Best 433Mhz transmitter/receiver pairs?

hugoxp1

Member
Hi,


Because and don't have success put to work my sparkfun transmitter and receiver i will try to use another type of transmitters and receivers.

I have found these three:

http://www.robotshop.ca/on-shine-high-sensitivity-tx-rx-4.html

http://www.robotshop.ca/seeedstudio-433mhz-low-cost-transmitter-receiver-pair-3.html

http://www.robotshop.ca/on-shine-low-cost-tx-rx-4.html


Can anyone tell me, please, what is the best choice?

And what is the best: ASK or OOK modulation?


(This is to use with 18x and 20x2 Picaxes)

If you have another suggestion, I will apreciate.

Thank you
 

manuka

Senior Member
Simply getting other modules may not be the answer, as your setup & approach may be the real culprit. TAME THAT FIRST - SIMPLY - with just a 08M!

It's horses for courses with modules. "Down under" here in Australia/NZ the Jaycar Keymark/SpiritOn modules are the most reliable entry level ones. However for demanding work go for GFSK, as this is superior to FM, which is usually preferred to OOK/ASK (same thing). Some modules (HopeRF types perhaps) however will need programming first.

Hence START SIMPLY! Almost any cheap & cheerful module will work- I have NEVER failed to establish wireless links with any offerings when powered by 3 xAA cells & controlled via a PICAXE-08M.

In general go for sensitive & narrow band receivers & S L O W data rates - AND START SIMPLY USING JUST AA CELLS & BREADBOARDS. I have numerous 433 MHz examples at => www.picaxe.orcon.net.nz. START SIMPLY !!
 
Last edited:

russbow

Senior Member
Your sparkfun TX / RX pair are good and well matched. They should do all you want for a project.

Download the relevant documentation / data sheets from the sparkfun site and read them.

Come back if you have specific questions about the modules or your circuit. There is a lot of 433mhz expertise on this forum that will help you
 

lbenson

Senior Member
I can second the confirmation that the sparkfun TX / RX pair are good and well matched.

Have you first confirmed that your two picaxes work with just a wired connection? (Make sure that you also connect 0 volts on the two systems.)

Are you using a qualifier to filter out the noise, of which there may be a constant stream?

Have you used womai's "433 MHz wireless link trick": http://www.picaxeforum.co.uk/showthread.php?t=11531&highlight=433

If yes to all of these, post your code for both picaxes, simplified, and good photos of both setups.
 

Dippy

Moderator
I completely agree with Stan and many of the dozens of threads giving advice on how to make any RF modules work.
There are many techniques on how to write 'good code' including premables, delays and qualifiers.
There's a nice FAQs section on the Radiometrix site which can give you good info. on general principles.

You need to read up , be patient and also adopt good construction and code techniques.
Unless the device is duff ,or you have local interference, you should be able to get some kind of link working.

Only yesterday did I manage get a reliable RF link between two devices using a TI RF chip.
It was totally new to me and took nearly 2 weeks of reading up, experimenting, measuring to get the 'meat' of the code up and running.
Time consuming and tedious, yes , but I got the result - with a little presistence and some 'educational' cockups along the way.


There are many here who have great experience with dumb, smartish and smart modules. It's a matter of searching and asking clear questions and for YOU to provide as much detail as possible. If you help others to help you then you will have success, so don't just thrash around picking modules at random.
(Personally, I would choose big brand names so I could get assistance from the Manuf's but that's just a perosnal choice.)

Keep at it , you'll get there ;) .
 
Last edited:

hugoxp1

Member
Thank you all for your help... and what a great support do you have in this forum :)

My "frustration" is: I have made a lot of experiments and I still can not receive a single byte (using preamble, qualifiers, pauses, ascii, etc...) that's why I thought to buy another pair of Tx/Rx.

Well... I'm going to start over and make the things very simple using the same sparkfun pair... i will keep you updated :)


To give an answer to lbenson questions:

Please, read my previous post:

http://www.picaxeforum.co.uk/showthread.php?t=15194

Thank you
 

Grogster

Senior Member
I agree with manuka and others - start simple - an 08M is ideal, as manuka has hinted...

