SMALL - espritmodel.com SMALL - Telemetry SMALL - Radio
Reply
Thread Tools
Old Dec 30, 2012, 05:36 AM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Cool
Mod-TRX Project (modern, cheap TRX system)

All the current mainstream TX/RX radio control units available today are overpriced, low-power, analog units transmitting on poor frequency bands.

I'd like to get a project started to create an ultra-low cost, modern, powerful, digital system with telemetry, that operates in more optimal frequency bands (433 and 915 mhz).

This may piss off some people, but IMHO the current systems out there are very poorly designed in many respects.

Current systems are...
-Very low power, operating at 60mW or so.
-Have poor receive sensitivity. (= low link budget)
-Are overpriced.
-Operate in a poor choice of frequency band, 2.4g
-Use analog PPM signals
-Use proprietary protocols, mostly with no telemetry capability
-Are closed source

Design goals for a modern system...
-Medium TX power, 100mW to 1W
-High receive sensitivity (-110 dBm or better)
-Ultra cheap (target < $10 per end)
-Operate in the longer range (and less LOS) 433 or 915 band
-Use an all digital data path (no analog PPM)
-Use open source firmware/software/protocol and hardware
-Capable of medium-high speed telemetry
-Capable of data stream processing and error correction

To produce a system like this we need to start with the best radio chips out there. TI has a three chip solution that meets the design goals, and SiLabs has a comparable one chip solution with the Si100X series.

HopeRF makes a module using the SiLabs chip for around $5-7. 3DR makes a telemetry radio using this module, and there is at least one other company making a 3DR compatible radio with an amplifier to produce a 1W unit.

Although I'm not in love with the HopeRF module, I think it would be a good starting point for a cheap design. We probably can't produce anything ourselves for this price, so we might as well start with what China has to offer.

3DR has a decent open source firmware available. It should be fairly easy to gut all the extra BS from it and use it as a starting point.

When I pitched the idea of using the SiLabs chip for a telemetry radio to 3DR they had one on the market 3 months later. So I don't think it would be too difficult to take this as a starting point and produce a decent TX and RX unit by the spring flying season.

Some people might ask why not just work from the Open LRS system. It uses the same chip radio in the non-MCU version. The reason I see against that is that they use an extra Atmel processor and are still dealing with analog PPM. Using a separate Atmel processor instead of the SoC Si100X (radio + MCU) increases the cost, size, and component count of the board. There's also no ready-made module that can be ordered cheaply and in quantity.

Analog PPM has a fairly large processing overhead and, of course, it has all the disadvantages of an analog signal. By eliminating this silly analog throwback in the data path we should be able to increase the quality and reliability of the control.


I'm throwing all this out there so somebody can shoot down the idea or we can start putting together a project. I'm still working on my CDI project here and need to devote most of my time to that for now, but I like to plan, discuss, and think about new project for awhile to flesh them out before starting.

If anyone's interested in this idea we can start compiling links, data, and design ideas right away!


-Jake
jakestew is online now Find More Posts by jakestew
Reply With Quote
Sign up now
to remove ads between posts
Old Dec 30, 2012, 07:42 AM
Registered User
Romania, Dolj, Craiova
Joined Sep 2007
14,952 Posts
Too much emphasis and hype... about things already done, tried, and abandoned.
OpenLRS is already here and does almost all you want, but ... is useless for me.
There are areas of the world where 433Mhz is so crowded that you can trash your wonderful -110dB receiver sensitivity, that RFM22 chip used by OpenLRS already have.
When you have 60dB noise floor, that cut your range at 2-3km from 100mW, you can only say ...... and look elsewhere.
Conversely, 2.4Ghz is not so crowded as you believe, especially where I fly, my Frsky receiver, featuring -110dB sensitivity and telemetry show me levels of RSSI often exceeding -100dB and ranges up to 5km.

