I have to second the RFM69 modules over the NRF modules. I was struggling to get more than about 7 metres in open air on the NRF modules that I had, however swapping over to trying the RFM69 modules for me will get me 30m distance through 2-3 brick walls with no hassle at all. I haven't looked back 
Capacitive switch sensor like the AT42QT1011 maybe?
Something along the lines of https://www.sparkfun.com/products/14520
Only thing I'm not sure of is whether the copper wire would trigger the capacitive sensor, although if its being inserted by hand then it should have a charge from your own bodies natural charge as well possibly? Or if it is being fed a charge by something else that is touching it, then it might be enough.
@sundberg84 Thank you so much for the advice, I knew that it would come down to small things here and there, and it's so much easier to tinker around in eagle than try to order a board, find the errors and fix/reorder over and over, so I appreciate it very much!
Time to get back in and make some changes
Good evening!
I have been doing some bits and pieces, built a few temperature sensor nodes that are happily collecting information throughout the house.
I wanted to turn my attention to designing some in-wall modules, as I have a specific light switch for my garage that the builders decided to put in a very odd spot, its on the wall right in the middle of where one of our car bonnets end up, so you have to lean across the car just to turn the lights on!
So this is the first module I have ever designed, its done in EAGLE, with a combination of imported libraries for things like the HLK-PM03, and I have modified a pro-mini 3v3 layout to remove some of the pins that I won't use in this project, to try and keep things a little easier to route.
This is also the first time I have done PCB work with 240V power, so I would appreciate any comments that you might have!
One thing to note is that currently there are just mounting holes for the Live and Neutral inputs, I haven't yet decided on the terminals for those, I'll have to pop down to Jaycar Electronics during the week and pick up some samples for measurement so that I can make a footprint for them. Also, the Serial Upload pins shown in the image are just because I didn't get around to removing them from the design, I would still have them on the end product, but I would spin them around so they run over the top of the pro-mini board.
Many thanks to everyone on the "Safe in wall ac/dc converters" thread, that has been very invaluable for research!
I know that I haven't got any voids in the board to section off the AC traces, that is my next learning experience too
Some things to consider, will be that I will have a capacitive switch sensor for the 3 pin header, I will mount that behind a blank faceplate, and have the LED lighting up the faceplate. Further versions will have more than one switch sensor and LED, this is just to get an idea of size and how well it fits behind a wallplate. I'm also using the HLK PM-03, which gives a 3v3 output, so I don't have to bother with voltage regulation for the RFM69, and the capacitive buttons are happy with a 3v3 input too, with no degradation in range/sensitivity.
image url)
Thats kind of where I'm stuck, I'm not sure what to put on to help it, whether its just that the power circuit is slightly higher output voltage than it would normally get from a battery pack or whether I need to have something else added in....Creating these circuits (While somewhat simplistic) is a bit of a first for me, so still learning as I go
Hi guys... I've been hunting around and hoping that you can give me a solution to my issue! Being christmas, we do a fairly big display at our house (Its a huge thing in Australia!).
I have some solar lights that have worn out over the years, so when the solar controller gets into pretty bad shape, I just knock up a little voltage converter to drop a 5v power source down to about 3V to power the solar lights, and give them a new life, not to mention that I can also turn them all on and off by the same power switches, rather than hunting for all the individual switches out in the garden lol.
I'm a little stumped though, because I have a couple of sets of solar stars that flash, and they would cycle in a roughly 2 second on/off pulse on solar, but now that I've converted it to low voltage mains, it becomes a half second on, 3-4 second off cycle now. I'd prefer to keep them to the same cycle if I can, as now they don't really show up very well.
I have seen some references to the flash being a result of the batteries internal resistance, so I might need to add something to my circuit, but I'm not sure what.... that's where I got stuck.
The circuit is a really simple LM317 with 2 resistors to drop the incoming 5V down to 3V, which then just feeds into the input side of the original light controller, so essentially it replaces the solar panel and the battery inside the compartment.
Any tips?
Not so much cost as an issue, but more being able to pack things into small size requirements, such as light switch casings or junction boxes. I'm happy to use phone chargers for some sensors, but things like light switch sensors and garage or outdoor sensors I need sealed options.
Got an email today from Mouser about a new series of AC/DC modules, and wasn't sure how the idea would compare to things like the HLK units that we are starting to see used.
The package details are at http://au.mouser.com/new/cui-inc/cui-pbo-3-pbo-5/
I'm still fairly new in electronics design, so some of it goes over my head, but on that page they have the diagrams with the application circuits. Does that design mean that in order to use this module, you would need the module itself, PLUS all of the fuses, varistors, caps etc in that application note? Does that then mean that this module would be analogous to just the HLK module on its own as well?
