Noise from multiple cheap RF transmitters?

Hi. I'm working on a project which involves eight Nanos communicating two way with a ninth master unit. The master polls each of the slaves in turn, which immediately reply with a small string of data which the master processes and updates other slaves accordingly. I'm using the cheap 315MHz eBay RF units and VirtualWire.

Initial tests using one-way communication gave me very reliable data transfer over a range much greater than required for my final project. However, now I have multiple devices at play and transmissions going both ways, my range and reliability has dropped significantly. Even when the units are only a 2-3 meters apart and in clear sight of each other, I can lose most of my broadcasts.

My question is this: do the TX modules create background noise when powered up but not broadcasting? Is it possible that several powered TX units in a confined space are creating a wall of background noise that the real signal is struggling to get through?

Should I power the TX unit from a pin, so I can switch it off when it's not broadcasting?

Should I power the TX unit from a pin, so I can switch it off when it's not broadcasting?

Try that and let us know how it goes. Some of the TX modules that I've worked with do not transmit at all when the data line is low, but others may leave the oscillator running and, as you suggest, generate some noise.

Thanks - I'll do that and report back.

I guess there's no way to 'listen' to what's happening on that frequency when nothing's being broadcast?

Tried switching power to TX unit on when it’s transmitting and it has made no difference to the overall reliability.

I only have two units running - if they’re placed within six inches of each other, practically no drop out. Move them two feet apart and almost nothing gets through.

What kind of units? Are you using antennas? Antenna wire for 315MHz is like 23cm long I think. Check the device datasheet.

These are the units: http://www.ebay.co.uk/itm/261355642647

Yes, antenna should be 23.8cm long.

However, many have reported successfully using these without an additional antenna, and my initial tests with one-way communication gave me range of 5m or more without antennas. I’m wondering why the range has dropped so dramatically now that I have multiple units in play at any one time.

I have tried adding antennas - they seem to improve performance slightly, but still nowhere near what I need to make this work.

Aren't those A.M. units (not real good for dependable data traffic)? Try polling your remotes. Means you nees a transmitter and reciever for each unit. Keep your remote in recieve mode until the master sends a code unique to the reote, then have the remote send it's data and wait for a successfull ACK from the master.

Or, go to something like the NRF24L01's or XBees. costs more, but much easier to ger working.

123Splat: Try polling your remotes. Means you nees a transmitter and reciever for each unit. Keep your remote in recieve mode until the master sends a code unique to the reote, then have the remote send it's data and wait for a successfull ACK from the master.

That's almost exactly what I am doing. As I say, it's just about the range - it works perfectly up close, but a couple of feet away things are very different.

From all I've read about these units, I should be able to get 10ft with out of them with 90% reliability. 90% is fine for this project - I can easily compensate for that with the software.

Could more than one be transmitting when it isn't supposed to be?

You mention trying 1-way with a single unit, does that also work with 2-way with a single unit?

Another possibility is to shield each of them in metal enclosures with an antenna for each, which should do a good job of isolating them all from outside noise.

In order to see if there is noise, if you have a gigahertz scope or spectrum analyzer available to you, you can just stick a wire (antenna) onto a probe and wave it around things and see what noise it picks up, especially . But since its at 315MHz, you'll need something that will handle those frequencies.

ArduinJoe: These are the units: http://www.ebay.co.uk/itm/261355642647

Those aren't two-way. Do you have a pair of them for each link?

I would use NRF24L01+s for this. It's what they're designed for. They have a 'listen' mode which doesn't transmit anything. They also have data transmission with full ack, retry, etc.

They're probably cheaper, too.

http://www.ebay.com/sch/i.html?_nkw=nrf24l01%2B

fungus:

ArduinJoe: These are the units: http://www.ebay.co.uk/itm/261355642647

Those aren't two-way. Do you have a pair of them for each link?

I would use NRF24L01+s for this. It's what they're designed for. They have a 'listen' mode which doesn't transmit anything. They also have data transmission with full ack, retry, etc.

They're probably cheaper, too.

http://www.ebay.com/sch/i.html?_nkw=nrf24l01%2B

Yes, I have a pair for each Nano. I have two-way comms working exactly as I want, just not with the range I want!

I didn't know about those other units - I thought all other options were considerably more expensive. Perhaps it's back to the drawing board (or eBay)...

Or yourduino.com:
http://yourduino.com/sunshop2/index.php?l=product_detail&p=188

Thanks. I'm in the UK, just picked up 10 for less than £14, delivered by Monday.

Isn't the Internet a thing of wonder?