

Is this too much of a bodge, could it work? Would the "T"  reduce the ARF receive sensitivity? I consider the two TX antennas could assist "position diversity" as called for, if connected to PAW using coax cables.

Regards,

Chris

[The Westmorland Flyer]

A thought, can the ARF board be remote mounted with the antenna and the control cable run back to the Pi, no coax needed, we could also do with decent antennas especially with the flex wigs as we don't won't stray bits n pieces going through the prop.

Probably, yes. The ARF only needs a few connections and none of them are carrying seriously fast data, although noise immunity might be an issue with only 3.3V signalling levels. You would probably also remote the 3.3V regulator to reduce the issue of voltage drop. Decent quality coaxial cable and short runs would be a simpler solution.

I'm not sure what alternatives to the ARF Lee may be considering. If these do not achieve any improvement in transmit power, I wonder if a narrow band ( or even broad band) power amplifier connected to the ARF TX output, might be the way to go... <snip>

Perhaps although it seems like a complex solution that is only fixing one half of the problem. Really we need the Tx AND the Rx to be able to work better and there is only one common component to that problem: the antenna. That said Lee's work on a replacement for the ARF is important because the 500mW would be handy to have and because of the apparent variability in Tx (and possibly Rx) performance between units.

[JCurtis]

Yes, the ARF in PAW is using the "high power" (sic) bit of the band. What we need and what we can get are unfortunately likely to be poles apart once metalwork gets in the way. The problem with bolting on amps (transmitter or receiver) is the need for T/R switching so that both can use the single antenna. All is possible, of course but the technology isn't commonplace (i.e. cheap). The ARF already has a PA-cum-RF amplifier that includes the T/R switching fabric. (Edit - I see that Jerry has covered some of these points whilst I was doing my one fingered typist trick.)

I'm working up to an experiment. I shall place PAW #1 in the visual control room at my local airfield, then trundle off with PAW #2. In the aircraft I shall have a) the standard whip antenna lurking on the coaming, b) a dipole taped to the canopy, and c) a quarter wave ground plane antenna protruding through a convenient hole in the underside of my aircraft's all metal fuselage (don't ask...!). I'll then do some tests, flying from and to the airfield with the different antenna configurations. All I need is for it to stop raining... should be an interesting test.

I'm not sure what alternatives to the ARF Lee may be considering. If these do not achieve any improvement in transmit power, I wonder if a narrow band ( or even broad band) power amplifier connected to the ARF TX output, might be the way to go (dependent also on TWF's experiment described above).

My thoughts are;
1) Existing ARF antenna retained and used for ARF receive (and low power TX) as at present, via the end feed SMA connector.

2) ARF pig-tail coax connected to ARF output U/FL connector (effectively "T" ee'd off (1) above), and connecting to power amplifier input, via cable's SMA female end.

3) Amplifier output connected to additional antenna (for high power transmit).

4) Separate (high power i.e. 450 mw) antenna and existing ARF antenna gets around Tx/Rx switching referred to by TWF.

Is this too much of a bodge, could it work? Would the "T"  reduce the ARF receive sensitivity? I consider the two TX antennas could assist "position diversity" as called for, if connected to PAW using coax cables.

Regards,

Chris

Not sure this would work, the additional amplification would probably slightly alter the phase of the signal, so you could make things worse for receivers at shorter range as the two signals contend.  Any additional amp also needs to know when to turn on & off, that signal is available on the ARF board to tell the CC1190 what mode it's in but not easily accessible.  An Amp will just amplify until you tell it otherwise, the carrier will still transmit.

Other modules exist, Lee and his team are looking at options, for me any replacement should be certified for a number of reasons including repeatable quality.  Certified doesn't mean expensive, they can be had for within a few £'s of the ARF.  The big players in this area make them by the truck load for use in all sorts of things, saw a few at an electronics show last week and that was on one stand.  Whats also handy is the offer of assistance with antenna selection and/or design to get a good match where needed.

[Moffrestorer]

Hi TWF and Jeremy,

I agree regarding antenna efficiency and position, however I think a big part of Lee's original concept for PAW was one of portability, particularly for renters or users of certified aircraft where mod approvals for installing antennae is for PAW difficult or undesirable.

