Alternative to ERF?

Jeremy Harris

Senior Member
HC-11 433MHz transceiver (was: Alternative to ERF?)

[I've edited the title to reflect the drift in this thread in later posts to include details of the HC-11 transceiver modules]

I've just heard that the ERF has been discontinued, which is a nuisance, as I've been using them and now have no spares available.

Does anyone know of a similar size and shape module that is a drop-in replacement for the ERF, please?

Essentials are that is has 0.1" pin spacing, operates at the same default baud rate (9600) and can be used just like the ERF to establish a fairly robust two way link. I'm going to have to replace all the ERF modules, to ensure commonality, so any replacement doesn't have to operate at 868MHz, but does have to just work as a transparent wireless serial link without needing any additional code.

So far I've not found anything that looks similar, most modules seem to need some form of setting up, and an unable to change the kit I already have to do that.
 
Last edited:

Circuit

Senior Member
I didn't know that it had been discontinued; it is still listed as available here; http://www.picaxestore.com/index.php/en_gb/rfa020.html
Or are you talking about the original Ciseco (now 'Internet of Things') module? That I can no longer find.

The latest version of the ERF that I have is merely an SRF module mounted on an ERF carrier board and SRFs are still listed. Given that a maximum of six connections are required to link the SRF to the ERF carrier board, I suspect that it would be pretty easy to cut a piece of strip-board six strips wide and glue on and connect an SRF module. You would then be best with a quarter wave wire antenna but ceramic antennae are easily available. Mounting a quarter wave antenna would involve a couple of surface mount components and therefore a custom carrier might be the best answer - I think that you are one of those people who can sort out their own printed circuit boards. You don't say whether you are using a PICAXE ERF or a Ciseco ERF; the protocols are slightly different - inverted serial etc. but these aspects are easily overcome in the software.
 

Jeremy Harris

Senior Member
I emailed Wireless Things to enquire and they've stopped production, so I would guess that the Picaxe version will cease to exist soon, too,

Thanks for that tip about the SRF, I'm using the Ciseco (now "Wireless Things") ERF modules as I didn't need the mods that are included in Picaxe versions to use them as a programming interface. As you say, I may be able to make a board to fit an SRF module, but I'm not 100% sure that I could also fit the ceramic antenna. I have around a dozen of these to strip out of units and replace, even though only a single ERF has failed. Given my experience with Wireless Things just withdrawing the ERF without warning I'm extremely reluctant to give them more money for SRF modules, go through the process of changing everything to suit these and then find that in a couple of months they withdraw the SRF too.

It looks like I could switch the network to run on the very cheap HC-11 modules, at 433 MHZ, as these seem to be a pin-for-pin replacement (more or less) for the 868MHz ERF modules. The HC-11 modules are so cheap that I could buy a couple of dozen to be sure that I had enough spares to keep things going. The only slight snag with that is I already have two other systems in the house that periodically transmit data on 433 MHz, which is one reason I'd opted to use the 868MHs ERF units.

I suppose the lesson here is to not buy stuff from a small company where there is no second source if they stop supplying. Much as I feel a bit uncomfortable with some of the Chinese stuff, their habit of cloning things and making them available from many suppliers can be a boon if you want more parts in future.
 

lbenson

Senior Member
HC-11 HC11 433mHz wireless module

Jeremy,

That HC-11 looks very interesting, with an "AT" command mode. I found a datasheet here: http://www.topelectronics.com.au/image-eb/HC11/HC11.pdf

Do you have any other useful links?

Any picaxe threads (a search on "HC-11" found nothing--not even this thread--I've put "HC11" in the title to this post, so future searches will find that if anyone thinks to leave out the hyphen)?

[HT changed to HC per tex below]
 
Last edited:

Jeremy Harris

Senior Member
As well as that document you've found I found this one, with some more detail: http://www.elecrow.com/download/HC-11.pdf . The things are very cheap, around £2.50 each from China, with some ebay sellers offering them in packs of ten at a good price, complete with a small helical antenna.

As far as I can see, in default mode they work as a straight 9600 baud half duplex link, so very similar to the ERF modules, but on 433 MHz. There seem to be some useful options that can be programmed via AT commands (with the end pin held low). One option is to set one module up to control relays or similar, on receipt of commands from another module, so being able to use the module as a cheap way of operating up to three circuits remotely. There are details in the data sheet linked to above.
 

Circuit

Senior Member
Thanks for that tip about the SRF, I'm using the Ciseco (now "Wireless Things") ERF modules as I didn't need the mods that are included in Picaxe versions to use them as a programming interface. As you say, I may be able to make a board to fit an SRF module, but I'm not 100% sure that I could also fit the ceramic antenna.
Do you need the ceramic antenna? The quarter wave antenna is only 82 mm and can be curved around if you don't need long range; a quarter wave will certainly have a greater range than the ceramic ones. The ERF unit IS an SRF unit; the latest embodiments are just simply an SRF on a carrier board so the SRF is very easy to patch in as a replacement for an ERF unit.


It looks like I could switch the network to run on the very cheap HC-11 modules, at 433 MHZ, as these seem to be a pin-for-pin replacement (more or less) for the 868MHz ERF modules. The HC-11 modules are so cheap that I could buy a couple of dozen to be sure that I had enough spares to keep things going. The only slight snag with that is I already have two other systems in the house that periodically transmit data on 433 MHz, which is one reason I'd opted to use the 868MHs ERF units.
Assuming that the PICAXE units are still available, I think that these can be reprogrammed to the standard Ciseco ERF mode. So one option would be to buy a single replacement and save yourself a lot of time - and cost perhaps.
Personally, I would just bang in an SRF unit and tuck the 80mm antenna wherever I reasonably could. Also, there is so much interference these days on 433 compared with 868 that you may find that you are replacing one problem with another.