Regarding overpriced... I doubt you can release anything closer to 9X/R radios.
Actually this could be your best approach: Get a 9X, software is open, write your own PPM replacement, but you already have implemented the Frsky PXX, a 16ch HDLC based two way protocol, then get OpenLRS and implement PXX on the module side, and you are done. Such system will comply 99% your desires.
Good luck !
renatoa is online now Find More Posts by renatoa
Reply With Quote
Old Dec 30, 2012, 05:03 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
I've got a 9X system and love it. The new 9XR comes without a TX module and the stock 9X module is pretty basic.

The idea of the project is very similar to the Open LRS, as you mentioned. The main implementation would be as a module for 9X controllers.

2.4g is as crowded as it gets in my area. I live in a fairly small town and every wifi channel is filled in most locations. That's bound to get worse and worse as time goes on.

People used to fight over the lower frequencies, squabbling over even a few mhz. Now everyone thinks more G's = better, so fortunately most new equipment and masses of users have moved to 2g+ frequencies. IMHO that means that the better, lower frequencies should be increasingly open.

If 2.4g is clear now in your area it probably won't be for long. Millions of 2.4g consumer devices are dumped on the public every year.

The $6 HopeRF module is almost ready to go. It's more or less the complete system on a cheap, tiny board. The only difficulty I see is how many PWM channels it can output. It might take a little work to get 8-9 channels out of it. Other than that level shifters and a regulator should be all it needs.

3.3v might be enough for the PWM signals and is probably enough for the serial TTL, so a couple resistors might be all that's needed to make the complete system.
jakestew is online now Find More Posts by jakestew
Reply With Quote
Old Dec 31, 2012, 02:37 PM
RC beginner
New York
Joined Oct 2008
6,054 Posts
Quote:
Originally Posted by jakestew View Post
SiLabs has a comparable one chip solution with the Si100X series.

HopeRF makes a module using the SiLabs chip for around $5-7.
link? or did you mean $5 PLUS $7 (the $12 one with $22 s&h)? if i could locate an si1000 module for $5 or even $7 shipped id order it right away and be forever in your debt.
dave1993 is offline Find More Posts by dave1993
Reply With Quote
Old Jan 02, 2013, 01:57 AM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Last I checked HopeRF didn't have any prices on their site. I emailed them for a quote awhile back.

Looking it up it was actually USD10.00/pc for the HM-TRP on 4/12. That might be up or down since then and I didn't figure out any of the shipping costs. I think the quote was FOB though, so I assume you can use whatever shipping you want.

They didn't give me a minimum order quantity for that price, so it seems they're fine sending them out a few at a time from Shenzhen.

Not as cheap as I remembered, but still not bad. If anyone requests a quote please post it here.

I imagine there will be similar modules from other China manufacturers before long.

I need to look over the couple modules I have, but it should be pretty minimal external components needed.
jakestew is online now Find More Posts by jakestew
Reply With Quote
Old Jan 04, 2013, 07:01 PM
Registered User
Joined Aug 2011
36 Posts
Quote:
Originally Posted by jakestew View Post
All the current mainstream TX/RX radio control units available today are overpriced, low-power, analog units transmitting on poor frequency bands.

I'd like to get a project started to create an ultra-low cost, modern, powerful, digital system with telemetry, that operates in more optimal frequency bands (433 and 915 mhz).

This may piss off some people, but IMHO the current systems out there are very poorly designed in many respects.
Hear hear. I am in agreement and I have actually been thinking about the design requirements for a while.

For what it's worth, here are my considerations regarding a possible modern replacement for existing RC gear, with some additional design parameters imposed by regulations in this part of the world.

Firstly, the radio spectrum is a limited commodity. If we disregard the crowded 2.4 GHz ISM band and above, my choices as per Finnish + EU regulations are, or seem to be as follows:

26,825 - 27,255 MHz at 100 mW ERP max (with internal antennas) or 500 mW ERP max (with external antennas);
35 MHz (legacy analog RC band);
40,660 - 40,790 MHz (100 or 500 mW also, depending on the antenna);
434,040 - 434,790 MHz at 10 mW ERP maximum, no duty cycle limitation;
869,400 - 869,650 MHz at 500 mW ERP with 10% duty cycle OR a transmission protocol compliant with ETSI standard EN 300 220.