I've been playing around with Radiometrix modules lately, NTX2/NRX2 pair.
They have given me excellent results and shock! horror! - they just work. :p

Depending on your specific modules, there is no reason why you can't get beeps from a polling arrangement - two 08M chips(one for TX control, one for RX filter) like so:

Code:
'TRANSMITTER CODE

start:
  high 1  'turn on LED
  high 2  'make sure pin in correct state for TT comms
  serout 2,T2400,("UUUUUUUUUU!@#")
  serout 2,T2400,("Hello.")
  pause 150  'delay for LED to be seen
  low 1  'turn off LED
  wait 5  'Tra la la la la...
  goto start
Code:
'RECEIVER FILTER CODE

start:
  serin 2,T2400,("!@#Hello.")  'look for qualifier and message
  pulsout 4,20000  'beep piezo buzzer
  serout 0,N2400,("Hello.",CR)  'send message to terminal
  goto start
Provided you put a piezo with built-in driver on pin4 of the receiving PICAXE, it should beep every time a message is received, and send the message to the PC screen(open Terminal with [F8] first)

Can you get something as simple as this to work?
 
Last edited:

D n T

Senior Member
Slow down and have a break

Try using a slower baud rate, I used T600 and it never fails (almost)
I have sent "UUUUU" the paused 10 then sent my preamble "ABC, followed by 1 variable( usually 12 for some unknown reason).
Once you get ABC then 12 coming up on you debug, then add another.
Look at the green house project and reverse engineer the code until you "get it", then add more data.
Once again, start with a wired connection with a preamble then remove the wires and add the modules.
One note, often it might help to turn off then on again before you try to recieve your packets, I bought extramodules because I thought mine were RS ("really stuffed") but turning them off and on again fixed that.

If you start slow and build up, you won't be wasting your time trying to crack it at an advanced level first (and we won't tell anyone you are having trouble, we will help though).
Start slow and build up. Your code needs to be simple first, the modules are most likely matched, I have used the same types from different manufacturers amd even I (hitek redneck)can get them going.
I you need simple code and circuit look through the forum.

We await your efforts, remember start simple.
Edison never failed when he tried to invent a filament for the globe, he worked out ways not to do it.
 

SAborn

Senior Member
In support of hugoxp1 i have bought 2 sets of 433 modules that would not work.

The transmitters would work with any of the previous receivers i had but the new receivers would not detect the transmitters.

On the Cro i could see a output signal from the new receivers during a transmittion period but the output wave was in a compleetly different code then any of the earlier receivers.

I spent a full weekend trying to get these new pairs to talk to one another with no success.

Monday i returned them to the place of purchase and went else where and bought 2 other sets and these worked first up.

There is wrongly matched sets of these modules in the system and you might have received a set like i did. ( I got 2 bad sets )

It would be smart to get another set from else where, and if they work then you can retest the first set.

Dont give up because these thing are super easy to use if they are working correctly,
and think of all the information you have learnt because of a fault.
 

hippy

Ex-Staff (retired)
Start really simple and note the comments about having the transmitter too close to the receiver. For transmission over a metre or so quality of modules probably does not matter, if any work they probably all will. Ideally start with two PICAXE of the same type.

Transmit:
Do
SerOut pin, T2400, ( "UUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU" )
Pause 10
SerOut pin, T2400, ( "ABCDEF" )
Pause 1000
Loop

Receive:
Do
SerIn pin, T2400, b0
If b0 > $20 And b0 < $7F Then : SerTxd( b0 ) : End If
Loop

See if you can get that to work and amongst the noise you should see "ABCDEF" or at least part of it. Adjust polarities ( four combinations ) and try different baud rates.

Then move on and see if you can get consistent "C" ...

Receive:
Do
SerIn pin, T2400, ( "AB" ), b0
If b0 > $20 And b0 < $7F Then : SerTxd( b0 ) : End If
Loop
 

Grogster

Senior Member
@ hippy - I note that you and others in the know tend to use 10ms pauses between the preamble and the qualifier. I have never done this, and the messages always work, so I am curious as to why that is used...