I suppose the lesson here is to not buy stuff from a small company where there is no second source if they stop supplying. Much as I feel a bit uncomfortable with some of the Chinese stuff, their habit of cloning things and making them available from many suppliers can be a boon if you want more parts in future.
Well, hold on there! I think that this was just a case of rationalisation of the range; the SRF core is now being used as a common base for the ARF unit (with additional RF amplifier for increased range); the XRF unit and some Arduino-specific toys. The SRF can be used as a direct replacement for the ERF so I don't think it can be inferred that there is any likelihood of the range disappearing anytime soon.
 

Jeremy Harris

Senior Member
These units are fitted into a small waterproof plastic box, so I only have very limited room to play with, hence the need to try and retain the ceramic antenna. They live outside and I'm reluctant to drill a hole in the case for even a very small antenna if I can help it.

If I fit the small helical vertically to an HC-11 board I think it will just fit in the box - I don't have one yet to try.

I agree, 433 MHz is crowded, which is why I'd like to stick to the quieter 868MHz band if I can.

I take the point about the SRF being used in other products from this company, but it looks as if they have just been taken over by a bigger company and seem to be undergoing some rationalisation. I'm not prepared to take the risk of the SRF module just disappearing in a few months, in fact right now I'm inclined just to never deal with this company again, as by just withdrawing the ERF without warning they have caused me to waste a lot of time seeking a replacement.

It looks as if they are designing for the Arduino form factor with a lot of their products, so I can't really see that they will continue to supply things that are outwith there core offering.

I've just found some other modules that might do the job, but need to do some more digging around to see if they can be made to work as "out of the box" replacements or not.
 

Circuit

Senior Member
FYI I just took this picture that shows the line-up of boards starting at left with the original ERF board. Next to it is the replacement based on the SRF piggy-backed onto an ERF carrier board. Next is the XRF unit, again just a carrier for the SRF module, and then the SRF module stand-alone. There is now little point in the ERF board when the SRF is so much more compact.
ERF-SRF modules.jpg
 

Jeremy Harris

Senior Member
FYI I just took this picture that shows the line-up of boards starting at left with the original ERF board. Next to it is the replacement based on the SRF piggy-backed onto an ERF carrier board. Next is the XRF unit, again just a carrier for the SRF module, and then the SRF module stand-alone. There is now little point in the ERF board when the SRF is so much more compact.
View attachment 19579
Except for the ERF having a properly mounted ceramic antenna, a power regulator and a set of 0.1" spaced pins at the end, you're right.

If I were to take the risk that the SRF is going to remain available, then I could design a replacement carrier board with the added components needed to make it work, then just fit SRF modules to them.

However, it means being able to design the antenna mount and fit the surface mount matching components, as well as the power supply components needed to work with a 5V system. It can be done, but it seems like a lot of work to do, when the chances are I can use a readily available module from another supplier that will work without needing modification.
 

Jeremy Harris

Senior Member
I've ordered some HC-11s to play with.

One useful looking function is the ability to turn a pair of these cheap modules into a simple three port remote wireless digital interface. The way this seems to work is that you programme up a pair of modules, using an FTDI or a Picaxe with a serial connection, and set the transmitting module up with a frequency channel and an address. You set the second module up with the same frequency channel and address, so they match. This is done by sending normal AT commands to the modules, with the CON pin held low. Once programmed the state is stored in non-volatile EEPROM, so it's a once-only process.

You can then programme the transmitting module to transmit either the state, or a level change (falling edge only) to the receiver module. The receiver module will then change the state of the appropriate output pin.

I can see this being useful if you wanted to either use a Picaxe to control up to three circuits remotely (say to sound an alarm, switch lights, or similar, or the other way around, where you wanted a Picaxe to respond to the switched output of a remote sensor, or a remote push button or switch.

There are lots of applications where a full blown data link isn't needed, just a reliable wireless switch, and these cheap little modules seem a neat way of achieving this, without needing to use the serin command and have to work around the fact that it's a blocking command for any Picaxe that doesn't have a good background hardware receive function.

I've immediately thought of a use for this at home, as I have a package treatment plant that has an alarm unit that senses either an air pressure drop or a effluent level too high condition and sound a local alarm (a flashing beacon and a beeper). These are at the end of the garden, and it would be more convenient to have the indication inside the house, some 30m away. It looks as if I should be able to use the switched sensors in the existing alarm box to switch lines on an HC-11 and then have another one inside the house with either LEDs directly on the outputs or the lines fed to a Picaxe so that I could have a more meaningful display,

As soon as they arrive on the slow boat from China I shall report on here how well they work.
 

Jeremy Harris

Senior Member
It seems a friend of mine lurks on this forum, read this thread and came around last night to lend me a couple of spare HC-11s, that I can replace when mine arrive from China.

So, I thought I'd start by doing some basics and learning how to programme them. First off, the better datasheet (this one: http://www.elecrow.com/download/HC-11.pdf ) isn't 100% clear, so I thought some really basic instructions may help.

Equipment used:

Standard USB FTDI module connected to PC, with these connections

FTDI HC-11

+5V +5V
GND GND
Tx Rx
Rx Tx

CON on the HC-11 to ground for AT mode, open circuit for normal mode

Terminal emulator used was Termite on Windows PC, with the following initial settings (important, as other settings caused some problems)

Baud rate 9600, 8 bits, no parity, no stop bit, no flow control, no forward

Transmitted text – nothing (so no added CR etc)

Echo on (just for convenience, could be off)

No word wrap, no function keys, no hex view, no log file, no status LEDS

Options settings seem unimportant, but I only had close on cancel selected.

To programme the HC-11 then power it on with the CON connection connected to 0V and leave it connected to 0V.

Follow the programme instructions in the data sheet when in AT mode. For example, here is a screen grab from Termite of me checking the setting of the unit (using the AT+RX command) and then setting the frequency channel to 15 (015) and the address to 150 (0 to 255 is possible)

First, a screen grab showing the bare module as received and its settings:

AT+RXU1
B9600
C001
A000
P8



The blue is the echoed command sent, the green is the data received from the HC-11
As you can see, the default settings are Baud rate = 9600, Channel = 001, Address = 000 and transmit power level = 8 (which is +10 dBm - see datasheet for other power settings).