Tough, huh? Some more bands exist with even tougher duty cycle or transmit power limits. For any serious application, conformity to the regulations is a must. Any dreams of commercial applications rule out radio amateur bands and power levels, which would otherwise enable more options. One would also be quite mistaken to assume these bands weren't already full of existing users of all kinds, depending on where you operate. If the airwaves are clear, it's probable that 2.4 GHz is also clear, wherever that place is.

Secondly, I would like to have two-way telemetry. Some existing 2.4 GHz RC gear offer some telemetry. Some of the existing LRS gear offer transparent serial links, but see my first point about my limited choice of bands and power levels that tend to rule them out. Which brings me to...

The third point is that I want hardware that is fully compliant, that is, has been CE tested. This means that whatever the system is, it needs to have a spot for dropping in an off-the-shelf, regulation-compliant module of whatever brand proves to be most suitable. Unless, of course, if we somehow ended up going through the testing process ourselves for a design of our own... Suitable RF chips such as those you mentioned exist, and would be the way to start, but only after integrating them into a design (even if it's the reference design) can they be tested to prove that they conform. Right?

Lastly, whatever we come up with needs to meet and exceed current 2.4 GHz hardware in ease of use and operation. Robustness, fast response, secure pairing (at least within reason). Nobody would like to see a UAV getting hijacked with spoofed telemetry guidance commands.

Basically, what I would like to have is this: a dependable radio link with two-way serial communication and a radio controller that has sticks, knobs, buttons and switches that accurately move servos and throttle motors, plus a few binary bits I can set on or off without sacrificing an entire analog channel from the limited amount available just for controlling a simple thing like autopilot mode or navigation lights. This, without bending the rules regulating the use of the radio spectrum.
Helldesk is offline Find More Posts by Helldesk
Last edited by Helldesk; Jan 04, 2013 at 07:10 PM. Reason: copyedit
Reply With Quote
Old Jan 04, 2013, 09:34 PM
Registered User
Mesa Falcon Fld, Arizona, United States
Joined Sep 2004
426 Posts
Jake,

I am not trying to poke holes in your boat, really I am not, but could you please explain your comment about the radios being analog. In my research PPM and PCM are both implementations of a FM FSK method. FSK is by its very nature digital. One contains addressed information, but a FSK signal is either on or off plus the carrier frequency. The filtering that goes on to extract the digital information can be done with discrete components or in software, though most commonly done with hardware. The pulse of the PPM/PCM signal are discritized/digitized. Some of these radios offering 1024 steps of resolution, or 2^10. In other words 10 bit resolution. The receiver also processes these and uses digital timers to translate the data and output it appropriately.

The only thing that I could see being argued about current transmitters being analogue is the fact that they use potentiometers in the sticks and not encoders. For the resolution necessary, you would need ~14 bit encoders to be able to have 10 bits of usable range. This is because the sticks travel about 90 degrees hence you are only using 1/4 of the arc. To be able to maintain 10 bits of resolution over 90 degrees, you would need an encoder that has 4 times the desired resolution.

Please understand, I am really not trying to be a jerk, I just want to be sure that we are using the same terms here and not marketing terms that many vendors tend to use very freely to make their product seem better to a user that does not particularly care how the guts of it works.

Adam
Aeroengineer1 is offline Find More Posts by Aeroengineer1
Reply With Quote
Old Jan 05, 2013, 08:45 AM
Registered User
Joined Aug 2011
36 Posts
Quote:
Originally Posted by Aeroengineer1 View Post
In my research PPM and PCM are both implementations of a FM FSK method.
Pulse position modulation is not related to frequency modulation, hence its own name and its own operating principle. PPM was chosen back in the day because decoding it in the analog receiver hardware of the time didn't require much, possibly just demultiplexing and then sending the individual channels' signals out to the servos. Some RC historian ought to chime in here about how the early servomotors worked. (What came after the ingenious one-button rc controller and the solenoid-and-rubberband-driven rudder actuator?)