My opinion is that the series of "U"'s sync up the receiver's data-slicer nicely, but then if you put a pause in between the preamble and the qualifier, could not some background "Unwanted" characters creep into the data stream during the 10ms waiting period, as the data-slicer would start to calm down again with the 10ms pause?

Perhaps it's just that in my case, the Radiometrix modules don't need the pause...
 

hippy

Ex-Staff (retired)
The "pause 10" is to cater for the case where SERIN may synchronise up midway through one "U" and never properly regain sync to the data sent and if that happens nothing will ever match the qualifiers. Some people have seen that happen. The pause can be reduced to whatever the time it takes for one byte to be sent at the chosen baud rate, 10ms is just an easy option, covers most baud rates and hopefully not too long that the slicer loses conditioning.

It's likely possible to send multiple preamble "U" and put pauses between them so SERIN does start to synch on an actual "U".
 

Zizka

New Member
Thanks for the mention, Dippy ! (I might as well declare my interest: I'm the author of those articles, as well as being the chief RF engineer at Radiometrix).

"Best" 433MHz link is a very subjective term. Longest range, cheapest, smallest and easiest to use are all very different parameters.

Narrowband units (FSK) will out-range widebands (often ASK) but have a lower maximum datarate and higher pricetag

Raw interface radios give you most flexibility (and coding 'fun') but require a lot more effort in coding, timing and formatting the datastream.

Radio 'modems' take the serial stream and do the coding and formatting for you (they often just 'look like wire') but often have advanced features which require a lot of setting up, and can cost a bit more.

Now the advert (sorry: my boss would never forgive me if I didn't get a 'plug' in):

Radiometrix has a few relevant products:

TDL2A (and the 25mW TDH2 for Australia) is a 9k6 serial port on a module. Just send it data, and 14mS later it comes out of the other end.
RXL2/TXL2 are the one-way version of the same (TDL2A is two way, a transceiver)

http://radiometrix.com/content/tdl2a

The M48A is a modem chip. Put it between the radio and the Picaxe and it does all the coding and formatting for you. It also has addressing facilities and some advanced network protocol functions (like aknowledge/re-transmit).
It's compatible with pretty much any narrow-band radio.

http://radiometrix.com/m48a-0

I'd better get off my sales counter now ...... :)

ps. Don't forget: in Australia you've got a wonderful 100mW 151MHz VHF allocation too. Amazing range potential. (I've got radios for that band too ...)
 

manuka

Senior Member
Zizka: Thanks for the insights & input. As it happens the latest (May 2010) "Silicon Chip" has a rear cover advert (& prices) for the products you mention. 151 MHz (usually quoted 151.3 MHz) is indeed a tempting frequency down under, but the Australian ACMA regulatory powers strictly assign it for animal tracking ! All manner of other spectrum slots are available - the following jolly account from some years back (when ACMA were called ACA) outlines a few -see the ACMA site for a more recent list.
The 70 centimetre band is not the only place where small transmitters such as so-called LIPD'S or "low interference potential devices" can be found. An ACA brochure called "Spectrum Opportunities for Short-Range Radio-communications" lists many more.

You'd be amazed at the sorts of applications to which low powered transmitters have been put.

Possibly the most common examples are, remote controlled car alarm systems and garage door openers. These operate near 304, 345 and 433 MHz. Personal and duress alarms, with medical and safety applications operate around similar frequencies as well as above 900 MHz.

Fancy some radio foxhunting? Well, animal tracking devices use 151 and 173.5 MHz. And if you're into tracking whales and/or sharks, you need to go down to the 48 to 49 MHz band for this.

What about locating lost joggers? Lost? We can only assume that they are - why else would anyone go round and round the same block a dozen times if they're not lost? No matter, 77.4 MHz is the place to be for athlete location systems.

What is termed "Auditory assistance" is found near 3 MHz, around 42 MHz, the FM broadcast band, and around 2.4 GHz. These are generally the systems where people put on wireless headphones for spoken information at museum exhibits and the like. The 2.4 GHz segment is also shared with other applications, such as barcode readers. (2010 update -2.4GHz is now heavily used by ZigBee, Bluetooth, WiFi, AV senders, toys, Rc controls & cordless phones etc.