I think this is the reason I haven't yet gone down the route of sensors powered directly from AC (As much as I want to!), is that there doesn't really seem to be a pcb module that has everything included.
@manutremo said in RFM69 range issues:
Hi @executivul ,
You mentioned in your post that the length of the antenna must be tuned to the circuit - would you mind sharing the process to do the tuning?
https://www.mysensors.org/build/connect_radio
Down near the bottom of the page is the lengths of antenna needed for the three primary frequencies used by the RFM board.
Isn't that what the network ID is for though? I thought the network ID was the channel number within the band....or is it more analogous to the SSID in a WiFi type network?
How many thumbs do you require to create one switch? Is there a good thumb to wheel ratio that makes for a good switch? Would you make the switch out of different wheels, say light passenger vehicle versus a late 16th century cart wheel? Would you be able to specify the amount of each type of wheel in the production of said switch? How about thumb types? Can I have one switch made from virgin female thumbs, and would it be more expensive than one made from 50 year old retired carpenter thumbs? I would have thought the carpenter thumbs would be more expensive than virgin female thumbs, as most retired carpenters have lost part of their thumb by the time they retire, so would be quite rare to obtain.
TL:DR, your products suck, and have been blacklisted from my stock ordering options too. So long!
Many thanks @tbowmo I'll sit down tonight and fire up the iron, looking forward to freeing up another pro-mini back into the pool for more sensors!!
Hi guys!
So my shiny little Sensebender gateway arrived today in the post, man am I excited!
Just had a question, I need to solder the RFM69 to the underside of the module, but I have never done that kind of overlay solder before, and just wanted to check. do I need any insulation between the RFM and the sensebender? Or does the green mask provide for the electrical isolation? So, do I just lay the RFM onto the sensebender and solder the pads?
And I'm guessing I still need to have the antenna wire soldered on to the RFM, even though there is a corresponding pad on the sensebender underneath the antenna pad on the RFM.
I have had the same issues as you with the RFM69 modules.
First ones that I bought were marked as RFM69HCW 433mhz, and when I put them together as required, I couldn't get range of more than around 40cm. After that, the reception dropped off a cliff. One thing I did note was that the modules looked different from all the images I could see on this site, which led me to ponder whether I just had crummy modules or not. Also, given the size of the antenna, I was a bit loath to continue with 433mhz radios.
so I decided to up and pick up a new batch of radios, and change over to the RFM69HW (without the C), 868mhz. These ones, when I soldered them up, give me range of around 30m through the entire length of the house, through every wall.
So the only thing I could put my effort down to was that I just had a crummy radio module batch first, and a proper set later, so not sure I can put too much more knowledge into things here.
@NeverDie said:
Strange. Here the shipping cost is $16.45.
Wow, wonder if its because I'm in Western Australia, that final step over the Nullarbor taps on another $135.00!
I thought it looked really good, until I saw the shipping costs of 3/4 of the price of the printer to get it to Australia! Ack! Back to the drawing board
Strange question possibly, but for modules like the RFM69's, do I have to have a fly lead to BOTH grounds? Or just one of them?
Hi guys,
I have built a couple of nodes that I'm trying to slowly add to, so wanted to start with just an arduino nano, level converter, 5v to 3v3 power adapter (LM33) and 434meg RFM69HC's that I have.
I have wired it all up, set one as a serial gateway, and another with mock mysensors, radio's power up and show all ok, but no traffic between them tested using MYSController. Gateway module is powered from my laptop, the remote unit is powered from a 2.1V USB wall wart.
Ok, lets try changing a few things, so I swap the LM33 for LM3940 LDO regulators that I also have (And I'm powering the regulator via the VIN pin of the Nano, so bypassing the 5V and 3v3 regulators on the nano to ensure thats not part of the problem). Still no traffic.
Next step, redo the antennas (Using 1/4 length, so 173mm as advised on some sites and down to 164mm), and on playing with the Adafruit RFM69 test sketch, I find that I can get traffic between the two units if the aerials are only 4-5cm apart max. Ok, a start!
So, tried replacing the stranded wire antennas that I had with same lengths of solid copper wire (Pulled from some cat5a STP cabling), and that has now gotten me up to a range of about 2m (7 feet).
So it looks like things are travelling between the two, but the range is really bad. I have noticed with the transmitter that if I touch the antenna, then it reaches further, so should I try again with longer antennas on the transmitter unit? 1/2 wave maybe? Seems a little odd, as I would have thought that a 1/4 wave, dead straight vertical antenna would be good enough for a pretty decent range.
Would there be any improvement if I was to coil the 1/4 wave wire like you find inside consumer 433meg devices? It seems like it would make things worse than having it straight, but I'm not much of an antenna guy, so I'm not sure what to do next. Is cat5a STP cable too thin a diameter to make a reasonable antenna?
