I would like to build an wattmeter with ACS714 and i need a low cost RF Transceiver that i can build inside each wattmeter so that it can transmit to another Transceiver it’s value. Since i need to build many of these then what is the cheapest one i can get ? I have following requirements :
Must be lowcost
Need a range that can cover a house (big)
Low powered
Multiple items must be able to exist close to each other without interfering.
Be able to connect to Arduino
I intend to have everything going in a prototype then be able to put an microcontroller, transmitter and the ACS714 on the same PCB so i can make some cheap wattmeters for my home and my parents.
If meshed and integrated into a whole house sensor network then you should look into the ZigBee protocol. Digi Int builds XBee modules (sold by SparkFun and others) that run the ZigBee protocol and are ideal for such a network.
If you are going to build thousands of wireless Wattmeters then you could use one of the ZigBee chip sets to get the price down.
If you want to go really cheap on a RF link then you will need to design a protocol and use a processor coded with the protocol.
waltr:
Is the RF links to be point to point or meshed?
If meshed and integrated into a whole house sensor network then you should look into the ZigBee protocol. Digi Int builds XBee modules (sold by SparkFun and others) that run the ZigBee protocol and are ideal for such a network.
If you are going to build thousands of wireless Wattmeters then you could use one of the ZigBee chip sets to get the price down.
If you want to go really cheap on a RF link then you will need to design a protocol and use a processor coded with the protocol.
Thank you waltr for the reply.
I guess when you say mesh network you mean if they ought to talk to each other and extend the coverage. If that is the case then no. I guess with RF it should be fine because all can call to the main transceiver to send their data.
When you say i need to design an protocol then what do you really mean ? As far as i can understand i should be able to talk directly to other RF transmitters from the receiver through different frequencies or ? I mean every transmitter is indentified by some frequency then i can get the data from it.
You need to go look at what RF modules are available. The very low cost modules do not tend to use different frequencies so two modules transmitting at the same time will interfere with each other.
Also, the very low cost modules use OOK modulation with doesn’t work well with just serial data such as async out of a UART. The data needs encoding (Manchester is commonly used) plus a starting sync values for the Receiver to lock on before it will receive the data properly. Then there should be some error check (Checksum or CRC) to ensure the data is correct. A more robust system would use transceivers so that it can receive an ACK (acknowledgment) back from the receiving end and so it can re-transmit the data if it gets a NAK.
If there are more than one unit sending data there should be an ID value sent so the receiver knows which unit the data is from.
Then there needs to be a way to prevent and handle collisions (two transmitters on at once).
All of this is what I mean by designing a protocol if you use very low cost modules.
Read the threads in this forum for more info and discussions about wireless.
I think Zigbee is definitely the way to go for a low speed house wide network. You don’t need to mesh them, though that would give you greater range.
You can use one as a master, and query for all listeners, then turn on one at a time for communication. Its done by AT type commands, like the old modems.
All the low level RF stuff will be handled by the modules and builtin software. You can start with modules, if you’re building more than 25K or so, you can switch to chipsets. But you get a lot for the price of the module.
viskr:
I think Zigbee is definitely the way to go for a low speed house wide network. You don’t need to mesh them, though that would give you greater range.
You can use one as a master, and query for all listeners, then turn on one at a time for communication. Its done by AT type commands, like the old modems.
All the low level RF stuff will be handled by the modules and builtin software. You can start with modules, if you’re building more than 25K or so, you can switch to chipsets. But you get a lot for the price of the module.
My problem is that they are quite expensive i mean the modules. Can you please enlighten me what i need ?
Do i just need many of these ? Is that all or i need something to program them ? Can i get away cheaper if i solder the modules or something like that ? When you say 25K i guess you mean more than 25000 pieces ? I guess i will be building around 20 devices. Would i be able to runt the xbee together with an Atmel ATmega168 or ATmega328 directly so i can cut cost by making some custom boards without all the stuff needed in an Arduino ? I could maybe build one of these instead https://www.adafruit.com/products/72 and use that together with the xbee and the ACS714.
20 pieces for a hobbyist is a lot, but in the engineering world that’s a small proto run.
