Developing a universal module

More
19 Jan 2014 16:27 #18690 by vlad_vy
Replied by vlad_vy on topic Developing a universal module
It looks like it was error in Hubsan protocol. With latest firmware, rewired XL7105-D03 module (GIO1->TXEN, GIO2->RXEN) works fine.

Please Log in or Create an account to join the conversation.

More
21 Jan 2014 22:00 #18873 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
I just saw this:
blog.cyberexplorer.me/2014/01/sniffing-a...ng-nrf24l01-and.html

This is (potentially) half of the holy-grail we've been looking for. It should allow scanning any protocol from any 2.4GHz device at low cost. Of course, it can't transmit, so you can't use it as a universal module, but it could make protocol analysis much easier.

Please Log in or Create an account to join the conversation.

More
21 Jan 2014 22:17 #18876 by SeByDocKy
Replied by SeByDocKy on topic Developing a universal module

PhracturedBlue wrote: I just saw this:
blog.cyberexplorer.me/2014/01/sniffing-a...ng-nrf24l01-and.html

This is (potentially) half of the holy-grail we've been looking for. It should allow scanning any protocol from any 2.4GHz device at low cost. Of course, it can't transmit, so you can't use it as a universal module, but it could make protocol analysis much easier.



Nice project :)

Please Log in or Create an account to join the conversation.

More
21 Jan 2014 22:41 #18879 by cstratton
Replied by cstratton on topic Developing a universal module
Certainly looks like a fun setup, but a bit much with the downconverter and software processing for embedded use.

For analysis I'm not actually sure it's superior to using a logic analyzer on the SPI interface. One thing is that as soon as you start dealing with raw RF, you see all sorts of things which aren't your traffic of interest. It might help with identifying unkown chipsets without opening the case, but with the exception of the chip-on-board Bekin/Hope systems (unlabeled dot of epoxy), that's not really very hard.

In terms of actually making a universal module, the key question would be if there's either a chip that is versatile enough to do everything, or else something that could function as a sufficiently frequency agile up/down converter to allow a $10-20 class FPGA (or possibly even very fast processor) to do the modulation/demodulation at a sample rate consistent with only the instantaneous channel bandwidth, and not the 100 MHz bandwidth of the entire ISM band. If not, there are faster FPGA solutions which could handle the whole bandwidth, but they start getting more expensive and needing more expensive ADCs.

But most of the modules in use are really cheap...

Please Log in or Create an account to join the conversation.

More
21 Jan 2014 23:13 #18881 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
the problem is the sheer # of them. Today you need 4 modules to support the protocols we have (CC2500, CYRF6936, A7105, NRF24L01). The CC2500 modules available apparently don't support the FrSky protocol (I never figured out why, but it seems to be related to frequency deltas, and I think the frsky chips may use a skewed oscillator). There are also a few (like the futaba ones) which are difficut to source. The cyrf is not cheap (at least when paried with a PA). It becomes difficult to really have a universal module without a backpack of extra hardware. The design I've got now with the AVR switch, seems like it'll work for us, but it is far from universal.

I already have a Ettus SDR, but I've never found the time to really learn how to use it properly. I got far enough to know that the CC2500 w/PA I'm using has an output that looks nothing like the result from the real FrSky module (even though the non-PA module I have matches, and works just fine with frsky)

Anyhow, being able to get into 2.4GHz SDR for ~ $50 is pretty nice.

Please Log in or Create an account to join the conversation.

More
23 Jan 2014 07:01 #18979 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
I got my board back from OSH Park today. They are really nice, as everyone said they'd be.
I found that the set of holes for the nrf24L01 were too small, so I had to drill them out (luckily I have sub 1mm drills for the task). Other than that and the known issues I already found, it seems ok. The board is quite compact when fully populated. My idea for surface mounting the cc2500 and a7105 with the AVR mounted on the back didn't work quite as I intended due to th pins sticky through the board too far. It looks like I'll need to cut the pins on the AVRso they'r flush with the bottom of the board as clipping them afterwards didn't work at all. Using a surface mount package for the AVR would probably make the most sense actually. But it would probably put the project outside the realm of most DIYers. Once I solder the modules down it'll be pretty hard to remove them, so I'm thinking I'll just spin another rev of the board fixing the issues I found and skip using these. I think the board I have now is probably still within reason for anyone witheven a basic amount of soldering skills.

