Using Multiple Cheap RF Modules Together

Thread Starter

elec_mech

Joined Nov 12, 2008
1,500
I'm attempting to use six latching 4-Channel RF modules together for a project for another member. Unfortunately, I can't seem to get consistent results. I'm using three of the four channels to light up blinking LED's and, for two of the modules, play an audio file.

When all six are side-by-side, they work well. Push a button on the transmitter and the corresponding LED blinks on all six. Press the same button again, they all go out, perfect.

However, when I separate them, say three to fifteen feet (no walls between them, only open air in a basement), repeatability goes out the door. Sometimes they all work, but often 4-5 work and one or two miss a signal.

I've added a length of wire for a receiving antenna, beefed up the regulator, increased the voltage to the transmitter, etc., all to no avail. Sometimes these helped a little, but the problem never goes away entirely.

The RF modules come from China through eBay:

s-l140[1].jpg

Am I expecting too much from these? The problem is having six modules - if they all aren't receiving the signal, then some won't come on when they are supposed to and some will go on when they are supposed to be off which defeats the purpose of the project.

Is there a better solution? I've looked at other RF modules on eBay and they all seem to use the same RX/TX IC pair, so I doubt I'll get a different result. Should I look at Linx (hate to spend that kind of money, but if this doesn't work reliably, it does the other member no good and I'm out of a lot of money anyway)? Anyone have a positive experience with a minimum 3-CH RF system working on multiple receivers simultaneously and reliably?

Here is the original thread for anyone interested in more background.
 

MikeML

Joined Oct 2, 2009
5,444
I am using those same (315MHz) EBay modules to open my garage doors and my motorized swing gate. The secret to getting more than 300m (1000ft) range in open air is the receive antenna. I am using a full-length 1/2λ dipole mounted up high on the outside of my metal aircraft hangar.
 

Thread Starter

elec_mech

Joined Nov 12, 2008
1,500
I only need to get 25 feet in an office-like environment, but I can't get reliable (and repeatable) range at three feet in the same room. Would a properly designed receiver antenna fix this? If so, could you suggest a design to look at or should I go through my rarely perused 2008(?) ARRL book? RF is black magic to me, but I'm up for making an antenna if this will fix the problem. Thanks Mike.
 

MikeML

Joined Oct 2, 2009
5,444
I only need to get 25 feet in an office-like environment, but I can't get reliable (and repeatable) range at three feet in the same room. Would a properly designed receiver antenna fix this? If so, could you suggest a design to look at or should I go through my rarely perused 2008(?) ARRL book? RF is black magic to me, but I'm up for making an antenna if this will fix the problem. Thanks Mike.
remember 468/Freq?

I have a picture of the dipole I made on my other computer. I will post a picture of it later today...
 

Thread Starter

elec_mech

Joined Nov 12, 2008
1,500
I'll have to study up. I had to Google dipole antenna - I'm not a ham or RF guy, just collected a lot of books in my short lifetime. :)

So a dipole has two wires and looks like the old FM antennas that came with receivers back in the 80/90's? I assume one wire goes the ANT pad on the receiver. Does the other go to GND or elsewhere? Sorry to ask such newbie questions - I know a little bit about a lot. I've heard the term dipole, but I've never made one or studied RF design in detail.
 

MikeML

Joined Oct 2, 2009
5,444
Center-fed 1/2λ dipole consisting of two 1/4λ elements. In my case, remoted from the receiver using a piece of CATV coax cable. Center conductor of the coax cable goes to the ANT pad on the receiver end, and to the vertical up element of the dipole. The coax shield goes to the ground trace on the receiver and to the down element of the dipole. If attaching the antenna elements directly to the receiver, the two elements need to be staight, and in opposite directions.

468/315 or 468/433 is the overall length of the dipole in feet.
 

Thread Starter

elec_mech

Joined Nov 12, 2008
1,500
Something like this?
Antenna.jpg

I believe I have 315MHz, so 468/315 = 1.4857 feet. Therefore, the top half with the insulation and ground removed would be 1.4857/2 = 0.7428 feet or 8.91 inches, ditto for the bottom half and I could use a few more inches beyond that to attach to the receiver module? I'm not sure how to keep them straight and in opposite directions. So RG6 would work?

At a foot and half, I'm not sure this will ultimately be a practical solution since this going into a vet's office and patient rooms, but I'll explore it if it solves the reliability problem.
 

MikeML

Joined Oct 2, 2009
5,444
That is a vertically-polarized coaxial sleeve dipole antenna. It would work just fine, and can be made from a piece of coax that has a woven braided shield sleeve. You can remove ~ 1 ft of the vinyl jacket, and then pull the exposed sleeve back down the coax on the outside of the vinyl that is still there, finally trim the sleeve length to the calculated dimension and tape it or heatshrink it. The whip portion is just the center conductor with its insulating sleeve intact also cut to length.

