Thread Tools
Mar 28, 2008, 03:37 AM
Registered User
my current programs are straight 802.15.4 apps, I have a feeling that if you create zigbee apps they use more ram.
Sign up now
to remove ads between posts
Mar 28, 2008, 04:09 AM
Registered User
you got me thinking so I looked this stuff up properly. The copy to ram takes 0.8ms/Kbyte so less than 10ms for my program.
Mar 28, 2008, 09:44 AM
That tree again!!!!
thanhTran's Avatar
Hi Alan

Thank you very much for your reply. Looks like we have plenty of space left for our stuff. 32KB left of flash is pretty good, I think .

How do you like the range of the on board chip antenna? You've gotten the range around 200m with that, correct?

Thanks again


ps. I was wondering why you had to do frequency hopping while the underneath 802.15.4 already uses DSSS (Direct Sequence Spread Spectrum) until I re-checked my understanding of what DSSS is
Mar 31, 2008, 02:26 AM
Have Fun !

DIY Low cost 2.4Ghz module and receiver

Hi !

A DIY low cost 2.4Ghz R/C project can be find here

Only on French for the moment, i have not enough funtime to translate in english all the website, but all you need to buid your own transmitter module and receiver are available in the "download" section of the web site -> PCB, Hex file and many more...

We start to build a downlink daughterboard now...

Mar 31, 2008, 01:10 PM
Registered User
yep, I get about 200m on the ground with the chip antenna, I am about to buy a high power module with an external antenna for my proper transmitter.

I started hopping as a simple way to allow for the chosen channel becoming bad. If you don't hop all the time, deciding when to change channel is quite tricky and there is a risk you can further delay communication by changing channels. It is also quite hard to sychronise the channel change without further delay when comms are already bad.

Hopping all the time also has its problems, if only a small number of channels are good a lot of time will be spent sending on bad channels.

Whatever you do you can always thing of a situation were another system would be better!

My current experiment involves a mix of both systems, for the first bit of each frame I use a channel that only changes if it fails after a number of frames, if this fails I use the rest of the frame time in a continuously hopping mode. This hopefully will combine the advantages of both systems, using one good channel most of the time but with fast response to change.

I have been wondering if our systems could be made to interoperate as the underlying radio spec is meant to be the same.

Mar 31, 2008, 03:47 PM
Registered User
Hi Alan, Interesting project. I've not read enough of the datasheets to find the answer to this question: Does the module transmit on the chosen channel when instructed by you or does it check for activity and only transmit if not in use?
Thanks, David.
Mar 31, 2008, 04:12 PM
Registered User
David, by default you ask it nicely to transmit, it waits a random length of time, checks the channels clear and only transmits if it is. If the channel is not clear it waits a for a possibly longer random period and tries again, it repeats this for a configurable number of times. You can adjust the delay period down to zero. I'm not sure if you can force it to transmit if the channel is not clear ( I don't want to, so haven't tried). The api does all this for you, however in my current solution I do all the delay and retry stuff in my software so I can timestamp each retry with something close to the actual send time, this lets me synchronise clocks on the rx and tx. This all makes the hopping stuff simpler.

What do people know about antenna diversity? Is there any merit in having two transmitting antenna rather than doubling up on the receiver?

Mar 31, 2008, 04:30 PM
Registered User
WIFI antennae are directional so presumably these are as well and 2 should give better coverage. I believe Spektrum uses two receivers to reserve 2 channels but also uses it to cover directional issues.

Thanks for the transmission clarification. I'd love to use these devices for a webcam project at our flying field but am trying to assess the implications at an active RC field. My understanding is that Zigbee uses 16 channels with 5mHz spacing and that RC radios use 1 or 2 from 80 channels with 1Mhz spacing. Both cover the same overall bandwidth spectrum. So it will only transmit if the chosen channel is free but when it does it will occupy the equivalent of 5 Futaba/Spektrum 1Mhz channels. Does this sound correct?
Apr 02, 2008, 01:22 AM
That tree again!!!!
thanhTran's Avatar
thanks for the info Alan
I think diversity on the receiver makes more sense since the multi-path issue (signal bouncing of different path could be out of phase and could eliminate each other) impacts it directly. In your case, the transmitter is also a receiver, so it should be diversified both ways, I guess.