Next, I will send commands to the HC-11, in AT mode (CON held low) to programme new settings:

AT+C015OK-C015
AT+A150OK-A150
AT+P6OK-P6

These settings have changed the channel to 15, the Address to 150 and the transmit power level to 5dBM (level P6).

To check that these settings have been saved, the HC-11 is unplugged and then plugged back in, so the setting should now be stored in EEPROM.
To check this, the HC-11 is reconnected, with the CON line held low, and the command AT+RX is used to read out the module saved setting:

AT+RXU1
B9600
C015
A150
P6


As before, the blue text is the transmitted command, the green text is the response from the HC-11. As can be seen, the previously programmed settings have been retained during power down and the module is still set to work at 9600 baud, on Channel 15, with the address 150 and a transmit power level of 5 dBm (P6).

The default settings can be recovered at any time by entering AT command mode and sending AT+RESET, which will restore the unit to the original settings. Be aware that doing this will set the baud rate back to 9600, so if that has been changed then you will need to reset your terminal emulator to the new baud rate to be able to communicate with the HC-11 post-reset.

Up to 20 frequency channels and 255 address channels can be set, giving lots of space for multiple links working in the same area

Please read this in conjunction with the attached datasheet for more information.

I still have more playing to do, but I can see these being very useful for use with a Picaxe. For example, there are three digital input/outputs when the modules are programmed with the transmitting module setting AT+FCMF and the receiving module setting AT+FCSF Up to 20 frequency channels and 255 address channels can be set, giving lots of space for multiple links working in the same area. This measn you could, for example, have a simple set of three push buttons (or a matrix of up to 8 push buttons) connected directly to the transmitting HC-11. The receiving HC-11 would then present a Picaxe with three lines (5V compatible) that would mimic the state of the remote push buttons.

The units can also be used as a stand alone full of half duplex UART link, with up to 255 addresses and 20 different frequency channels, allowing normal comms links between devices, or broadcasting (to all units on the same channel and address) if required.

I have to say, from the little bit of playing around I've done this morning I have to say these look to be very easy to use and versatile, and they are cheap if you buy them in significant quantities, I ordered 10 from China for around £30 including postage, so around £3 each, which seems very good value, and they include small helical antenna.

I'll report back more once I've established how well they work "in the wild".
 
Last edited:

techElder

Well-known member
Ibenson, is there some reason you are specifically referencing/searching for "HT-11" rather than "HC-11?

Jeremy,

That HC-11 looks very interesting, with an "AT" command mode. I found a datasheet here: http://www.topelectronics.com.au/image-eb/HC11/HC11.pdf

Do you have any other useful links?

Any picaxe threads (a search on "HT-11" found nothing--not even this thread--I've put "HT11" in the title to this post, so future searches will find that if anyone thinks to leave out the hyphen)?
 

lbenson

Senior Member
Ibenson, is there some reason you are specifically referencing/searching for "HT-11" rather than "HC-11?
Well, I'd try to claim "finger fault", but if pressed, I'd have to admit that it was mental.

"HC-11" still finds nothing; "HC11" now finds this and unrelated threads.

[Post above corrected with credit to Tex.]

Very interesting, Jeremy. Thanks for posting your results, and thanks to your lurking friend.
 

Jeremy Harris

Senior Member
More information on HC-11 transceivers

I have some more info gleaned from playing with these units that may prove useful.

First a legal warning. The 433 MHz ISM band in the UK and Europe covers 433.06 MHz to 434.79 MHz. The HC-11 module defaults to 433.4 MHz in channel 1 and each channel is 400 kHz wide. There are 127 channels that can be programmed, but the somewhat skimpy data suggests that only channels 1 to 20 are stable. This neatly hides some of the legal issues!

The power output of the HC-11 seems OK, at +10 dBM, so for short range applications there should be no real risk of interference.

Legally, you can only use channels 1 to 4 though, as channel 5 (435.0 MHz) and above are outside the licence exempt band. I'm guessing that the warning regarding not going above channel 20 has to do with the matching components on the CC1101 RF IC.

The modules use FSK, with the packet format being a pre-amble (not clear on exactly what this is, yet) followed by the unique programmable device address (in the range 0 to 255), then the data byte. You can therefore have up to 256 pairs of these modules all on the same frequency, and as long as they don't all transmit at the same time you should be able to maintain interference-free operation.

So far I've only being playing with two modules, but can confirm that they interface to a Picaxe very easily and allow semi-duplex data transmission. The technique I've tested it to transmit a data sequence from the master unit and wait to receive a "received OK" message back. If that OK message isn't received back the master transmits it again. This is just a part of my application where I want to switch something on remotely and have confirmation that it has really switched on.

I've done some experiments with the HC-11 as extended wireless push buttons, too. I sent the appropriate AT commands to two units to set one up as a state transmitted and the other as a receiver. There is a quirk in that the buttons must pull low to work and the loads must be on the high side (i.e. between the pins and +5V) but I have managed to get three push buttons lighting 3 LEDs over a few metres range with remote push button control. I have an application in mind where this will be very useful, as it gets around the blocking problem of waiting to receive a serial data signal. I can use one of the three outputs of the HC-11 to trigger an interrupt, so having near-immediate control of the target Picaxe.

I have a lot of signals floating around on 433 MHz, yet these cheap little modules seem pretty good at not allowing that to upset them. The output seems very clean, with it only responding to a specifically addressed command. I'm sure that that are still going to be subject from blocking from time to time, but for me the important thing is that they don't give any spurious outputs. Using the handshaking system I have gets around short duration blocking OK, by allowing for resending of commands that don't get through.

