RF Module Communication for Hobby Project: Using TTL or Encoders/Decoders?

[aquitaine]

For a hobby project I got a couple of RF modules, a QAM rx2 433 receiver and a DSQAM tx1 transmitter. I understand they need some kind of TTL to work, would just using some NAND gates hooked up to the transmitter and receiver be enough or do I need something more fancy like encoders and decoders?

 

[davenn]

have you looked for datasheets for these TX and RX modules ?
They came up on google with a number of hits
they are likely to tell you all about interface and signal requirements Dave

 

[Merlin3189]

I understand they need some kind of TTL to work, would just using some NAND gates hooked up to the transmitter and receiver be enough or do I need something more fancy like encoders and decoders?

Looking at the meagre data sheets I'd say, what goes in comes out. You put a digital signal into the TX and get the same digital signal from the RX.

The sheet says the circuits will accept / provide CMOS/TTL levels for both input and output, though since the voltage levels are given (for the RX) as 0.3 Vcc and 0.7 Vcc, I'd have said that was CMOS levels rather than TTL. But these are max and min, so they may be TTL compatible.
Edit: so you don't actually need TTL or CMOS chips, it's just that these boards work with logic signals not analogue and these are the voltage levels which work.

The other strange feature is the minimum data rate. This suggests AC coupling so that DC levels are lost. I'd guess that is in the demodulation circuit, because with AM* they need to suppress spurious signals from changes in RF due to movement, reflections, or what have you between the TX and RX. So what goes in comes out will only be true if you observe the minimum and max data rates.

Whether you need codecs depends on what you are doing with the signal. This is just the physical layer, like a bit of wire.

* QAM seems to refer to Quasar's Amplitude Modulation rather than Quadrature AM. But that's hardly surprising at $1 a board.

 

[aquitaine]

Looking at the meagre data sheets I'd say, what goes in comes out. You put a digital signal into the TX and get the same digital signal from the RX.

The sheet says the circuits will accept / provide CMOS/TTL levels for both input and output, though since the voltage levels are given (for the RX) as 0.3 Vcc and 0.7 Vcc, I'd have said that was CMOS levels rather than TTL. But these are max and min, so they may be TTL compatible.
Edit: so you don't actually need TTL or CMOS chips, it's just that these boards work with logic signals not analogue and these are the voltage levels which work.

The other strange feature is the minimum data rate. This suggests AC coupling so that DC levels are lost. I'd guess that is in the demodulation circuit, because with AM* they need to suppress spurious signals from changes in RF due to movement, reflections, or what have you between the TX and RX. So what goes in comes out will only be true if you observe the minimum and max data rates.

Whether you need codecs depends on what you are doing with the signal. This is just the physical layer, like a bit of wire.

* QAM seems to refer to Quasar's Amplitude Modulation rather than Quadrature AM. But that's hardly surprising at $1 a board.

Thanks for your feedback. What I was attempting to do was replace a nand latch circuitry to remotely control an H Bridge circuit for a small electric motor to get it to turn on and off remotely albiet in just one direction.

So I took the switch, wired it to the inputs of one of the gates, whose output is connected to the transmitter's data pin. The data pins of the reciever are tied together and connected to one of the h bridge's data pins. The other h bridge control pin is also connected to a nand gate but is hardwired to provide an constant logical 0. After plugging in power the end is result currently is for some reason the motor is on constantly regardless. I'm not totally sure if I wired the h bridge wrong or if my concept is flawed. I'll look into it again when I have some time.

 

[Merlin3189]

What I was attempting to do was replace a nand latch circuitry to remotely control an H Bridge circuit for a small electric motor to get it to turn on and off remotely albiet in just one direction.

So I took the switch, wired it to the inputs of one of the gates, whose output is connected to the transmitter's data pin. The data pins of the reciever are tied together and connected to one of the h bridge's data pins. The other h bridge control pin is also connected to a nand gate but is hardwired to provide an constant logical 0. After plugging in power the end is result currently is for some reason the motor is on constantly regardless. I'm not totally sure if I wired the h bridge wrong or if my concept is flawed. I'll look into it again when I have some time.

So you had originally, a momentary push switch, to a latch, to the motor inputs?
And you now connect the two radio modules between the latch and the motor?
If so, I think your problem could be the static logic level output of the latch. As I mentioned in #3, the radio link is AC coupled, so cannot pass a static logic level. It must change at least 200 times per second. Otherwise the output logic level will revert to its quiescent level (probably over a few 10's msec.)
You could send isolated pulses of opposite polarity to the quiescent state. But I'm not sure how you'd handle these at the motor end.