Listening to cordless phones is illegal, yet baby-monitors are not. You can find these around 35 MHz, although we have also heard them near 27 MHz.

Cordless phones, by the way, use a variety of frequencies, such as 1.7/40, 30/39, 915, 1900 and 2400 MHz. Cheap two way radios can be found near 27 MHz, with 55, 152 and 433 MHz also in use.The 40 channel UHF CB ("PRS") band around 476 MHz is however the most popular.

Radio controlled models use a variety of frequencies, such as 27, 29, 36, 41,433 MHz and 2.4GHz. And somewhat bigger models, trains to be exact, use frequencies near 900 MHz for tracking purposes.

Perhaps someone is bugging you? Well, listen for yourself by tuning around 39.5, 88 to 108, 174 to 230, 520 to 820 and 915 to 928 MHz.You'll hear a feedback howl in the receiver if your room is being monitored.

We already know about those pesky crane controllers near 433 MHz and an alternative frequency for them is near 472 MHz. Keeping tabs on crims is easy if you tune to around 314.2 MHz, that's
the "home detention" monitoring device frequency.

Have you ever thought about communicating with your bodily implants? No, well, apparently 262 kHz is the place to try for 'implantable medical devices'.

Biomedical telemetry however also uses various VHF and UHF frequencies including the 70cm LIPD band. We just hope no one is using a 70cm. hand-held in the vicinity!

Moving up, we see the Gigahertz segments used for applications such as distance and speed measurement, cruise control systems, radar fluid measurements and handheld data terminals.

But we've barely scratched the surface. We could go into e-tags, product security tags, computer networks, underground communications, wireless weather stations, video surveillance and more.

It's amazing what has a transmitter or receiver attached to them these days. (ACA website)
(APCNEWS)
Although higher powers of course tempt,I tend to favour the restriction free 25mW 433 MHz for data links to ~1km. This arises from the (usually!) low UHF band noise,sensitive receivers, abundant gear, significant obstacle penetration (a major reason why crowded Asian cities use 433MHz for utility reading in preference to 2.4GHz),"decent" data speeds and compact antenna. This slot in fact covers 433.05 - 434.79 MHz so there's even an appealing ~1.7MHz wide spectrum slice. Expect to see more frequency hoping & spread spectrum devices arising to utilise this!

But horses for courses- our tolerant regulations & wide open spaces down here in the South Seas naturally differ from what may be the norm elsewhere. Stan.
 
Last edited:

Dippy

Moderator
Wise words from Zizka and an interesting list from Stan.

I fully appreciate budget and availability problems for many people.
But , that and hobby interests aside, I often wonder why some some spend weeks trying to get cheapo low-quality modules to work for important projects.

But as I'm a complete hypocrite, I've just spent 2 weeks (and counting) on a project using a low cost TI module.
I've even included routines to do an automatic site survey, ACK, retry , remote control , Echo and some AT.
The result is a transceiver which outperforms and is easier to use than XBee (in most typically used respects).
I hope Stan will be proud of me.... maybe ;)
 

Tyro

Member
The Sparkfun 433MHz transmitter pair work well. Some hardware things that are not documented clearly are that the transmitter has to have an antenna (6.8 inches, 17cm). If it doesn’t, the leading edge of any pulse has some timing jitter (30us to 0.5ms) relative to the trailing edge. This gets worse as the supply voltage increases.

The supply current as documented does not indicate that it is the average current with a 50% mark to space ratio. When it states the current is 50mA, what is actually happening is during the positive pulse 100mA is drawn and zero current during the low period. The power supply needs to be designed with this in mind.

The receiver output impedance is high and is incapable of driving much capacitance. Unloaded it has 5 to 10 us rise and fall times. I have used a schmitt trigger input to minimize potential spurious signals being generated.

One last thing. If there is a lot of local RF interference, it may pay off to have no antenna on the receiver to reduce its sensitivity. Only the, hopefully, larger signal from your transmitter will be received.
 

Dippy

Moderator
Sounds pretty poor spec then. Cheapo?
Looks like you should spend an extra quid or two and get something decent.
I'll stick to the proper makes.
 