So, in short, these modules allow 4 legal radio channels in the 433 MHz band to be used, with 256 unique addresses. They allow half duplex data transmission and reception, or three digital state signals to be sent an received. All for a price of under £2.60 each including shipping.

Hopefully the above may be of use to someone else.
 

Jeremy Harris

Senior Member
Small query - NO stop bit?
That seems to be an error, both in the documentation and my re-iteration of it. Getting hard data on these is difficult, as it often is with generic Chinese stuff, but so far I've found the following out about programming them with the AT+U instruction. Valid AT+U commands seem to be:

AT+UN = No parity
AT+UO = Odd parity
AT+UE = Even parity

AT+UX1 = 1 stop bit
AT+UX2 = 2 stop bits
AT+UX1.5 = 1.5 stop bits

Where X is the parity designator. In practice the parity designator is only echoed back with the "OK" reply when set to anything other than no parity. For example, sending AT+UN1 seem to just reply "OK-U1". Sending AT+UO1 gets the reply "OK-UO1", indicating odd parity, 1 stop bit.

As usual, it is as confusing as hell trying to get to the bottom of the various form of Chinglish documentation. If I get time I will try and test and document all of the commands to make life a bit easier for anyone else that want to use these modules.

It's worth noting that they have a "big brother", the HC-12, with a +20dBm output, that is very similar in appearance to the HC-11, but apparently not compatible with it. Whether this is because it uses a different channel arrangement, or perhaps Manchester encoding or some other variation in data formats I don't know. I have no real need for 100mW of power, so probably won't bother looking at the HC-12.
 

Jeremy Harris

Senior Member
Thanks very much, that makes life a lot simpler for windows users, I'm sure. The threads on that forum, particularly this one: http://www.thebackshed.com/forum/forum_posts.asp?TID=8246&PN=8 are well worth reading, too.

I've had a look those utilities on my old Windows netbook and it seems that the main difference between the HC-12 and the HC-11 (apart from the greater power) is that the HC-12 can be set to operate with 7 data bits, rather than 8 and it seems to use a different RF IC, hence the incompatibility, perhaps, between the two when they are both in their default mode. The frequency range looks to be the same, with 433.4 MHz being the lowest channel and the steps being 400 kHz. There is a data sheet on that same forum for the HC-12: www.thebackshed.com/forum/uploads/robertrozee/2016-01-14_122335_HC-12_v2.3B.pdf that may be of help for those looking for greater range.

I plan on doing some more testing later today, but right not I'm pretty impressed with these cheap transceivers. They are as easy to interface to a Picaxe as the ERF modules, seem to be as reliable and interference free and for around the same price as a cheap and nasty pair of ASK modules you're getting a fully functional multichannel, addressable, FSK transceiver, with UART interface. These things are even overcoming my natural suspicion about "no name" Chinese modules!
 

Jeremy Harris

Senior Member
I can add a little more about using these modules with two Picaxe variants, the 20M2 and the 08M2. The well-known Picaxe baud rate error problem when using some baud rates and software serial ports does rear its ugly head again with these modules. It seems that there is sufficient baud rate error to stop some combinations working. This isn't something new, and has been reported here before by others, but it should be noted that if you wish to use the default 9600 baud reliably then you are advised to up the Picaxe clock speed to get reliable operation if sending more than three or four data bytes (The error seems to be cumulative and gets worse the greater the number of bytes sent or received).

I've seen this exact same problem with other devices with a hardware serial port, when interfacing with a Picaxe software serial port. It isn't a show-stopper, but just needs to be remembered if trying to send a lot of data in one transmission.

My test setup (intended to do some ad hoc range testing, but I've spent hours sorting the baud rate problem instead!) is a 20M2 fitted to an AXE091 development board. This is running a simple bit of code that transmits 6 bytes at 9600 baud, then looks to receive back from the remote end 9 bytes. The remote end is an 08M2 that receives the six transmitted bytes, then prefixes them with "OK " (three bytes, there's a space) followed by the 6 received bytes.

The initial testing used just three bytes transmitted to the remote transceiver, with the Rxd and Txd connections linked, so that the HC-11 just retransmitted whatever was sent. That seemed to work very reliably, so I thought that a more challenging test would be to add an 08M2 at the remote transceiver end to receive the bytes, add the "OK " prefix and re-transmit them, increasing the number of bytes to a realistic message length for my application. It was when doing this that I ran into problems, with bytes after the first three or four being corrupted fairly randomly.

Initially I just increased the 20M2 clock to 32 MHz (adjusting the serout and serin baud rates accordingly) and this made a very small improvement, perhaps the first four bytes would get through OK, with the rest corrupted. Increasing the 08M2 clock to 32 MHz (with the same adjustments to keep the baud rate at 9600) completely fixed the problem, and I can now send, retransmit and receive 9 bytes with no errors.

The next stage will be some range testing, but I thought it might be useful to flag up the baud rate problem, especially as it is mentioned on one of the Youtube video about these transceivers. It may well be that the HC-11 is over-sensitive to slight baud rate errors, or may itself have an error, but my money is on it being a Picaxe issue, as I know that exactly the same problem has occurred in other applications.
 

Jeremy Harris

Senior Member
I've now completed some range testing, using a "remote" transceiver that was connected to an 08M2 and an indoor "base station" transceiver using a 20M2, both running at 32 MHz with the transmit and receive baud rate at 9600, the default condition for the HC-11 as it arrives. The default channel (433.4 MHz) and address (000) were also used. Both the indoor and outdoor unit used the supplied small helical antenna arranged vertically. Software serial ports were used at both ends.

To provide a robust test, the remote end expected to receive 6 bytes (the ASCII characters "Jeremy") and was programmed to transmit an error message if the serin command timed out after 2 seconds, or if the received message was garbled. If all was OK, the transmitted message from the remote receiver was three bytes (the ASCII characters "OK ", with a space after the OK) followed by the six received bytes echoed back. If the serin on the remote receiver timed out, or if the message was garbled, the remote transceiver transmitted nine bytes ("ERROR ", padded with ASCII spaces to make 9 bytes).