I think your simplest solution might be to use your latch output to gate a 200 Hz signal to the TX. That would give you a 200 Hz pulse train at the RX to drive the motor. Varying the mark to space ratio could give speed control, with a minimum and maximum of about 1:10 and 10:1 (to stay below the 3kHz max data rate.)
Or send a train of pulses of say 200 usec duration at any rate up to about 3 kHz.

 

[rbelli1]

If you only need one bit you can attach the output to the clock of a flip flop. You will want the flip flop to latch on the asserted to non-asserted edge. You may need an inverter. Then also through an RC circuit connect it to data so the data line is delayed. If you have a short pulse the data line doesn't have enough time to go to the asserted state and you latch in the non-asserted state. If the pulse is long your data line goes to the asserted level and you latch the asserted level. You need to be sure you get your timing right as you can get outside of the datasheet setup and hold times really easily.

Or put one of these (or any micro-controller) at either end:
https://www.adafruit.com/products/1501

and you get all the control you need and more. Get a pair of the alternate channel radios and you can monitor too.

BoB

 

[aquitaine]

So you had originally, a momentary push switch, to a latch, to the motor inputs?
And you now connect the two radio modules between the latch and the motor?
If so, I think your problem could be the static logic level output of the latch. As I mentioned in #3, the radio link is AC coupled, so cannot pass a static logic level. It must change at least 200 times per second. Otherwise the output logic level will revert to its quiescent level (probably over a few 10's msec.)
You could send isolated pulses of opposite polarity to the quiescent state. But I'm not sure how you'd handle these at the motor end.

I think your simplest solution might be to use your latch output to gate a 200 Hz signal to the TX. That would give you a 200 Hz pulse train at the RX to drive the motor. Varying the mark to space ratio could give speed control, with a minimum and maximum of about 1:10 and 10:1 (to stay below the 3kHz max data rate.)
Or send a train of pulses of say 200 usec duration at any rate up to about 3 kHz.

No, it's not a pushbutton switch, it's a rocker switch. Originally what I had was an SR latch wired directly to power to just make it spin. I'm still ok with it spinning in one direction but I'd like to use the switch to control when it's on and off wirelessly. I was trying to replace the latch entirely, but evidently that's not such a good idea. What I did was use one gate, with the inputs tied together, running through the switch to power. The output of this gate going to the data pin of the transmitter. Having no experience with wireless at all I never took the signal type into account.

Theoretically it would be simpler to use microcontrollers but as little as I know about hardware I know almost nothing about programming. I did try to use the codec circuits in the datasheets but the only IC's available were surface mount packages and without access to a soldering iron I didn't see anyway to convert them to be used for breadboarding. So I'm trying to go with a pure hardware approach.

If you only need one bit you can attach the output to the clock of a flip flop. You will want the flip flop to latch on the asserted to non-asserted edge. You may need an inverter. Then also through an RC circuit connect it to data so the data line is delayed. If you have a short pulse the data line doesn't have enough time to go to the asserted state and you latch in the non-asserted state. If the pulse is long your data line goes to the asserted level and you latch the asserted level. You need to be sure you get your timing right as you can get outside of the datasheet setup and hold times really easily.

Or put one of these (or any micro-controller) at either end:
https://www.adafruit.com/products/1501

and you get all the control you need and more. Get a pair of the alternate channel radios and you can monitor too.

This is an interesting idea, so the RC circuit would be connected to the data pin of the receiver? I'm sorry, I'm not really experienced with electronics so I'm having a hard time wrapping my head around how this would work.

 

[Merlin3189]

How about getting something like this: http://www.ebay.co.uk/itm/DC-12V-4CH-Channel-200M-Wireless-RF-Remote-Control-Switch-Transmitter-Receiver-/141916545969?hash=item210ae2a3b1:g:sAkAAOSwQYZW1Dj3
Only 12V output, but you could use it to drive a relay for whatever you need.
I looked around for some encoder & decoder chips, but by the time you've put them with your radio modules and boxed them up, you might as well get the ready made units. Maybe your receiver module is a bit better than this one, but you can look around for a similar device with better spec if you need it. They all seem to use the same codec chips with 1- 4 channels, so if you get one with at least two, you can have on/off and fwd/rev. At the sort of prices these sell for, you could afford to experiment with them.