RF, Zigbee, SMAC

Go To Last Post
23 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've been thinking about adding an RF interface to a project of mine to make it easier to use without wires to the PC. The 2.4GHz ISM band looks easiest for international use and simple antenna design.
Yesterday I found the site of the split off company from Motorola called Freescale at www.freescale.com. I thought I'd try for some samples and filled out an request for 4 of the MC13191 RF tranceivers. While I was there I asked for some accelerometers aswell. Today I got a mail saying they have shipped 4 of each which I thought was quite generous of them.
It doesn't look like a huge amount of work to get the RF chip SPI interface to an ATmega working but its quite a way from trivial.
I thought I'd mention it here as it seems an inexpensive way to get into embedded wireless systems and perhaps if others are interested we can share code, bugs etc. :-) I don't want to get into Zigbee at the moment - just a simple point to point protocol so I'm thinking of setting up a couple of test boards with ATmega32 chips.

Ralph Hilton

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It OUGHT to work. As I recall, those radio ICs are really built to with with quadrature up/down converters. But, often, those "companion" chips are designed to work with a specific protocol, at least at the RF level. This may make it difficult to do something that is not Zigbee.

By the way, Freescale is just Motorola Semiconductor. All of it. Except the discretes and basic logic spun off to multiple purchasers several years ago. Except they kept RF power discretes. Motorola retained telephone/cellphone, radio, and similar busineses.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The MC13191 is advetised as not using Zigbee which is why I chose it. They have C code on ther site for their µcontrollers which looks modifiable and much simpler than a full zigbee setup.

Ralph Hilton

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You may want to have a look at the Nordic rf devices if you want to roll your own comms protocol.

Keep it simple it will not bite as hard

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just took a look and it looks pretty cool!

It says it is a full transceiver and has a full packet modem with a 128 byte payload at 250KBps.

I have been looking for something like that for a long time! Wow. It seems to be the transceiver that is used for Zigbee.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Outchh

32 pins on a 5x5 mm chip.

8 pins on 5 mm , not for my "HomeWork" :-(

damm :-)

What do you guyzz do to solder and make PCB for those ???

/Bingo

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

...32 pins on a 5x5 mm chip...

Are you referring to the MC13191? If so, the Freescale dev. boards aren't too pricey, as they are pushing their chip series pretty hard.

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you wan't 2.4GHz then you must use a "complex" transceiver such as IEEE 802.15.4 (Zigbee Physical / Mac layers). Your puny little signal doesn't stand a chance of being received in the face of such heavyweight competition as Bluetooth, Wi-Fi, Wireless-USB, microwave ovens and in the USA there are cordless phones to deal with.

It takes some believing but 802.15.4 does actually work in the presence of all those other users of the 2.4GHz band. (I haven't tested Wireless-USB yet)

BTW If you use a 802.15.4 transceiver then you have the option of writing a Zigbee stack, debugging and testing it and then of course sharing it with the freaks. :) :)

Nigel

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm looking at applications needing only a few meters range without intervening walls so the low power devices should be quite useable. At the moment I don't envision needing anything as complex as a full Zigbee implementation.

Ralph Hilton

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you are interested in the Freescale Zigbee parts without the trouble of making a breadboard, you could look at:
http://www.maxstream.net/zigbee.php
Supposed to ship next month.

Scott

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ralph, I think you missed my point.

I think you'll need a decent (probably 802.15.4) transceiver for good comms reliability, as expained above. However you certainly don't need to write a full Zigbee stack. The simplest protocol will do. (Although I'm sure there are many freaks who would be very happy if you did)

Nigel

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is there something about the MC13191 that makes you think it is inadequate?

Ralph Hilton

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Did you see www.kcwirefree.com?
Bluetooth modules with integrated antenna, and USART / USB inrterface (plus GPIOs). I'm experimenting with them since a few days and they are simple to use.

Julián.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

But what's the price tag? I daresay its quite a bit more than making my own modules with an MC13191. Having gone through the datasheets more an ATmega88 is probably quite enough to handle the processing.

Ralph Hilton

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm in Argentina, the prices to get them here are different than yours. Write to them just as I did, I'm sure they will answer you.
Regards,
Julián.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How about answering my question?? Prices don't vary a lot. Or are you working for them?

Ralph Hilton

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

About u$s 35-40 for a Class 1, but buying at least 10 (this price is not new, It's of a few month ago, so I do not really know the current price).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

rhilton wrote:
Is there something about the MC13191 that makes you think it is inadequate?

After a quick browse on the Freescale site my answer is. No !
It has all the required bits expected of my "complex" transceiver as mentioned above.

However why dont you use the MC13192. It looks identical to MC13191 except that it confirms to IEEE 802.15.4.
Look also at MC13193 which has a complete Phy/Mac and Zigbee stack built in.

I imagine the 802.15.4 / Zigbee stuff will sell in far greater numbers than MC13191 so will probably be cheaper.

(I cannot understand the thinking behind MC13191, perhaps it was Freescale's first big toe in the water for Zigbee and they simply got it wrong. I wonder if it will be obsoleted!)

Nigel

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The MC13191/2/3 were intended as degrees of increasing features. The 13191 - I have yet to verify this - should be able to operate with around 2k flash and only a small RAM whereas the 13193 [edited] full Zigbee implementation needs far more.
It would seem like overkill to implement Zigbee on a mouse, for instance..
The simpler MC13191 means faster development on a smaller processor. Full 802.15.4 compliance isn't really needed for my apps at the moment - it would just be bloating the firmware with useless functions. The PHY and SMAC layers are enough.

Ralph Hilton

Last Edited: Mon. Apr 25, 2005 - 11:26 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Aah I think you've misunderstood,

IEEE 802.15.4 refers to the physical layer and media access layer only. Zigbee refers to everything else above. Therefore 13192 should be no more complex to program for than 13191.

I haven't read the datasheet but from looking at the summary spec the chips both do the same job so probably have very similar registers etc.

Nigel

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sorry - I typed in one of the numbers wrongly.
The 13191 registers are a subset of the 13192.
13191 = 802.15.4 PHY layer $2.28 in qty 1000+
13192 adds functionality 802.15.4 MAC layer which is implemented in firmware $2.70 in qty 1000+
13193 adds Zigbee support $3.26 in qty 1000+
Programming the 13192 should be just as easy as the 13191 if only the SMAC is used.
Programming the 802.15.4 MAC layer would add complexity and size.
So I just want the PHY layer and SMAC software which the 13191 can handle ok. I don't see the point of buying a more expensive chip with functions I won't use even if its only $0.83 (Digikey)

Ralph Hilton

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Do not forget the Chipcon device.
This has a slight edge on the Freescale part in that it switches the RF path internaly so only one antenna is required.

Keep it simple it will not bite as hard

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You don't need two antennas but switch the signal using a GaAs analog switch, the transceiver should have a output you can use as a control line.

Nigel