At the indoor "base station" end the same HC-11 set up was used, with the small helical antenna being used, mounted so that it was vertical. The code in the base station 20M2 repeatedly transmitted the six ASCII characters "Jeremy" using serout at 9600 baud (so using the "T9600_32" instruction in serout and serin). After transmitting these six characters a serin command waited for a response from the remote transceiver for two seconds. If no response was received it timed out and sent "No signal received" to the terminal, using the sertxd instruction. If a response was received by the base station then the response was echoed as received as nine bytes, plus an added Cr and LF, to the terminal using the sertxd instruction.

The usable range, transmitting through one internal house wall and one thick external wall, was modest, about 13 metres. At 14 metres the outdoor transceiver started throwing errors, at 15 metres there was nothing being received at the indoor end. I found this range disappointing, but strongly suspect that the cause is primarily the small, low gain, helical antennas that are supplied. I have cut some lengths of 1.5mm thick copper wire to 173mm and have some pieces of old SRBP copper board that I can cut into ground planes, so I'm going to make a couple of decent 1/4 wave monopole antennas, fitted with a decent ground plane area for that frequency and try the test again on Monday to see what, if any, improvement I can get. Based on using similar 1/4 wave wire antennas with the cheap ASK modules I'm anticipating a significant increase in range.

One major advantage of these modules is that they don't seem to outwardly respond to any interference. I have a weather station that transmits on 433MHz every ten seconds (using the cheap ASK modules, with a 1/4 wave antenna) and also some remote controlled lights in the house that also use 433 MHz, yet the indoor receiver gives no spurious outputs at all. Despite being around the same price as the cheap ASK modules, these transceivers seem to be a far better bet for a simple, cheap, short range Picaxe to Picaxe link, even if you only use them as a one-way link.

If all goes well I should have time on Monday to do another range test with the better antennas, and I'll report back on how much of a difference, if any, these make.
 

lbenson

Senior Member
If your serins were timing out, and so enabling the sending of error messages, it sounds like these modules might not suffer from the problem which the cheap ASK/FSK modules have of failing to time out because the reception of spurious signals.

Is that your experience? Were you using a qualifier on serin? Were the errors because of "no message received" or "garbage received"?

Do the modules link, so there is one-to-one communication, or is many--to-one possible without disconnecting from a first sender and reconnecting to a second?

So many questions. Sorry, but these do look intriguing.
 

Jeremy Harris

Senior Member
If your serins were timing out, and so enabling the sending of error messages, it sounds like these modules might not suffer from the problem which the cheap ASK/FSK modules have of failing to time out because the reception of spurious signals.

Is that your experience? Were you using a qualifier on serin? Were the errors because of "no message received" or "garbage received"?

Do the modules link, so there is one-to-one communication, or is many--to-one possible without disconnecting from a first sender and reconnecting to a second?

So many questions. Sorry, but these do look intriguing.
You're exactly right, With no transmission from another module on the same channel the output of these modules is signal free, not a trace of a blip of anything on the output, when looking at it with a 'scope. This is unlike the cheap ASK modules that have a great deal of random noise on the output when there is no signal.

The errors I got at the margins of reception were garbled characters at the end of the message, probably because the signal to noise ratio was right at the limit and the preamble and address were getting through, together with a couple of bytes of the message, then rubbish. The drop off when I moved the distance apart by another metre was complete, it went to having no signal at the output at all.

I didn't use a qualifier on serin, just loaded 6 byte variables with whatever came in the serial port, compared it to the 6 bytes that were expected and if the two were the same I re-transmitted the received 6 bytes with the addition of "OK " in front.

The modules are addressable, as well as having frequency channels. This means you can set two up with a unique channel and a unique address and they will just talk to each other, with others in the area being on a different address and not responding. You can also implement a "one to many" link, by setting several modules to the same frequency and address, but you would either then have to accept that the broadcast was unidirectional, or have some other way of sequencing replies on this net. For example, you could have one module as a master, transmitting to, say, four other slave modules. If the message sent by the master was only intended for one of the four slaves you could include an identifier in the first bytes, so that only that slave responded. If you wanted confirmation that the slave had received the data you could use handshaking, where the correctly identified slave transmits a unique identifying handshake to the master.

An example would be if you wanted to switch on some outdoor lights, in sequence. The master transmits an identifier (say "1") followed by a command ("say "ON") and then listens for a response. The slave transmits back it's ID and state, so "1" followed by "ON" so that the master then knows that the particular slave has responded correctly. The process could be repeated for every slave on the net.

Alternatively, you could just use this broadcast unidirectionally, without bothering to check whether the slaves have all responded, by broadcasting an "ON" signal that all the slaves on that address would respond to, without any handshaking (in the same way that the serial displays work without handshaking).

If you connected the "SET" or "CON" terminal (the labels seem to vary) on the HC-11 to another Picaxe output pin on the master, then you could change the address or channel of the master HC-11 using AT commands. This would allow unique control of data from a master to individual slaves, as the master Picaxe could put its HC-11 into AT mode, change the frequency, or just the address (there are 256 addresses available, 000 to 255) and then transmit and receive unique data to one of the slaves only.

You could also daisy chain units, so that the first receives an instruction, acts on it, retransmits it so another down the line receives it and acts on it, etc, etc. Doing this could extend the range and allow data from, say, an array of sensors, to be collected and then passed back to the master.