Receiving FM transmissions requires a phase locked loop at least, and decoding with fourier analysis was certainly beyond available light-weight hardware in the '60s. PPM receivers on the other hand didn't even need a clock to synchronize, all you needed was counting the time between pulses (differential pulse position modulation).

When affordable digital hardware did appear, it was of course easy to build and market something to encode the signals put out by existing Tx hardware (by using PCM or whatever you chose). Replace the old radio module, read the ppm sum signal coming from the controller, and transmit the data to the receiver with whatever your new encoding scheme was and market it as something-or-other. I say it is time to start on developing something more flexible instead of starting with a piece of the legacy architecture. Gaming joysticks of today don't work by directly shorting pins of the joystick port anymore either, they talk USB or connect via bluetooth. We can fit more controller channels in the airwaves with digital methods, just like TV did.

Please excuse my ranting guys; I don't mean to hijack the thread or your discussion, but I wanted to share this.
Helldesk is offline Find More Posts by Helldesk
Reply With Quote
Old Jan 05, 2013, 09:25 AM
Registered User
Mesa Falcon Fld, Arizona, United States
Joined Sep 2004
426 Posts
Helldesk,


This is not the best reference source, but here is a wiki on PPM. You will notice that it states in the beginning that it is a digital method. Also another on FSK which describes PPM as variant of FSK. From what I read in other links, PPM is just a subset of FSK (Frequency Shift Keying ie Morris code broadcast over a FM carrier frequency).

Adam
Aeroengineer1 is offline Find More Posts by Aeroengineer1
Reply With Quote
Old Jan 05, 2013, 11:03 AM
Registered User
Joined Aug 2011
36 Posts
I don't know what Morris code is, sorry. Anyway, all by itself, using a binary representation of data doesn't grant a system a label of "being digital", as it were. Consider punch cards in the early computer systems, or even their counterparts in any mass storage device of today. Individual bits and bytes are represented as best we can while their representation is deeply rooted in the physical world, neither analog nor digital.

(Until we consider the most basic of properties, the spin of elementary particles and their quantized energy levels which one might think seem digital (discrete, with no continuous or in-between values) in nature... but the crazy quantum physics they obey throw all such ideas out the window).

The position of a stick on a radio controller is a continuously variable value. This was easily represented as a pulse whose position is shifted in proportion to the amount the stick position differs from one of its endpoints. Even though that pulse is a discrete, non-continuous value, I think it doesn't matter to be honest. The first hobby radios did the the job with their analog radio circuits and got the control surfaces turning without a single integrated digital chip in them. As an example, did using Morse code with its basic amplitude-shift keying (on-off keying of a carrier wave) make the telegraph system digital, in the sense that we commonly use the term? Check it out, the ASK is listed next to PPM in that wikipedia article you linked to.

Here, read these sections, maybe they'll do a better job at saying something about what I'm trying to say:
http://en.wikipedia.org/wiki/Modulation#Aim
http://en.wikipedia.org/wiki/Digital...al_information

Digital methods like error correction (or detection, rather), secure pairing and such came along with 2.4 GHz hardware, but all that new good stuff only exists between the radio module and the receiver. The majority of controller internals and operating principles are still firmly in the analog era, as are the components after the Rx. That's how they maintained compatibility back when hardware was way too expensive to replace all at once!
Helldesk is offline Find More Posts by Helldesk
Reply With Quote
Old Jan 05, 2013, 05:19 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Quote:
Originally Posted by Aeroengineer1 View Post
This is not the best reference source, but here is a wiki on PPM. You will notice that it states in the beginning that it is a digital method.
Good question. The digital vs analog thing can get a bit confusing.

PPM uses digital voltage levels (on or off). However the information is NOT encoded in the voltage level, it is encoded in the LENGTH of the pulse.