Please Log in or Create an account to join the conversation.

More
26 Jan 2014 01:32 #19125 by Pattaya01
Replied by Pattaya01 on topic Developing a universal module
I have 2 questions:

1. Will this module be compatible with the Devo12S?
2. If yes, will it maintain telemetry functionality?

Please Log in or Create an account to join the conversation.

More
26 Jan 2014 04:11 #19129 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
I ordered the ver1.1 of my boards today after fixing a variety of issues. I added support for the xl7105-sy (no PA) since I could do so without making the board bigger. I also fixed the drill-holes and wiring issues. Hopefully that will take care of all the issues.

The board will be compatible with all radios supported by Deviation. On the Devo 6/8/10/12, I'd recommend leaving the stock module in place and adding this (it mounts in basically the same way as the previous modules)

For the Devo7e, you could choose to buy a new CYRF6936 module (at ~$25) to get the full output power without the tricky diode mod.

For the X9d (still in development), you need to add the $25 CYRF6936 if you want to use the DSM, Devo, WK, or NineEagles protocol.

The module does not need to be populated all at once, but due to its design, you must add the A7105 module before soldering in the nrf24l01 module, or you need to use a socket for the nrf24l01 (thus increasing the thickness significantly, and maybe making mounting more difficult)

Please Log in or Create an account to join the conversation.

More
27 Jan 2014 11:40 #19265 by mpicpu
Replied by mpicpu on topic Developing a universal module
can we order this board yet

Please Log in or Create an account to join the conversation.

More
27 Jan 2014 14:48 #19273 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
nope. I'll likely have the next iteration some time next week. If they work as intended, we'll see what the next steps are.

Please Log in or Create an account to join the conversation.

More
04 Feb 2014 05:41 #19813 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
I got the Rev1.1 boards today and they seem to be basically functional. I have only wired up the a7105 so far, but it binds fine. I don't have the headers to hook up the nrf24l01 or cyrf6936 yet so those may need to wait, or I may jury-rig something. For some reason the holes for the 2mm headers are smaller this time than last time, and the male pins don't fit (but female headers do, so I'm ok for now)

Please Log in or Create an account to join the conversation.

More
04 Feb 2014 07:10 - 04 Feb 2014 07:11 #19820 by SeByDocKy
Replied by SeByDocKy on topic Developing a universal module
Hi PB,

A lot of deviationx's users are waiting very attentively such universal module. Definitively since your return into the scene, the deviationx developpement have again a big boost.

I noticed, a new RF chips appearing on some toy quad, the LT8910 (precisely the LT8910S5C)

www.nst-ic.com/f/1/1311114648.pdf

Do you already know it ?


Attachments:
Last edit: 04 Feb 2014 07:11 by SeByDocKy.

Please Log in or Create an account to join the conversation.

More
04 Feb 2014 19:32 #19831 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
That is interesting. It looks like it may overlap in functionality with the CYRF6936 such that in some cases you may be able to use one chip to talk to the other. I don't read Chinese, so I can't verify that, but it looks like it supports GFSK at 1Mbps, which is something the CYRF6936 can do. There are other modes that may not be supported, so if it is possible at all, it is likely protocol specific.

Please Log in or Create an account to join the conversation.

More
04 Feb 2014 19:54 #19832 by SeByDocKy
Replied by SeByDocKy on topic Developing a universal module

PhracturedBlue wrote: That is interesting. It looks like it may overlap in functionality with the CYRF6936 such that in some cases you may be able to use one chip to talk to the other. I don't read Chinese, so I can't verify that, but it looks like it supports GFSK at 1Mbps, which is something the CYRF6936 can do. There are other modes that may not be supported, so if it is possible at all, it is likely protocol specific.