I'm sure there are a lot more possibilities. One thing I've found is that they will echo if the Txd and Rxd connections on a slave are just linked, with nothing other than a power supply and antenna. What this means is that when a byte is transmitted by the master, the slave receives it and then retransmits it back. I can see this being useful when you just wanted confirmation from a remote unit that data had been received and didn't want the overhead of using another I/O line or bit of code to send confirmation of data having been received. In the example of turning garden lights on, you could have the remote switch receivers set with just an HC-11 and an 08M2, with Rxd and Txd on the HC-11 linked and that link going to a serial input on the Picaxe. Three spare Picaxe outputs could then switch three lights, say red, blue, and green, individually. The master transmits "RED ON" and if the signal gets through the slave turns the red light on and sends back "RED ON" as confirmation. this could be replicated for the other outputs and for "OFF" commands, and could be expanded with custom commands, like "ALL ON" or "RED & GREEN ON", allowing both control and confirmation of control by a cheap and simple link.
 

lbenson

Senior Member
Thanks, Jeremy. These devices look very useful. One more question--about battery operation.

What kind of quiescent current draw do these have? Or what maximum? For instance, could they be turned on briefly by a picaxe, either directly from a picaxe pin or through a transistor, allowed time to connect, send a brief message (e.g., change in some sensor status), perhaps await an acknowledgement, then be turned off again, with little draw on 3 AAs or 4 with a regulator?
 

Jeremy Harris

Senior Member
Thanks, Jeremy. These devices look very useful. One more question--about battery operation.

What kind of quiescent current draw do these have? Or what maximum? For instance, could they be turned on briefly by a picaxe, either directly from a picaxe pin or through a transistor, allowed time to connect, send a brief message (e.g., change in some sensor status), perhaps await an acknowledgement, then be turned off again, with little draw on 3 AAs or 4 with a regulator?
They have a programmable current mode, so you can set them up so that they draw a very low quiescent current (80µA according to the data sheet) and they "wake up" transmit or receive data, then go back to sleep again. The downside with this mode is that there is a long delay (the datasheet says 380mS) between sending or receiving a character and it actually getting through. The datasheet isn't at all clear on how this very low quiescent power mode works, it's described as mode FU2. The default mode is FU1, with a quiescent current of 3.5 mA according to the datasheet, but with the peak current being a bit higher when transmitting. I'll try and do some measurements on Monday to see what it really draws. There's also an FU3 mode, which increases the quiescent current to 22 mA but reduces the standard 20mS transmission delay time to 2mS.

According to the data sheet for the CC1101 RF chip this board uses, the current drawn during transmit is 22mA, so that would suggest that the total current in the default FU1 mode would be around 3.5mA when idle and a bit over 25 mA during transmit at full power. You can programme the transmission power down from the default 10dBm, though, so it may well be that if set to, say 7dBm, or even 5dBm, the total current during transmit might come in under the 20 mA pin limit for a Picaxe. Alternatively, if you can spare a Picaxe output to drive the control pin low and put the module into AT command mode, you can then send the command AT+SLEEP and the module will shut down to sleep mode, drawing 20µA. However, before you can use it again you need to switch it back on by going to AT mode (taking the control pin low) and then back to normal mode (control pin high).

If you can spare an output pin in addition to the data transmission output, then it would seem that you could do what you need to do by switching the module to AT mode, sending "AT+SLEEP" and that will reduce the current to 20µA until you need to transmit data. When you want to transmit, you just wake it up and it's ready to go, in normal mode with 3.5 mA quiescent current.

I haven't had time to test this out yet, so this is just going on what's in the datasheet, but it sounds as it it might do what you're after for battery operation.
 

RonnS

Member
thanks jeremy

I have achieved similar results with the HC12. It would be useful to test even lower baudrate...

for my PicaxeParts i prefer Hserin with interrupt....

Ron
 

Jeremy Harris

Senior Member
I have some HC-12 modules coming from China along with the order I have in for HC-11 modules, so it will be interesting to see how they compare. I've heard that the receiver part of the CC1101 RF IC used in the HC-11 isn't as sensitive as that used in the HC-12, so it may well be that the HC-12 is the better of the two, although to be legal it would need to be re-programmed to a lower power output.

I've just completed some more tests with the HC-11 modules, but with the small helical antenna replaced with a 1/4 wave monopole (just a 173mm long bit of stiff copper wire). At first I tried just changing the antenna on one module, leaving the other with the small helical antenna. The reliable range improved slightly, to about 16 metres, including going through a single internal wall plus an external wall.

Next I fitted 1/4 wave antennas to both units and the reliable range is in excess of 20m, again through an internal and external wall. This is about as far as I can go at the moment, as I have the remote end right at the end of our small garden.

This seems to show that one of the main range limitations with these modules is the antenna, so if used with a simple 1/4 wave monopole they should be good for at least 20m going through walls and probably upwards of 30m in the open air. For any application I have this is plenty of range, and I've no doubt that if I was to reduce the baud rate to something lower than the default 9600 baud I'm currently using then the range may well be extended.

As before, there is no interference on the data lines at all, if the link is working you get solid data, when it is out of range, or the remote unit is turned off, nothing gets through.

In effect these modules work just like a long 5V "RS232" protocol serial cable, with data seemlessly transmitted both ways in semi-duplex. They probably can't be used to programme a Picaxe, though, as I doubt they properly support the break command, it's not something I've tested yet, though.
 

srnet

Senior Member
If it were me, I would forget about building antennas on the assumption that the radio module is precisely 50ohm output.

For a simple vertical or dipole start with an antenna that is maybe 50% longer that theory suggests, then cut bits off whilst you measure the received signal at a distance, say 50M.

The idea is that you match the antenna to the module, not a theory.

A ground plane can be as simple as 4 x 1/4wave ground radials at the base of antenna, that can have a significant effect on antenna performance over just a 1/4wave vertical.

And yes, all made from cheap bits of wire.
 

manuka

Senior Member
At UHF height can be everything, so elevating the transmitter (or moving it to a signal "sweet spot") could be more effective than antenna wrangling. 433 MHz ISM regs. may -ah- prohibit extra TX antenna anyway ! Modules with better receivers could hence be more attractive.