It seems like the RFM is getting enough power, or should I try a larger cap than the current 47uf thats installed on it? Have I gotten an odd version of RFM? I have an HC according to the unit, and most of the stuff I see talks about HCW or HW....
Where do I go from here? I'm aiming to be able to get these to the point where I can be confident enough to start fitting one inside a wall plate, but at present thats a long way off, I want to get the basics right first, and figure out how far I can get rangewise from these.
@vladimir , you are correct, in that you can only connect a load that is less than the rating of the relay's, or you can cause damage. Realistically what you would do is use this unit for lighting where the load required can be calculated and won't really change, or if you were using it for a switch, you would need to have some kind of labelling noting that it was only for a maxmium xxxW output.
What about using the battery packs used in cordless drills / power tools like the De Walt 18V packs or Ryobi One systems?
Find a damaged drill and cut the base off to re use the socet, or just solder direct to the terminals and use a charge module to charge it up. You can buy the batteries separately too.
They are lightweight, have a good charge storage and are pretty rugged, so put up with a bit of pounding.
@dbemowsk try searching for watercutting, you get very nice edges without burnt colouring, and cost wise its generally cheaper.
@jbjalling Perfect, thank you so much for the confirmation! Time to get ordering
Hi guys, I have some 5v arduino nano's that I want to play with the RFM69's (3v3)
On the logic converters, there is this link that has 4 hv and 4 lv links: http://www.ebay.com.au/itm/351686714460
But it talks about having 2 inputs and 2 outputs on either side? So if I need to convert at least three links on the RFM, do I just need one of these that has 4 low and 4 high voltage points?
I guess what i'm asking, do these modules work in both directions or do I need to wire certain links to only specific channels on these modules?
Sorry if its not too clear, let me know if you need more info!
Saw this one come up in my digest today, and looks to be an interesting development:
http://www.theregister.co.uk/2015/12/16/huawei_iot_home_is_pretty_neat/
It sounds like a company with a fair bit of drive, looking to actually create open standards for IoT gear, and still at a reasonable price.
It would probably make for an interesting tie in with MySensors, as you would be able to still create the MySensors network for reading of sensor data, then fall back to a more robust or certified level for things like light switching and mains power level interaction.
Good for those of us who might not have the ability or trust in our own skills to hack apart mains voltage devices, but still want some of the flexibility that MySensors offers?
@pdey What is the per unit cost of those? And do they fit easily inside the wall box say sideways to allow for arduino etc to also fit?
@LastSamurai, the codes look like 16 bit hexadecimal RGB codes similar to the colour formatting used in web pages, formatted as RRGGBBWW. Looks like if its not white it just transmits as RRGGBB though? That way you can have all colours in RRGGBB, and if white, you would be able to transmit that as 255 steps of light level on white only.
Might need some more info, such as car model etc to be able to give more feedback.
I'd love to implement some voice recognition in my setup at the new house, but ideally I would prefer a mic or pickup in each room feeding back to my server which runs 24/7 already. So for me, I would be interested in finding what kinds of mic setups would pick up voice at around 5-6 metre range. I'm worried that might also come with a large price tag too... 
Not sure if it fits the cheap option too much, but I have been looking to use:
http://www.noushouse.com.au/store/product-info.php?Clipsal-CBus-30-Mech-Wall-Switch-pid602.html
I will be using the 5031NMS, which is the slave button, so it has a very shallow depth, fits the normal faceplates and has lit buttons that run on 5v.
So not really cheap ($20 AUD per button), but worth it for my install. The HLK power supplies, an arduino nano and RFM69 module should fit in the wall box behind these switches.
Cheap ones not really... possibly some from ali-express, but the problem you run into is that any faceplates with microswitch type buttons are made for home automation such as clipsal C-bus, so the price suddenly goes through the roof (who would seriously pay $500 for 4 x $0.20 microswitches and 4 x $2.00 single colour lcd screens in a c-bus faceplate??). They have to still get certification, and the demand is low.
You can buy the swith buttons to go into a normal faceplate, but each button will cost more than the microcontroller and radio... It comes down to the question of cheap, good looking or good quality. You can pick any two of those
Great to see it working! Sometimes its definitely a case of going back to the simplest design idea to get the best result. Congratulations on getting the right solution
So (and not to cause issue), is the sketch as it stands returning an incorrect value, and just inefficient? Or is the value being returned incorrect as well?
I'm definitely all for streamlining the code, especially if it is storing arrays of a lot of unused figures. I use a BMP085 around here and in testing, and it is in a location where I want to have a few sensors on the one unit, as it will essentially form my main weather station, so the reworking of the code would be interesting.