Since the information is encoded in the time domain, which is analog, the signal is an analog signal. Every time the PPM signal is generated and decoded there is analog-to-digital sampling error present.

If I understand most current systems right the signal path is: Controller sticks -> PPM signal encoded -> TX module -> PPM signal decoded -> digital data RF transmission -> RX unit -> RX encode PWM for servos.

Some systems also use a PPM stage in the RX or use PPM in the RF signal.

In any case, there are errors and lost resolution with using PPM. There's also decoding and encoding overhead on the processor. Writing highly accurate timing sensitive code is much more difficult, while dealing with digital data through hardware peripherals (like I2C, SPI, or UART ports) is very easy.

If we want to use a non-hardware supported signal like PPM the coding becomes more difficult.
jakestew is online now Find More Posts by jakestew
Reply With Quote
Old Jan 05, 2013, 05:44 PM
Registered User
Joined Aug 2011
36 Posts
Quote:
Originally Posted by jakestew View Post
Good question. The digital vs analog thing can get a bit confusing.

PPM uses digital voltage levels (on or off). However the information is NOT encoded in the voltage level, it is encoded in the LENGTH of the pulse.

Since the information is encoded in the time domain, which is analog, the signal is an analog signal. Every time the PPM signal is generated and decoded there is analog-to-digital sampling error present.
PWM encodes it in the length of the pulse, PPM encodes it in the time between rising (or falling) edges of pulses that don't change in length. Other than that, thanks for wording the rest of it better than I could.

I wonder how little the guts of a fully digital radio could cost... sticks connected to digital encoders that don't have pots that wear down, simpler circuits to read and combine it all... something you could affordably mod into a proven casing of an existing controller, which is the hard part anyway (industrial design and ergonomics).
Helldesk is offline Find More Posts by Helldesk
Reply With Quote
Old Jan 05, 2013, 06:12 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Quote:
Originally Posted by Helldesk View Post
Hear hear. I am in agreement and I have actually been thinking about the design requirements for a while.

my choices as per Finnish + EU regulations are, or seem to be as follows:

434,040 - 434,790 MHz at 10 mW ERP maximum, no duty cycle limitation;
869,400 - 869,650 MHz at 500 mW ERP with 10% duty cycle OR a transmission protocol compliant with ETSI standard EN 300 220.

Tough, huh?

The third point is that I want hardware that is fully compliant, that is, has been CE tested. This means that whatever the system is, it needs to have a spot for dropping in an off-the-shelf, regulation-compliant module of whatever brand proves to be most suitable. Unless, of course, if we somehow ended up going through the testing process ourselves for a design of our own... Suitable RF chips such as those you mentioned exist, and would be the way to start, but only after integrating them into a design (even if it's the reference design) can they be tested to prove that they conform. Right?
Here is a link to the HopeRF HM-TRP module page...
http://www.hoperf.com/rf/fsk/HM-TRP.htm

Here is the HM-TRP datasheet...
http://www.hoperf.com/upload/rf_app/hm-trp.pdf

HopeRF makes 433/470/868/915MHz versions of the module. The difference is in the PA matching network, aka the low-pass RF output filter.

Any of these frequency ranges is software selectable, but if your low pass filter is too low you won't get any output. I think the 915 should be able to transmit on 868 fine, but otherwise you'd have to get the right band module.

The TX power is also software selectable, so you should have no problems meeting your requirements. Whatever TX protocol requirements there are should also not be difficult to meet, so that you can use the higher power levels in those countries.

One thing I'd like to do is use the RSSI to constantly adjust power levels. There's no point in wasting power when you're at close range.

Another feature I'd like to implement is a "power boost" button. This would allow you to operate at low, legal power levels, but if you lost signal you could briefly break the regulations just long enough to regain control. That should allow you to not have to worry about range checks either since you'll always have a bit of extra reserve power even after you find your range limit.


As far as the certifications... you might have to relax your requirements a bit there. IIRC the HopeRF modules are have some sort of CE marking, but that doesn't really mean much.