Grogster

Senior Member
My personal favourite at the moment, is the Radiometrix NTX2/NRX2 pair. http://www.radiometrix.co.uk/products/ntx2nrx2.htm

Quite reasonable price tags for 25mW of juice. The receiver sensitivity is around 116dB depending on data-rate, coupled to an external aerial, the range performance is fantastic. Far better then the really cheap Jaycar modules, and also far better then the HopeRF units.
Go with professional pre-assembled aerials unless you know what you are doing - it saves SO much hassle.

All this is only my 2c, naturally...

The cheap modules can be perfectly usable for basic stuff, but for anything serious, you should consider the more expensive(read: reliable) range of Modules. Radiometrix have been building modules for years, so they have the tools and the talent to help you succeed, unlike some of the cheaper options.
 
Last edited:

SAborn

Senior Member
Grogster,

They look to be good modules and like you said some good data on the modules.

You quote they are a bit dearer than jaycar, What price did you pay for these modules, as an idea of their cost.
 

manuka

Senior Member
"Silicon Chip"'s May 2010 rear cover RFMA advert lists the NTX2 at Aust$32.45 & NRX2 at Aust$48.40. A data link pair hence will run to Aust$160 ! While certainly great performers (& the RSSI tap would be particularly handy),these Radiometrix modules are "only" FSK & fixed to just a single freq. They do not come with antenna sockets or antenna, & fitting these (especially the sockets) may end up costing $$ extra in a professional project.

I really don't want to keep going on about the flood of hi spec. Asian offerings,but consider perhaps the likes of the recently mentioned Sure RMB-CM12111 transceivers (rebadged Appcon APC-220). These retail at just Aust$45 a pair (c/w socket & antenna),are superior GFSK types,offer inbuilt error correction & buffering, & can be externally reconfig. for different freqs./data rates/power levels etc. A nifty USB adapter is even available.

Perhaps the closest Radiometrix offering is their RDL2, but these are WBFM,FSK,10mW TX & demand a 5V supply. Allow ~Aust$100 EACH I'd say - my past dealings with Radiometrix have shown them frustratingly cadgy about revealing prices. I almost suspect some of their products are Asian sourced items embedded in shrouding epoxy under the casing, & they're wary that someone may do some digging & reveal the truth ...

But- wireless shootouts aside -as mentioned surely a zillion times on this forum,it's horses for courses (& budgets) guys! Cheap & cheerful $($) offerings are perfectly OK in many 433 MHz applications, especially for those learning the wireless data ropes. It's possibly akin to car purchase- some buyers may be well heeled enough to consider the asking price as incidental to brand or perceived "performance".

Stan.
 

Attachments

Last edited:

Dippy

Moderator
Yes, proper and reputable companies will tend to be more expensive. But one of things in that extra cost will be R&D and Regulations testing. Another thing will be design support and warranty.


Aspects that (sadly) rarely get covered in this Forum include; quality, specification, support and warranty.
Price seems king here and sod the quality. Good'nuff is fine and a pox on my neighbours.

Companies like Radiometrix (and many others) probably have a 100 man-years more experience in design and testing than many of these recent tuppeny-ha'penny far-eastern firms.

So, dear readers:-
  • When you buy your $3 Anonymous specials can you tell me the operational specs?
  • Do they genuinely conform to Radio and Emmissions Regs (before you integrate them) ?
  • Will they give you technical support like Radiometrix and other well-known companies?
  • (...and if so, why do you keep coming to this Forum to make it work?)
  • Do people realise their legal responsibilities if they use RF in the 'real world'?
  • Are their range claims anywhere near real life? (or is that 1000m more like 100m?)
I suspect the answer to all/most of the above will be NO.

I don't know where Radiometrix stuff is made. I expect 'from all over' covers it. But, one thing is for sure, they will NOT sell stuff that doesn't conform to International/Local Regs - does that apply to a £2 Jaycar special?

I have had a huge amount of technical help from TI during my dev. Free Gerbers and BOMs and component comparisons. They have been superb. And they'd have had top marks if they'd written the firmware too.
I wonder how much help I'd get from a garage just outside .....?