Whatever, the performances of this RF chips seems to be not as good as for the nRF24L01


SOC solution with integrated on-chip components including transmitter, receiver, frequency synthesizer, GFSK modem
Power-adjustable transmitter with a max transmit power of 6dBm
Lower-middle-frequency receiver with spread-spectrum digital communication mechanism and -96dBm sensitivity
Energy detection of digital channel for channel quality monitor
Data rate: 1Mbps / 250Kbps / 125Kbps / 62.5Kbps
Transmit distance: 200m
Supply voltage: 1.9V to 3.6V
Operating temperature: -40℃ to 85℃
Sleep mode current:1μA (while maintain the register values)
Frequency-Hopping Spread Spectrum
SPI / I2C interface
Built-in CRC, FEC, auto-ack and auto retransmit
QFN24(4*4mm) and SSOP16 Package (in accordance with RoHS)
Applications: remote control, wireless keyboard and mouse, wireless networking, intelligent home furnishing, industrial and commercial short-range communication, IP phone, cordless telephone and mutual communication between machines

Please Log in or Create an account to join the conversation.

More
04 Feb 2014 20:45 #19833 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
The availability of the SSOP package is nice. So far every transceiver I've seen has been QFN which is a pain to deal with as a hobbyist.

Please Log in or Create an account to join the conversation.

More
04 Feb 2014 23:03 - 04 Feb 2014 23:07 #19840 by Hondawasaki
Replied by Hondawasaki on topic Developing a universal module
This topic is very interesting . :blink:
was wondering if it would be possible to design the board in two pieces.
and connect them together using a cheap ribbon cable of sorts .
thus making each Half circuit useless with out the other half and maybe not law defiant. as a single board would do nothing making the individual boards legal. It would be up to the buyer to obtain both and put them together.???
possibly made by 2 different manufactures
Last edit: 04 Feb 2014 23:07 by Hondawasaki.

Please Log in or Create an account to join the conversation.

More
05 Feb 2014 03:14 #19847 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
Let's put it this way:
I am personally not going to play games with the FCC.
I am not about to pay for a lawyer's services to determine what would or would-not be a violation of the law in this regard.

Also, actually soldering the board is a hassle, and I seriously doubt that we could create enough volume to warrant sending it to a pick-and-place machine (besides that the board isn't designed in a way you could use such a machine), so I am not interested in trying to produce them for others.

So if you have a soldering iron, and a little patience, putting it together yourself shouldn't be too hard, or perhaps you can find someone willing to sell them pre-assembled, but it certainly isn't me.

I'll get some time later this week, and will fully assemble the 1st board with headers. If everything works, I'll try actually mounting the parts without headers (the way the board was designed), and will report back.

Please Log in or Create an account to join the conversation.

More
05 Feb 2014 20:59 #19894 by nusbr
Replied by nusbr on topic Developing a universal module
Of its working, could you please send me some pictures of them? Is it also possible to get the schematic? That would be very helpfull for me. Tank you

Please Log in or Create an account to join the conversation.

More
06 Feb 2014 04:36 #19907 by PhracturedBlue
Replied by PhracturedBlue on topic Developing a universal module
I verified today that everything but the cyrf module is working. Due to the board layout, I can't easily plug the board in with the ribbon cable I'm using.
To use the AWA24 board, I need to spend some time working out the differences in board configuration, so that'll take some time. But so far so good.

Please Log in or Create an account to join the conversation.

More
07 Feb 2014 07:19 #19964 by nusbr
Replied by nusbr on topic Developing a universal module
Does i understand this right? GIO1 is connected to TXEN, and GIO2 to RXEN? Are the other connections as described in Moduleinstallation.pdf?

In TX.ini, i activate the module.
with enable-a7105
Using TMS: ‘A13’(Is this the Chip enable Signal?)
And for what is the "has_pa-a7105" command?

Sorry for those many questions and
thanks for answer

Please Log in or Create an account to join the conversation.

Time to create page: 0.149 seconds
Powered by Kunena Forum