These designs will be homemade experimental designs, so there's not much hope of certification. You can probably figure out your laws to where you are compliant with the restrictions on homemade-type electronic equipment.

You should also feel better knowing that these modules are ARE used in some FCC certified designs. With a little pressure on HopeRF they might get the design certified.

3DR slaps these modules on a carrier PCB and sells them with no certifications. I guess they're assuming that the FCC will consider it only a component of a larger system and not "end-user equipment".

As you will be the one building the equipment you'll be considered the manufacturer and the certifications will be your responsibility. You can self-certify the device and put the CE mark on it yourself if you want.

For FCC or other certifications the easiest and cheapest way would be to find other certified devices that use the module and contact them. If you can get them to issue you a LOA (letter of authority) to use their FCC number, then you can put it on your devices. If your device is similar to their's and doesn't compete in the same market that's a possibility worth exploring.

Alternately, you could contact the company and HopeRF. They might be willing to issue HopeRF a LOA, which means HopeRF can put the FCC number on the module or packaging, which you would then also put on the case of your device.

In any case, I think there are lots of devices that will be using this module. 3DR will also probably be interested in getting their's certified. At some point there should be enough people interested in getting the module certified for us to get together on the costs.

I'll have to check if there's any FCC numbers for the actual SiLabs chip and see how appropriate it is to put that on the device case.

Another consideration is that it would be easy to install this inside the 9X case as an internal mod. People seem to be doing this with other TX modules.

While this might seem a little deceptive, any authority would simply look at the back of the 9X case and see whatever certifications and numbers are listed there.

I've never come across a cop with any interest in poking around someone's radio gear, and I've certainly never heard of one that would know enough about RF devices and FCC rules to make the judgement that you're breaking the law.

The way I look at it, if the firmware controls the TX power to the minimum you need for the desired RSSI/signal strength, then 95% of the time you'll be putting out much LESS power than a standard TX module. You'll only be using higher power when you need it, and when you need the power you need it for safe operation. You can't justify risking the safety of anyone because of some regulations, so if you accidentally go out of range the only moral and legal thing to do is increase your power to regain control.

In practice, the TX power regulations may have to be violated in some circumstances to avoid breaking much more serious laws. I'll take a FCC warning and/or fine over "reckless endangerment / negligence" any day of the week.
jakestew is online now Find More Posts by jakestew
Reply With Quote
Old Jan 05, 2013, 06:33 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Quote:
Originally Posted by Helldesk View Post
PWM encodes it in the length of the pulse, PPM encodes it in the time between rising (or falling) edges of pulses that don't change in length. Other than that, thanks for wording the rest of it better than I could.
Good catch, that's what I meant to say. Measuring the length between pulses is essentially the same thing as what I said though since you can also interpret/implement the decoding by measuring the length of the low going pulse. (Same as considering the inverted signal)

In any case, PPM works by measuring signals that encode data in the time domain. It's an analog signal since you have to repeatedly sample the voltage on a line and convert the time between pules into a digital value.

Quote:
Originally Posted by Helldesk View Post
I wonder how little the guts of a fully digital radio could cost... sticks connected to digital encoders that don't have pots that wear down, simpler circuits to read and combine it all... something you could affordably mod into a proven casing of an existing controller, which is the hard part anyway (industrial design and ergonomics).
Mice are pretty cheap. They use optical encoders to measure the turning action of the rollers that contact the mouse ball.

You could probably pretty easily take an old mouse, remove the ball and case, then attach sticks to the rollers. Or you could just flip it over and screw a RC stick into the ball. Those components must also be available somewhere. I'd think it would be pretty easy to simply attach the optical encoder to the sticks of a 9X in place of the pots.

You'd end up with a fully digital, highly accurate stick position measurement. If we can figure out the parts I bet quite a few people would be interested. It would be a great project for everyone with work out 9X units out there.


BTW, here is the link to the 3DR telemetry radios...
https://store.diydrones.com/3DR_Radi...try-3dr915.htm

That link contains Eagle schematics and other good info. You can also find the links to the open source SiK firmware they're using. The 3DR radio is almost exactly what we're trying to do.