Is it really akin to buying a car? :confused:
Maybe it's more akin to buying a car from an unknown manufacturer, without reading a brochure first, having a 2 day warranty and not being sure if it's actually road-legal.
Yes, that's closer to the truth I'm sure ;)

Just remember, gentle readers, your RF link may be interference to others.
 

manuka

Senior Member
Ahem-such party lines are wasted in a mere forum! You should have stood in the recent UK elections.
Maybe you did...
 

Dippy

Moderator
Honest Guv, I was just trying to be realistic.

I only mentioned Radiometrix as you had. I was not intending bias towards any one company, hence: "like Radiometrix and other well-known companies".
There are many quality companies from all over the place.
Quality costs.

I fully understand about budgets (and availability). But RF can affect neighbours. I admit a tatty bit of FR4 with inductors wound by my cat are certainly cheap.:)


UK elections? What an 'Ed Balls' up.
No I didn't stand. The Monster Raving Loony Party have given in...

Anyway, this is another one that will go round and round so I'll go and put the kettle on instead. Coffee anyone? No, it's not the cheap stuff. :rolleyes:
 

moxhamj

New Member
There are many questions to answer re the "Best" units, including;

* Tx power (general consensus seems to be a max of 10mW is legal?)
* Transmitter, receiver or 'transceiver' in one module.
* Receiver db sensitivity (correllates reasonably well with range)
* Antenna included or not included
* Real tested range for data (as against tested range for hearing something on a scanner)
* Inbuilt microprocessor (RF 'modem') or raw RF
* Data buffer size (important for packet and file transmission)
* Price including shipping to your domicile.

I tend to fix a few of those parameters (eg the Tx power at 10mW) and then work out the best unit in terms of dollars per metre for a byte transmitted then received back again. Different units are best for different distances - eg 10 metres vs 100 metres vs 1km.
 

manuka

Senior Member
I'll add -

* initial "out of the box" operation
* ability to shift freq. if heavy local interference
* ability to change on air data rates (slower gives greater range)
* user friendly config. software
* flexible supply (at least 3-~5V)
* preferably narrow band GFSK
* low current hibernation
* standard antenna socket
* TTL or TTL/RS232
* compactness

We're allowed 25 mW TX power on this 433 MHz band in NZ/Aust. Stan.
 

Dippy

Moderator
And I'll add:
As options...
  • Fine tuning of frequency.
  • User adjustable bandwidths.
  • Ability to give a success/failure error/warning message to Host.
  • Optional retry number.
  • Acknowledge - inlcuding Data Quality indication
  • Ability to preset packet header or allow host to provide packet header
  • Allow Host to preset Source/Destination or change it on-the-fly.
  • Automatic recalibration due to ageing of crystal or crystal tolerance.
  • Echo setting for testing.
  • CRC on/off
  • Remote control of 'other end'.
  • Automatic site survey
  • RSSI interrogation
  • Data Quality interrogation
  • Ability to adjust settings from 'Terminal' with responses in English.
Blimey, I can do all those on my design :rolleyes:
 

Grogster

Senior Member
"Silicon Chip"'s May 2010 rear cover RFMA advert lists the NTX2 at Aust$32.45 & NRX2 at Aust$48.40. A data link pair hence will run to Aust$160 ! While certainly great performers (& the RSSI tap would be particularly handy),these Radiometrix modules are "only" FSK & fixed to just a single freq. They do not come with antenna sockets or antenna, & fitting these (especially the sockets) may end up costing $$ extra in a professional project.
Your calculations are based on the fact that you would be using one TX and one RX as a bi-directional link, which I don't want. My use(again: MY use - perhaps not suitable for others!) is point-to-point - one master receiver, lots of transmitters. Therefore, 1x $48 cost for the receiver(and it's a hell of a receiver!), and $30 per transmitter node. For me, the fact that they don't have SMA sockets on them is perfect, as I can then design-in exactly where I want the socket to be, which you can't necessarily do with the HopeRF type units, as they come with the socket, and if space is a problem, you just can't make them fit.

