Hi i am currently want to use one arduino to talk to many slaves, the idea is that the slaves each operating a small LED panel. The idea is that the master sends a small wireless signal and the slave will know what to display, currently its one master with about 10 slaves.
The range of the master and the slaves is about 30-50meters max, this could be more in the future.
What type of wireless device should i be using? I read alot about UWB but i dont think its suited for my use, what do you guys think?
I've made something like that using generic nRF24 modules. I have one base station controlling 3 auxiliary devices, but it's expandable. A dozen or so would be feasible, depending on your specific requirements.
How small of a signal? (I assume by 'small' you mean it's a small data packet?)
How fast do the auxiliary devices need to respond; i.e. how much delay is acceptable between the controller sending the packet and the aux device displaying the new data?
Do the aux devices need to be strictly synchronized, or is it OK if they don't update at exactly the same millisecond?
The idea is that everthing needs to be as fast as possible, delays wouldnt be feasable. The packet indeed is just a small data packet with for example what color the panel should be. The devices dont need to be synchronized although this would be cool ofcourse, it just isnt a requirement.
The packet would contain the address of the auxiliary and the command. The auxiliaries would be listening all the time. You haven't mentioned if there needs to be an acknowledgement back to the base station.
The protocol could be a unique address for each remote and an address for “everyone.” I would start off with the controller base station and one remote display. Get it working with the one remote.
You could either send an address/command and assume that the remote received it or you could wait some maximum amount of time for a response from the remote acknowledging receipt. Two way communication obviously complicates matters and will affect the time it takes to command all the remotes. You could simplify matters by sending a remote’s command multiple times to improve the odds that the command was received.
If you needed to synchronize the remotes you would send out all the commands and then send the command for “everyone” to execute.
For starters a simple one way communication would be the easiest, in the end a simple handshake would be best, send the command to the panel, panels sends back an OK or a failure command to let the base station know if the command was a success or not.
More would be 100- 300 meters max, maybe some important info is that i would like to use it outside aswell, currently im using WIFI to talk to the panels, this has its benefits but the range and delays are too much for my use.
That rules out nRF24 but also anything BLE-based; HC12 might/should work but I have no experience with it. LoRa may or may not be fast enough if you expand the network with more nodes, but I'd expect it to be OK for a dozen or so.
There's always a delay. A human observer will not generally notice a delay of up to 50ms (or a little more) and in most cases delays of 100-200ms are still experienced as fairly smooth if you have several displays/panels spread around the same space. So the question is how much of a delay is acceptable, given the fact that there will be a delay, always.
Will the same person in this space see several of your panels/displays at the same time? What kind of environment is it and what dictates the maximum manageable delay?
I'd act on the default that yes, there needs to be. If there's no ACK, you can't know for sure if the auxiliary device has received the command. Since delays must apparently be kept to a minimum, I assume that lost frames are an absolute no-no. Then again, maybe not...?
And the viability of a send and ack system, especially if the radios are close to the distance limit, really depends on how quick this 'system' needs to react.
Precisely, which is why I jumped on the delay & sync stuff. It becomes a concern especially once the number of nodes grows, the environment is more noise (radio/EMI-wise), etc. etc.
So the environment is basically a karting track, as said this can be indoor or outdoor. There can be some walls or other stuff in the way but most of the time this is a quite open space. The delay talked about should be within 0,5-1 sec as any longer the delay would be so much that the panel will be lit up to late.
The size depends, indoor tracks are usually smaller but sometimes have walls, this ofcourse wouldn't be ideal but here in the Netherlands the tracks im focusing are quite open and only have like steel beams or tire walls standing in the way.
Outside ofcouse is more open, here there shouldn't be as much inteference with other stuff like wifi signals etc.
I think that there will be about 6-12 panels in most cases.
Let's see, steel beams will be interfering with the signal. Tires with steel cores will be interfering. Why not just develop a test communication system and try it long before developing any other parts that will be irrelevant unless the wireless system works?