Here are pictures of the way I build mine. Mine is outside in the weather.

Ant1.gif

antuc.gif

The receiver is through the wall. Total coax length is about 12".
 

tracecom

Joined Apr 16, 2010
3,944
I don't think signal strength (or the lack thereof) is the problem, and don't believe the fix is better antennas. My guess is that you are getting the equivalent of switch bounce, which is actuating the errant receiver, and then immediately turning it off.
 

Thread Starter

elec_mech

Joined Nov 12, 2008
1,500
MikeL, I like it. Very professional looking.

My guess is that you are getting the equivalent of switch bounce, which is actuating the errant receiver, and then immediately turning it off.
That had occurred to me earlier and perhaps I wrote that off too soon. Would this work as a suitable debounce circuit? The top circuit is how the transmitter works now.

Debounce.png
 
Have you tried tunning the transmitter. I was asked to look at my neighbours garage doors openers as he said they were faulty.

The doors did open but only if the transmitter was 2 feet (under 1 meter) away from the receiver. I spent several hours looking at the circuit before I opened the transmitter and retune the trim cap. Both doors opener now work fine.
 

bertus

Joined Apr 5, 2008
22,266
Hello,

Are you using one transmitter or more?
When using more transmitters on the same frequency, they will interfere and give unwanted results.

Bertus
 

bertus

Joined Apr 5, 2008
22,266
Hello,

It should be ok to use one transmitter at the time.
As said tuning might help to extend the range.
First use 1 transmitter to tune ALL receivers.
Then tune the other transmitters to the receivers.

Bertus
 

MikeML

Joined Oct 2, 2009
5,444
Transmitters are self-contained, and the momentary push switch is internally de-bounced in the encoder chip in the transmitter. Nothing needs to be added there.

At the point of the signal being too weak to be properly decoded, if you hold the transmit switch down, I get multiple outputs from the decoder in the receiver as the signal is decoded, then not. My receiver is a momentary (not latching) type, so I see the pulses come and go. I don't know what would happen with a latching receiver...

Solution to the multiple decoding is simple: improve the receiver signal-to-noise ratio by putting a better antenna on it and dont hold the transmit switch down; just tap it. I have been using these same units to open/close two garage doors and my drive-through motorized gate. If I key the transmitter when I am about 100yds away, if I hold the switch down, I see the gate start, stop and reverse. If I just tap the switch, I get the desired action. When at close range, the multiple decodes dont happen, proving to me that it is an SNR problem.

I would wait on tuning the transmitters until I put antennas on the receivers. The extra loading of the antenna on the receiver will pull it a bit.
 

Thread Starter

elec_mech

Joined Nov 12, 2008
1,500
Transmitters are self-contained, and the momentary push switch is internally de-bounced in the encoder chip in the transmitter. Nothing needs to be added there.
I've added a larger momentary push button in parallel to the tactile switch on the transmitter, but if the encoder chip is de-bounces the button, it shouldn't make a difference. I have noted the datasheet for the transmitter IC calls for a pull-down resistor to the encoder IC, but none is present on the transmitters I have.

The button I'm using does have a larger travel path than a tactile switch, so a de-bounce issue might be present. It won't hurt to try a de-bounce circuit on one just to see if the results are the same, though assembling a temporary circuit and connecting it isn't a two minute task either.

As the all seem to work at various distances, just not consistently, I don't suspect tuning is an issue, although there have been times I pressed the transmitter and the receiver in the same enclosure didn't seem to pick up the signal (this was when I was using 9V instead of 12V supplies to the transmitter, so I think I've corrected this).

For the latching units, I put them on a scope and, at least when the receiver and transmitter are close, it works exactly like a toggle flip flop. You can hold the transmit button for several seconds and the receiver only toggles once as soon as it sees the transmitted signal. Press again and the receiver toggles again. I have noticed if I attempt to press the transmit button too quickly, the receiver doesn't see the signal. Likewise for pressing the transmit button multiple times rapidly - it catches the first signal, but doesn't react to the very next 1-2 transmits. I get more reliable(?) results by pressing the transmit button for about a second or so.

I had to take a break last night from this project (as I have half a dozen others right now), but I'll give the antenna a shot tonight, and if time allows the de-bounce circuit. Does the above de-bounce circuit appear correct?

Regarding the antenna, I apologize, but I don't quite follow your description. From my (poor) memory on coax, there is an outer vinyl jacket, then metal braiding, a foam core and the copper wire center. Assuming my calculations above are correct, would I strip off ~8.9" of everything off the copper center then strip off ~8.9" of the vinyl jacket (so the braid is exposed)?
 
Top