And speaking of receivers-on top of it's 9600 bps timing issues,it transpires (as I'd recently PM mentioned to Jeremy) that the HC-11 uses a less sensitive TI CC1101 RF IC. Although the CC1101 is far more versatile & less prone to interference than RF ICs in modern ASK receivers,the HC-11's reliable range may only be similar to superhet ASK offerings using the likes of Princton Tech's PT4302.

It's tricky to compare sensitivities of these transceiver RF ICs just from data sheets,as values are variously stated as -112dBm for the HC-11's TI (Chipcon) CC1101 & -126dBm for the HC-12's SiLab's Si4463. I've not yet made a shootout, but it's apparent that the superior Si4463 RF IC beats the CC1102 by a good 6dB at similar settings,so the HC-12 should immediately give perhaps twice the range of the HC-11 for the same TX power.

Experiences here in NZ pleasingly support this,with work largely thanks to Christchurch based E.E. Robert Rozee (who has also updated the "Chinglish" data sheets). Many of Rob's HC-12 findings are posted at the Australian based Back Shed forum.

As the HC-12 module is also free of 9600bps woes,has higher TX power capabilities (adjustable to suit local regs.) & is a similar price,why would you now bother with the HC-11 ? Chinese firm satisfyelectronics has been extremely sharp on both HC-12 prices (under US$10 a pair) & delivery. Orders are globally p&p free.

Extra: A HC-12-USB dongle is also available at ~US$10-$15 each. This may suit "IoT" style PC based data logging etc. Yellow, blue & red LEDs indicate pass thru' or AT config. mode - the latter chosen by a small side push. Naturally module placement could be enhanced with cheap & lossless USB cable as well.

Stan. (Forum irregular)
 

Attachments

Last edited:

Jeremy Harris

Senior Member
Thanks for the heads up on that other ebay supplier, they are selling the HC-12s cheaper than the place I ordered from a week ago, so I've taken a risk and ordered half a dozen more from them.

I've also been playing with antennas, and srnet seems to be right, the HC-11 output almost certainly isn't 50 ohms at the default frequency of 433.4 MHz. I did find that, instead of chopping wire antennas to length, making a sleeve dipole from thicker brass tube gave a better match, probably because of it's very much wider bandwidth from using thicker elements (I used 2mm brass tube for the radiating element and 4mm brass tube for the sleeve, both sourced from our local B&Q (a big hardware store, for those outside the UK). The problem is that the sleeve dipoles are big, and not well suited for a lot of applications.

If the HC-12 has a much better receiver than I may be able to revert to a smaller loaded antenna, like the small helical antennas they come with, which would be more practical.

By coincidence, my garage door opener is getting unreliable, and that uses an 868MHz rolling code system. Sadly it is a system unique to the manufacturer, SWS, so uses a different rolling code system from the more common keeloq one, which makes getting a replacement expensive, as I'd have to buy the genuine product, at around £45, rather than one of the very much cheaper programmable ones that are available. It would be cheaper to buy a complete new receiver and two handsets than buy a replacement handset alone from the manufacturer,

As these small HC-11 and HC-12 modules are transceivers, it has occurred to me that I could make a fairly secure replacement using a pair of them with a Picaxe at each end, using handshaking between the transmitter and receiver to add another level of security, I haven't worked through the details yet, but I was thinking of having a look up table hard coded into the receiver and transmitter, or perhaps just a fairly complex algorithm, so that the transmitter sends a key, the receiver looks this key up and sends a response and then the transmitter looks up the response and sends another key, at which point the receiver operates the relay to open or close the roller door,

It's not perfect, or super high security, but if I turn the power level down to just enough to get the range needed I think it should be adequately secure from snoopers who gain access by just sniffing signals, especially if I make the look up table or algorithm change after each successful operation, such that a repeat of the same code wouldn't work.

I'm planning on trying to prototype a system like this over the weekend, using the HC 11 modules I have at the moment to experiment with, but will start a new thread when (or if!) I get something working reliably. Meanwhile I'm going off to do some research on how the rolling code systems work, so I have a better understanding as to how to make a secure remote link like this.

Thanks for all the helpful comments and suggestions, they have all been very useful.
 

techElder

Well-known member
Jeremy, for less than $5, this might still interest you. It's from SparkFun.

PS. There's also a 915MHz version for the "Americas." The 434MHz frequency is not legal for this kind of application here.

RFM69HCW Wireless Transceiver - 434MHz
COM-13910 RoHS In Eagle Library

Description: This is the 434MHz base RFM69HCW Wireless Transceiver that is found on our RFM69 Breakout. The RFM69HCW is an inexpensive and versatile radio module that operates in the unlicensed ISM (Industry, Science and Medicine) band, a set of frequencies set aside for low-power, short-range, license-free radios. It’s perfect for building inexpensive short-range wireless networks of sensors and actuators for home automation, citizen science and more.

This RFM69HCW module operates on the 434MHz frequency and is capable of transmitting at up to 100mW and up to 300kbps, but you can change both of those values to fit your application. For example, you can maximize range by increasing the transmit power and reducing the data rate, or you can reduce both for short-range sensor networks that sip battery power. At full power and with simple wire antennas, we can get messages from one side of a large office building to the other through numerous internal walls. In open air you can reach 500 meters or more. With more complex antennas and modulation schemes, similar parts have successfully transmitted from space to the ground (by very smart amateur radio enthusiasts; your mileage may vary)!

The RFM69HCW uses an SPI (Serial Peripheral Interface) to communicate with a host microcontroller, and several good Arduino libraries are available. It supports up to 256 networks of 255 nodes per network, features AES encryption to keep your data private, and transmits data packets up to 66 bytes long.

SparkFun sells two versions of the RFM69HCW: a 915MHz version and this 434MHz version. Although the ISM band is license-free, the band itself is different in different areas. Very roughly, 915MHz is for use in the Americas, and the 434MHz version is for use in Europe, Asia and Africa. Check your local regulations for other areas.

