EEVblog Electronics Community Forum
Electronics => RF, Microwave, Ham Radio => Topic started by: Psi on September 11, 2021, 09:09:40 am
-
Does anyone have any recommendations for ~$2 RF modules to implement reliable glitch-free control over 50m with low enough current to run from 2x CR2032 for ~30sec (no more than 25mA transmit current)
I've been testing with aliexpress 4ch 433mhz modules but they always seem to be glitchy when you walk around with them while transmitting. Even with good antennas you occasionally get a glitch that causes the RX output to stop for 200ms.
I think their internal data rate is just so slow that any glitch results a noticeable period of off-time that cannot be filtered out externally without causing human-perceivable lag in the remotes response time.
I think the solution is probably a data module rather than a 4channel module. With a start/run/stop packet to decouple signal reception yes/no from output on/off.
Any thoughts? or recommendations. Ideally links to known good 433mhz data modules with a metal can, but i'm open to anything.
-
Missing tx data and slow to start up? Check power stability, decoupling and battery voltage, so that it not exist fast or slow voltage drops.
Continuously transmitting while walking around => From time to time will it be interference when reflections cancels out each other.
Weak signal level due to poor antenna efficiency and low TX level and non LOS will increase problem.
Cheap TX units may be in-stable and less precise in frequency and frequency stability.
3-pin modules for $0.10 is just oscillators that is very depending on its surrounding. Do not use if function is important.
These thing are measurable.
Better then guessing wrong for each evaluated TX module in infinite loop.Cheap and good quality is often different things.
Metal can for $2 and you will probably need to dig deep in chinese market to find anything useful.
Metal can is hardly needed at 433 MHz if you design PCB well and most keyfobs do not use metal can.This can be a good start if cost is important: https://www.mouser.com/ProductDetail/Microchip-Technology/MICRF114T-I-OT?qs=%2Fha2pyFaduhnQWX%252B%2FToJdq4MCV4Ig85buIxADa3NHnM= (https://www.mouser.com/ProductDetail/Microchip-Technology/MICRF114T-I-OT?qs=%2Fha2pyFaduhnQWX%252B%2FToJdq4MCV4Ig85buIxADa3NHnM=)
-
I think you have to make a choice.
buy the cheap stuff and live with it's limitations, or buy something decent and pay the price for that.
You can also buy the cheap stuff and then spend a significant amount of time and effort to setup a redundant protocol with error detection and automatic resending of packets.
-
The nicest transmitters i've found are Linx TXM-433. They are not cheap -- $12 or so -- but are good quality with nice features.
The cheap ones are hard to pass up at $2 or less even, but I have found them highly variable in power output, output impedance, and most frustrating for me, start-up time. Startup time varies from 10us to as much as 80us unit to unit and this affects signal timing. I have fallen into the habit of individually characterizing the startup time and putting this in a calibration constant for the signalling protocol. It's a major PITA, but seems to be sufficient for my needs.
If you find a source for cheap consistent modules, please post here.
I don't know what you use for receivers, but the Linx receiver is a nice one too. One feature it has is an RSSI indicator output that is really useful in debugging and unit placement. Most of the cheap receivers are actually quite good superhet designs, but lack RSSI output. I found some based on a Princeton PT4301-X superhet receiver chip that I've managed to tack a line onto one of the signals to get an RSSI output, but that's a major hassle and I've destroyed more than one attempting the job.