I did try out the hope-RF units, and while they are attractive from the point of view of their built-in "Intelligence" and their cheap cost, they simply did not have the range in my case, and when you only need to transfer about 6-bytes of information, I have found preambles and qualifiers on the NTX2/NRX2 to be 100% success rate during my testing on the bench AND on site before the system went in.

Cheapest is not necessarily best...
 

Grogster

Senior Member
Yes, proper and reputable companies will tend to be more expensive. But one of things in that extra cost will be R&D and Regulations testing. Another thing will be design support and warranty.


Aspects that (sadly) rarely get covered in this Forum include; quality, specification, support and warranty.
Price seems king here and sod the quality. Good'nuff is fine and a pox on my neighbours.

Companies like Radiometrix (and many others) probably have a 100 man-years more experience in design and testing than many of these recent tuppeny-ha'penny far-eastern firms.
I did want to say that myself but didn't, as I thought I would be criticized for it, but that is pretty much my feeling, so thanks Dippy!. Radiometrix have been around for ages, not that I am trying to plug them here, but can the same be said about HopeRF or any of the other cheaper options for that matter in a few years?(rhetorical)

I guess the same could be said about them - no-one can see in to the future, so perhaps something of a moot point...

During my bench tests, I was able to get about 150-200 with the HopeRF units through-walls, but the Radiometrix units gave me more then 1km through-buildings with not a single byte dropped(I was logging all this back at the receiver). I guess a good part of that can be attributed to the 25mW power of the transmitter module, however, I was really surprised at the fact that the Radiometrix units "Just worked" - they did exactly what the spec sheet said, and the range was exactly(actually more) then listed - most cheap units over-estimate their range, and I am so used to that being the case, I was genuinely surprised as to the performance of the Radiometrix units.

Now, don't get me wrong - the cheap $5 modules are just fine for classroom work, me thinks, but as something of a personal choice, I would not use them in final designs simply due to their cheapness. "You get what you pay for" comes to mind. Now I know that you can pervert that statement easily into saying you are getting ripped off, but I think there is a line that can be drawn between them without getting too carried away with expense...
 
Last edited:

MFB

Senior Member
That’s was pretty much my thinking when I posted the following comment some month back.

“I can’t help coming to the conclusion that getting a basic rf link to work by trying to balance bits is like training a dog to walk on it’s hind legs. Although the result is not very elegant, it is amazing the dog can manage it at all.

Unless we want this topic to keep appearing on the forum, it may be time to come up with a more reliable alternative (e.g. code for Manchester or FSK encoding).”
 

Dippy

Moderator
Oh don't worry - it'll keep on appearing :)

The trouble is, as Stan says, Horses for Courses.
There are so many diferent types of RF module that one size won't fit all.
Obv there are certain techniques in common for dumb modules and these have been covered so many times that my bits are unbalanced.
Talking of which, the principle of Manchester aren't difficult.

I could PM you a drop of non-PICAXE code which you could easily translate.
As long as you don't imagine it'll convert a rubbish dumbo module into a magic module. Again, it's efficacy will probably depend on the module design.

Some smart modules have it as an option.
 

hippy

Ex-Staff (retired)
Unless we want this topic to keep appearing on the forum, it may be time to come up with a more reliable alternative (e.g. code for Manchester or FSK encoding).”
One could simply say use a product which avoids the issues like XBee or other wireless modems, requires nothing more than serial in and serial out. End of issue.

Unfortunately no one can force someone to use what would be best and/or easiest and people will still want to do it some other way for any number of reasons.

The PICAXE is not specifically designed to interface with dumb wireless modules so there will always be compromise and limitations in making that work. There is it seems no 'perfect PICAXE solution', nor a definitive and universal 'how to' which fits all cases, but there is an accumulated wisdom of best practice and so on to hopefully achieve best results. There's nothing to stop people trying to short-cut even on that.

Unless prepared to say, at whatever level, 'do this or we're not interested', discussion will always arise.
 