Features:

+20 dBm - 100 mW Power Output Capability
High Sensitivity: down to -120 dBm at 1.2 kbps
High Selectivity: 16-tap FIR Channel Filter
Bullet-proof front end: IIP3 = -18 dBm, IIP2 = +35 dBm,80 dB Blocking Immunity, no Image Frequency response
Low current: Rx = 16 mA, 100nA register retention
Programmable Pout: -18 to +20 dBm in 1dB steps
Constant RF performance over voltage range of module
FSK Bit rates up to 300 kb/s
Fully integrated synthesizer with a resolution of 61 Hz
FSK, GFSK, MSK, GMSK and OOK modulations
Built-in Bit Synchronizer performing Clock Recovery
Incoming Sync Word Recognition
115 dB+ Dynamic Range RSSI
Automatic RF Sense with ultra-fast AFC
Packet engine with CRC-16, AES-128, 66-byte FIFO
Built-in temperature sensor
 

Jeremy Harris

Senior Member
Thanks for taking the time to add the above. I had looked at the RFM69 modules, as their RF spec looked really good, as you've mentioned. The only reason I chose to to buy some to try was because of the rather awkward SPI interface. What I was really looking for was a module that would be a pretty good straight swap for the ERF modules that are no longer being made, so a 5V capable transceiver with a straightforward UART serial interface, so that at most I might need to re-programme the modules and add an antenna before using them as replacements for the ERF modules.

For someone looking for a very good RF module and that's prepared to work around the SPI interface, then I think the RFM69 modules look to be a very good choice.
 

neiltechspec

Senior Member
Was going to buy ERF modules until I found they had been discontinued.

On watching this thread, I have ordered 5 of these myself (HC-12, from the far east).

It's just a pity they don't do an 868Mhz version (lots of 433Mhz crap around me, including 3 of my own), still, from the testing Jeremy has done they look pretty good as a 'drop in replacement'.

My application will be battery operated outdoor wireless PIR's.

Neil.
 

srnet

Senior Member
For someone looking for a very good RF module and that's prepared to work around the SPI interface, then I think the RFM69 modules look to be a very good choice.
The only advantage I can see of using the RFM69 is that it will run at 300kbps.

The LoRa devices are 'limited' to 37.5kbps, but otherwise they perform so much better than FSK devices such as the RFM69, good in RF noisy environments too, chalk and cheese really.

Do not underestimate the difficulty of writing the required software from scratch, they may well be Arduino libraries for RFM69, but PICAXE ?

There is PICAXE code for the LoRa devices.
 

techElder

Well-known member
There is PICAXE code for the LoRa devices.
I'm sure it doesn't matter, but the Chinese LoRa devices aren't available for the USA. Legally. All I ever see is 433MHz devices aplenty.
 

lbenson

Senior Member
Here a question I posted in 2008 (# 6 in the linked thread) concerning permitted U.S. use of 433mHz. While the post following recommends use of other frequencies, I haven't yet heard an explanation of what is permitted to picaxe users in the U.S. in the 433mHz range under the noted exceptions.

http://www.picaxeforum.co.uk/showthread.php?11057-Continuous-outside-433mhz-transmissions&highlight=regulation

This document here: http://fjallfoss.fcc.gov/edocs_public/attachmatch/FCC-04-98A1.pdf along with much other information about an FCC change relating to RFID in the 433.5-434.5 mHz range, states in paragraph 25: "The 433.5-434.5 MHz RFID band we are adopting falls within the 433-435 MHz band that ARRL has designated for auxiliary and repeater links. Auxiliary stations are required by the Commission’s rules to operate on a point-to-point basis and are permitted to operate with a maximum power of 50 watts. Because point-to-point operations typically use directional antennas, there is less likelihood of interference from other sources."

This document states that unlicensed 270-460 MHz band must have at least 30 times the dead time as opposed to active, with 10 seconds minimum between transmissions: http://www.radiotronix.com/downloads/RCTcommunicationtomicrocontroller.pdf

This ARRL document states: "The current Section 15.231(e) provisions for periodic radiators, however, permit field strengths of less than 5,000 uV/m at that frequency (measured at 3 meters), with duty cycles of less than one second, assuming a silent period between transmissions of at least 30 times the duration of the transmission." http://isotc.iso.org/livelink/livelink/fetch/2000/2122/327993/327946/4495289/4495082/4495496/31N14883.pdf?nodeid=4499240&vernum=0

I had heard of the 3-second-per-hour rule applying in Australia and, I think, Europe, and had heard that a similar rule applied to unlicensed 433 mHz transmissions in the U.S. Apparently, and loosely, "similar" means a maximum of 1 second on and 30 seconds off--much more liberal than the Australian regulation. An expansion of that time specificaly for RFID in commercial and industrial areas was FCC-approved in 2004. ARRL objected to that but did not prevail. There appears to be some power restrictions to the licensed ham use at this frequency.

Disclaimer: I don't know anything definitive about this, and would be glad to hear from those who do.
I'd still be interested to know, since workable (serial) 433mHz modules are still much more available and less expensive than those explicitly available to hobbyists in the U.S.

In particular, how does "field strengths of less than 5,000 uV/m ... with duty cycles of less than one second" relate to the parts (and transmission power terms) we are familiar with, e.g., the HC-11 or HC-12 433MHz transceivers?
 
Last edited:

lbenson

Senior Member
Thanks for the link, MikeM100, but price is $47.89, so not quite what is desired: http://www.saelig.com/category/MFR00054.htm

I don't wish to hijack this thread, but I'd certainly like to use the HC-12 modules in the U.S. From what I quoted and the links, it seems clear (to me) that irregular 433mHz transmission of up to 1 second in duration, with 30 times the off-time, is legal at some transmission power. How the legal transmission power in mW relates to "field strengths of less than 5,000 uV/m ... (measured at 3 meters)" is the mystery to me.
 
Top