I'd assumed the TX carrier is switched off whilst the ARF is in Rx mode due to Tx/Rx switching in the ARF. Accept phase differences between TX signals potentially a problem.

Regarding replacement for ARF, just a little bit concerned over ability to talk to Pi and amount of reprogramming this might mean for Lee. Accept your comments over certified performance. If a suitable device with appropriate output power and input sensitivity can be found this is likely to be a big step forward for PAW.

(Reading open micros.org description of ARF/SRF including TI admission about unpredictable behaviour, had me slightly concerned and made me wonder if some of the variability the testers are seeing might be caused by such factors as well as RF propagation issues)

[JCurtis]

I'd assumed the TX carrier is switched off whilst the ARF is in Rx mode due to Tx/Rx switching in the ARF. Accept phase differences between TX signals potentially a problem.

Yes, it is, the CC1190 on the ARF flips between Tx and Rx based on a signal received from the small SRF board on the ARF module.  When in Rx is operates as a high gain receiver, when in Tx a power amp.  If you add anything external to the ARF to try and boost things etc. it would also need to 'see' these control signals to be able to switch the amp on and off.

(Reading open micros.org description of ARF/SRF including TI admission about unpredictable behaviour, had me slightly concerned and made me wonder if some of the variability the testers are seeing might be caused by such factors as well as RF propagation issues)

I've not seen that admission, but would not be surprised if production variances caused enough of a problem to explain some of the oddities.

[SteveN]

Have people run a scanner where they are testing to check what else is transmitting and at what duty cycle ?

Trouble with 869.4Mhz is 500mw could end up the base level.

example:

http://xlsystems.co.uk/html/869mhz.html

If a few of those were within 10 miles or so....

just a thought.

[The Westmorland Flyer]

I too have only heard my PAW(s) but I live in a very rural area, my nearest neighbour 400m away, so I wouldn't expect to hear anything that I hadn't generated myself.

The 868MHz ISM band is, however, free for all to use and it's inevitable that as time goes on this relatively new allocation will become busier. At 2000ft we'll be able to pick up ground-based systems over a wide range. The main thing that will help to reduce the problem is the very short duty cycle that most, if not all these devices will be using.

It would be instructive to take the scanner flying some day!

[JCurtis]

Have people run a scanner where they are testing to check what else is transmitting and at what duty cycle ?

Trouble with 869.4Mhz is 500mw could end up the base level.

example:

http://xlsystems.co.uk/html/869mhz.html

If a few of those were within 10 miles or so....

just a thought.

Yes, and around Cambridge there is something else either on 869.4 or very close by.  From memory it transmitted briefly every couple of minutes.
User the ISM band the duty cycle can be no more than 10% (per hour) so that shouldn't be an issue.  A device can transmit for a total of no more than 36 seconds in any given hour.

[Moffrestorer]

I was wondering about "interference " from other ISM sources. Am I being overly simplistic in my thoughts that when PAW receives transmissions from non-PAW sources, data will be unrecognisable as not conforming to PAW protocols, therefore nothing is output to display or otherwise?

[The Westmorland Flyer]

...the ISM band the duty cycle can be no more than 10% (per hour) so that shouldn't be an issue.  A device can transmit for a total of no more than 36 seconds in any given hour.

10% would be 360 seconds but perhaps there is an overarching limit of 36s/hour as well, effectively making the long term maximum duty cycle 1%?

Edit: I see on further investigation that the P3i protocol has a 2.5% duty cycle, meaning it is transmitting for a total of 90 seconds/hour. From what I can tell there are two parameters: an overall duty cycle of <10% and, separately, a maximum single transmission time of 36s.

[The Westmorland Flyer]

I was wondering about "interference " from other ISM sources. Am I being overly simplistic in my thoughts that when PAW receives transmissions from non-PAW sources, data will be unrecognisable as not conforming to PAW protocols, therefore nothing is output to display or otherwise?

Yes, the packet protocol (packet length, CRC) should ensure that non P3i packets are discarded although I am not clear how reliable that would be when there is lots of different traffic. Hopefully there is also lots of data field integrity checking as well, so any non-P3i packets that do manage to pass length/CRC muster are ignored.