The main issues with using the 3DR radios for RC control is that most of the Si1000 pins are not exposed. The firmware also needs some work as it is concerned with dealing with the MAVLINK telemetry protocol (useless overhead for us), and the binding method could probably use some work to make it faster and more reliable (reconnections after signal loss could be an issue). We'd also have to add the PWM output code.

IIRC the Si1000 has hardware support for 6 PWM channels. So the other two we'd have to implement in software, or use those channels for other sorts of output like digital signals, relays, etc.. A nice feature of the Si1000 is that it has a "digital crossbar" which lets you map almost any function to any pin.
jakestew is online now Find More Posts by jakestew
Reply With Quote
Old Jan 05, 2013, 08:23 PM
Registered User
Joined Aug 2011
36 Posts
Regarding certifications: good points there, and I'd have to make some personal inquiries to the regulators anyway at some stage if I ended up self-building radio gear. About authorities checking up on things, well, I don't know how common it is but just recently I noticed a bulletin about our FCC equivalent possibly doing actual inspections or observation regarding some permit-requiring radio gear used in the hunting season. Of course those things bring in money as you have to licence them so it's a different kettle of fish...

A practical factor and a possible good reason for me to be conservative about these things is that if this radio gear ended up being useful enough in a flying machine usable for SAR in cooperation with authorities, or some such semi-official thing, well, it could get rather complicated if I couldn't answer truthfully about the nature of the comm link, if I was asked about it.

Morally there's no problem if a hobbyist knows what they are doing, and there are no legal problems either if the kit is built to spec. Nobody probably cares about a lone, intermittent, relatively low-power transmitter in the wild; there are bigger fish to fry to boot.

Checking RSSI levels for dynamic adjustments should be a good trick; you would make sure the received signal is decoded a-ok too, right? A high RSSI level alone might as well indicate a high noise floor or another transmitter on the same band, which is bound to happen at some point.

I am familiar with the 3DR radios, I was seriously considering them when they came out. I ended up waiting for a while when we were figuring out the compliance questions back then, and then also for the usb noise issue on the 433 MHz ground units (worked around by using two air modules, one on the ground with an ftdi cable). I've actually considered purchasing them and still am, for testing purposes at least.

From what I gathered from elsewhere, I got the impression that the 3DR implementation is a bit overengineered with a whole extra chip on them, when the Si1000 could possibly take care of everything alone. I guess it's the issue of non-exposed pins then that dictated that design choice?

On the self-built controller front... I found this: http://www.ersky9x.net/ - basically a 9X board replacement with extra input and output possibilities. Could make for a nice start for a fully digital controller.

Oh, I meant to ask you, Jake... do you think a 10% duty cycle I listed earlier could possibly work for controlling a flying machine? You could transmit for 100 ms total per each second. If you wanted 50 updates of control surface positions each second, you would have to get a full frame sent in just 2 ms... 10 Hz updates would let you get it done in 10 ms. Not much to work with, eh? Then again, a 2 ms ping over a (fiber) network hop or a few is perfectly doable today, even with consumer hardware.
Helldesk is offline Find More Posts by Helldesk
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Mod my 7CAP or buy a new cheap radio to use FRSKY system? OrlandoP Radios 2 Jul 18, 2011 05:12 PM
For Sale Werks Racing .28 Ceo Mods/Trx 3.3 Engines FT/FS Jamminracer18 Cars - Cars and Parts (FS/W) 1 Apr 18, 2009 11:17 AM
Build Log Project Gatika, Modern Voith Tug yacht boy Scale Boats 198 Feb 19, 2009 04:05 PM
Sold modded mini t arr for sale "cheap" or trade for mamba system jessnlaurie Cars - Cars and Parts (FS/W) 2 Oct 28, 2005 10:03 PM
FORMOSA project! Cheap, over 1:1 power system PunkerTFC Foamies (Kits) 12 Jun 16, 2005 08:45 AM