The Digi modules are a bit pricey, but they are also what I would call a module on top of a breakout board. You can buy the modules seperately from other vendors for around $22. Digi has promised just the module (was suppose to be available July 2010, but I still haven’t seen them yet).
For that $22 you get a tested radio with software to handle all the low level communication protocols including frequency hopping, error recovery … Like I say a lot of engineering for the money. I see MicroChip has a PICtail version for $10, and maybe they offer the software for the protocol stack to run on a PIC, but then you’d be spending time on getting that going, and whether you have any room for your app on the PIC is another story (someone here may have used it).
The saying in the engineering world is you can have fast, cheap or well designed, pick 2
One alternative the comes to mind … not that’s it’s better, just an option to look into … is powerline networking. Since your devices will be near powerlines, you might as well use that to your advantage. I don’t know about the cost of these products vs low cost RF transcievers but the concept was to be cheaper and less prone to blocking due to walls, ceilings, floors, etc than RF. I’ve not used any of the devices so I can’t say from personal experience whether the reality comes close to the sales pitch. The bandwidth of these things is generally less than you’d get with RF but given your intended usage that may not be a problem. Anyway it’s something to think about.
Bluetooth is nice because it has RS232 so it is easy to communicate from Arduino, Bluetooth has also the advantage that many devices can transmit at the same time without interfering but it lacks in range beeing i think about 10m.
The second option is RF but here there are problems with identifying an device, sending at the same time, protocol etc. What it’s good about it is that i can extend the range by increasing voltage.
I could use RF if i could solve the problems with concurrency and device identification. Is there any already made protocol which supports this ? The concurrency can that be solved ? I mean i do not want to change frequency for every single one because that would mean that i need an receiver for every frequency so that’s not viable option for me.
i found this : http://www.glacialwanderer.com/hobbyrobotics/?p=291 and it looks promissing. It sound like it solves the problem with concurrency but is that true ? I mean would it not interfere with eachother since two devices will transmit at the same time on the same frequency ?
relatively low cost for low volume ($20each in 25 quantity)
I know that but it’s still way too expensive for this simple use. I have ordered 4 pieces of bellow transceiver for 14USD so way bellow 80USD which would cost me for xBee modules.
Again as i have mentioned i need Low cost. With the above and with the http://www.arduino.cc/playground/Interf … e/Nrf24L01 and other libraries for Arduino i think i have it coverred. I would like somebody confirming that i am right about this module.
leon_heller:
Those are what I’d use. They are similar to the much more expensive ones made by SFE which I’ve used in the past, but have a different connector.
I’ve designed a PCB with a PIC18 on it for them, I should get the boards in a week or so.
Yeah. Please do post when you do so. I am interested to see it.
I don’t have any specific knowledge of that module, and its probably fine for your application. But you do realize that the one on ebay is a discontinued version, so grab as many as you need with some spares. As when you go to add more they will be around $20.
viskr:
I don’t have any specific knowledge of that module, and its probably fine for your application. But you do realize that the one on ebay is a discontinued version, so grab as many as you need with some spares. As when you go to add more they will be around $20.
That is not entirely true. The nRF24L01 is discontinued but not the nRF24L01+ which is the one i have bought for 3.50USD. If you check the listing it says
1x NRF24L01+ 2.4GHz Wireless Transceiver Module
. I pressume this is correct but will se when i receive the product
I would like to build an wattmeter with ACS714 and i need a low cost RF Transceiver that i can build inside each wattmeter so that it can transmit to another Transceiver it’s value. Since i need to build many of these then what is the cheapest one i can get ? I have following requirements :
Must be lowcost
Need a range that can cover a house (big)
Low powered
Multiple items must be able to exist close to each other without interfering.
Be able to connect to Arduino
I intend to have everything going in a prototype then be able to put an microcontroller, transmitter and the ACS714 on the same PCB so i can make some cheap wattmeters for my home and my parents.
I guess i will be building around 20 devices. Would i be able to runt the xbee together with an Atmel ATmega168 or ATmega328 directly so i can cut cost by making some custom boards without all the stuff needed in an Arduino ? I could maybe build one of these instead 