MFB

Senior Member
Maybe its time for Rev-Ed to provide guidance, in the form of a ‘Low-Power UHF Evaluation Board’ along the lines of their Zigbee offering. Basically a board with a regulator, PICAXE with sockets for standard TX and RX modules and some example communications code. With all the past posts that outline peoples low power 433MHz communication requirements, it should not be too difficult to select RF modules that meets most needs. I can already hear all the objections (which PICAXE to use and antenna connector etc) but then that’s all part of the ‘fun’ of designing a new product.
 

kd5crs

Senior Member
Maybe its time for Rev-Ed to provide guidance, in the form of a ‘Low-Power UHF Evaluation Board’ along the lines of their Zigbee offering. Basically a board with a regulator, PICAXE with sockets for standard TX and RX modules and some example communications code.
I certainly would have bought a pair of these back when I was starting out with wireless. Anything to make the initial process less painful.

There is certainly a lot of knowledge that could go into a PDF document to point out some of the common pitfalls with these things. For instance, back when I was having communication trouble, I thought I had all sorts of crazy problems, when in reality I was simply receiving something different that what I sent, due to the receiver picking the wrong bit to start the character. A simple change of putting $FF between data bytes made all the problems go away.

Oh, and by the way, everyone keeps saying "horses for courses". I know what a course is, but what the heck is a horse?

Brian
 

hippy

Ex-Staff (retired)
Speaking as an individual, not as an official representative of Rev-Ed - would people pay for a prototyping wireless board when most people often seem to be looking for the absolute minimum in cost when it comes to wireless modules ?

For me, plugging wireless modules into a purpose built PCB doesn't seem so different or have any real advantage over using bread board, veroboard or flying wires and soldering the modules direct to the PICAXE.

For the AXE210, for XBee and GPS, a feature and benefit there is it overcomes the issue of the modules being on unusual and somewhat difficult to handle pitch (1mm) plus requiring 3V3 operation.

I'll admit I'm having difficulty in seeing where such a PCB for wireless use would have benefited me or what particular difficulty it overcomes. After all it was only three wires for transmitter and receiver to solder up, two power and one signal line for each.
 

kd5crs

Senior Member
I'll admit I'm having difficulty in seeing where such a PCB for wireless use would have benefited me or what particular difficulty it overcomes. After all it was only three wires for transmitter and receiver to solder up, two power and one signal line for each.
My learning curve went like this:

1. Are the damn things broken?
2. Ok, they aren't broken. Am I actually transmitting?
3. Ok, I am transmitting. Am I actually receiving?
4. Ok, I'm transmitting and receiving. Why isn't my data making through?
5. Ok, my data is sometimes making it through. Why isn't it always?
6. Ok, now it is always making it through. Woohoo!

Items 1, 2, 3, and 4 could be illuminated (pun intended) by LEDs on the boards. On the receiver board, for instance, you could have:

1. A light for power, would always be lit
2. A light for anything being received. In most cases of this noisy RF world, should always be lit or flickering.
3. Then a series of lights that lights up when you send and receive progressively more difficult strings. Such as,
3.1 Light A lights up when "ABC" makes it through.
3.2 Light B lights up when "ABCDEF" makes it through.
3.3 Light C lights up when "ABCDEFGHIJKLMNOP" makes it through.
That would let you know if you have a total lack of comms, or just unreliable comms. You could walk around with the remote unit and see what effect distance/obstacles would have on reception.

The advantage to using the boards would be that you'd know the hardware part (simple though it may be) is up and working, and the diagnostic bits would help with the software. I probably would not have bought these right away, but after point #4 I definitely would have so I'd know I wasn't crazy.

Hardware aside, even a guaranteed bonafide certified code sample in manual 3 would go a long way.

My $.02.

Brian
 

SAborn

Senior Member
Lots of pretty flashing lights you have.

For me i stay plugged in on serial com's and use F8 on the keyboard to see exactly what i receive.

Once all is working as planned then one led could be use to confirm checksum.
 

kd5crs

Senior Member
That's a lot of walking back and forth to a computer, plus it can't tell you why the comms don't work from your actual use location, unless your computer is in your attic or garage or tree. Or what, put the remote unit in the tree, climb down, read screen, climb up, move antenna, climb down, etc.

I guess you could drag a laptop around, but I like my idea better. Blinking lights are always a plus.

Brian
 
Top