David, I think your understanding about the frequency cover is correct. The 5Mhz bandwidth signal will overlap the 1Mhz bandwidth channels.
Apr 02, 2008, 03:11 AM
Registered User
David, whilst the channels overlap I'm not sure that it automatically means that a transmission will block 5 of these other channels. The best way to make it friendly to other systems is to restrict the amount of data you send. If you only use say 1/10 of the 250kps available you would have (I guess ) roughly the same impact as a single xps system.

Apr 02, 2008, 12:31 PM
Registered User
Further research reveals that the zigbee transmission deliberately has a fairly broad distribution peaking on its target/center frequency for the chosen channel. Here's an example: The 5MHz spacing seems to cover only the strong part of the signal with further (weaker) spillover into adjacent channels. So I believe it will dominate 5Mhz of bandwidth while transmitting and depending on it's power output potentially a bit more.

The '4km' Jennic which I'd be interested in is rated at 19dBm which seems to equate to 79mW. RC transmitters are 100mW so the more useful module is going to compete with comparable power.

Alan, I could be wrong but I now believe the amount of data will determine transmit durations, not the shape of the frequency distribution but I agree the result is similar. So I'd expect data transmissions for remote control to be brief but the webcam type application I thought this might be useful for would probably take longer if I included image type data.

Judging by the small size of the RC receiver aerials, I assume they are only receiving and not transmitting any acknowledgements back. I believe Futaba keeps hopping to spread the risk of interference and Spektrum have two channels and presumably the Tx checks occasionally and changes them if they become busy. Either way my instinct is to avoid having any other 2.4GHz based transmissions at the flying field.
Apr 15, 2008, 11:06 AM
Registered User
Alan, any update on your project?

I have been following Powerbook's site for a while, and have been contemplating taking the plunge and trying to get it to work. A couple of differences, though, the obvious being that it uses the Xbee transceiver. It seems that the the jennic modules you have been working on are more flexible than the Xbee. They are also smaller and have some different antenna connection options.

As for diversity, obviously the route that others have been taking are with multiple receiver antennas (or receivers). I've tried to find off the shelf antennas that would incorporate diversity, but I haven't found anything. I wonder if you could actually put multiple modules in the aircraft and have them all communicate with each other in a kind of mesh network with the transmitter sending the signals, and the multiple receivers operating in a mesh to get the signals to the receiver with the servos attached.

One more question: How are you programming the Jennic module? Do you have a development board or did you build a programmer?

Apr 15, 2008, 02:59 PM
Registered User
I've been a bit swamped with the day job for a while but hopefully should get moving on this again soon.
It would be fairly simple to connect a couple of these modules together to get some rx diversity. My thoughts on using two tx is that you save weight and cost from the bit that is most likely to end up lost. For a more robust park flyer you could have two tx and 1 rx and for more serious stuff you could have two tx and two rx.
I did not buy the jennic dev kit, I simply soldered wires directly to the sides of the module and connected them to a pololu logic level usb serial device to program it.
I have been very impressed with the on board processor and so far have seen no need to use an external one. The one thing I have realised is missing is a watchdog timer, for my rx I have been looking for a simple watchdog part, any suggestions anybody? Also I still don't have a clue as to capacitor values to put across servo outputs.
Apr 15, 2008, 03:20 PM
Registered User
A Sot23-6 PIC 10F202 or 206, perhaps an 8 pin 12F, could poll the Jennic for something sensible perhaps via SPI and reset it if it fails. The micro just needs an internal xtal to save space; SPI does not have to be hardware based.
Apr 23, 2008, 06:08 PM
Registered User
Hahaha all you need to do is stuff it in a box and put a antenna that can go 2 directions and all the Hitec and Futaba pilots would love you...

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Idea Spektrum DX6 DIY Transmitter antenna replacement using 2.4GHz Wireless network parts? jaskew Radios 18 Jun 17, 2009 12:28 PM
Discussion Sort of DIY 2.4Ghz Spectrum Analyzer village_idiot DIY Electronics 11 Feb 23, 2008 11:55 AM
Discussion DIY +12dBi, increase the range of your rubber ducky 2.4Ghz antenna dalbert02 Aerial Photography 1 Jan 01, 2006 11:58 AM
Antenna for 2.4GHz video transmitter Wile E Power Systems 7 Sep 25, 2001 04:42 PM
2.4Ghz xmit system for a small fortune jas_Qfix Electric Plane Talk 28 Jul 02, 2001 08:37 AM