I'm not too up to scratch on the code side, still only just learning the semantics of arduino, so I would be keen to see how this one develops. Nice work!
I'd say more likely the garage doors were originally "barn" type swinging doors, rather than sliding doors, and that personnel door was original.
Also, something to remember with your test example, is that the isMetric function is only called in the setup portion of the standard code, in other words when the arduino is first started. So if you are going to swap the functions on the fly then you would need to modify the standard code.
If you have been restarting the arduino's to check then I apologise in advance! Just something that occurred to me on reading your issues.
@Prateek-Gupta I wasn't suggesting that you have the Pi as a controller, simply as a device pass serial/MQTT/MySensors data from the local sensors to the hosted VPS.
The reason I suggested the Pi as a gateway was more for available processing power at a still reduced power useage. Sure, you could use something like an ESP8266, but then you are working within very finite constraints, With something like a Pi and an NRF radio plumbed into the GPIO of the Pi, you could create a gateway that can route MySensors packets over the local network, route the packet into to the hosted VPS, and also do things like create buffers for times where internet connection is out, or add other packages to the Pi to allow for expanded useability that you couldn't do with an ESP8266. That extra memory and flexibility for a few Watts of power? Priceless.
Just remember we are giving ideas based on a pretty vague question. Are you looking to have openHAB integrate with MySensors only? Sure, and ESP8266 could probably do the job. Do you want to add Zwave support too? Uh oh, the ESP is starting to look pretty underpowered at that point.
You would need a local system in order to transmit the local gateway data to the hosted VPS. Something like a raspberry pi would handle that easily, you just need to mirror the serial data to the remote systems, and relay the remote requests to the local mysensors network. The location is essentially irrelevant. It does introduce more points of failure though, so depending on your technical abilities it may be worth doing or it may not, your call!
@Moshe-Livne Depending on whether your incoming power is replaceable or hard wired in your module...
You could look into picking up one of those dual voltage wall warts, that have one output with 1A, and one output with 2.1A. You could then just use one USB lead from the 1A output to run the Arduino and radio, then use the 2.1A output to power the relay board with a sufficient loading ability.
I have done this with my 8 port SSR relay card, and so far in testing it has been super stable, and I have had no glitches in radio performance due to power drops from turning on or off the relays.
Hi all,
I am starting out with building up my MySensors modules, and I am finding that the range on the NRF24L's seem to be quite variable, as well as quite small!
I have my controller radio with the SMA connection and larger antenna, and the remote modules have the NRF24L with just the PCB antenna.
I find that some modules are working at ranges of approx 8m through a single leaf brick wall, and others are struggling at that distance. I had another unit that works at a range of around 13m through either 4 single leaf brick walls, or on a reflective route of around 15m if the signal was travelling through the hallway (I'm not sure which path the signal was taking, so it could be either!).
I have fitted 22uF caps that I salvaged from an old UPS circuit board as I haven't had a chance to hit the electronics store (Might try and get out there this weekend if the significant other lets me), to trial a 4.7uF cap - is there a risk or difference to be gained between a 22uF and a 4.7uF cap?
I haven't tried powering the radio's from the 5V supply run through an LM33 yet, for the same reason above (No access to the electronics store) so these are all running from the arduino nano's 3.3V output, but do people find that it makes a huge difference in the effective range?
What kinds of range do other people get from their modules? I'm just interested to know if other people are getting the same kind of range as me, or less or more.
Thanks!
Just to check, did you upload the DallasTemperatureSensor.ino, or did you upload the Temperature sketch inside the MySensors subfolder?
I'm at work currently so I can't give exact details, but I noticed that if you unpack the MySensors package into the Arduino libraries, you get a heap of example sketches for various items, but then inside the MySensors subfolder are the sketches set up for the MySensors stuff.
So check which one you have used for the remote module with the dallas sensor.
Also on the serial gateway, check the serial output on that (at 115,200 board speed), and check, it should say "Gateway setup complete" or something close to that.
My reading of that main loop is that every time it runs the loop, it will create the message and send it. I think you need an IF statement in there to say "If I receive this message, THEN process the packet and send, ELSE do nothing".
Just a thought, those internal portions of the split system airconditioners have diffuser vents that open when the unit turns on and close when its turned off. You could fit a magnet to one of the vent blades, and use a magnetic reed switch to tell the system whether the vents are open or closed. That could be translated by MySensors as the unit running or turned off.
The plus side of that is that a reed switch costs pennies, and can be bought in a number of colours, so you should be able to colour match the aircon unit pretty closely. You can attach them with strong double sided tape if you cant screw into the unit as well (Use Automotive grade double sided tape, that they use for fixing badges on cars, works a treat), so it shouldn't void any warranties.