💬 RFM69 Livolo 2 channels 1 way EU switch(VL-C700X-1 Ver: B8)

[jirm]

@mtiutiu

Hi all

Can I suggest some improvements for that project?

For sure I agree with @tonnerre33 about make a version board for only one gang switch.
I buy regularly (once a month or so) some parts from livolo and from last 4 or 6 months I see that the switch plate boards (for the EU version) are the same on hardware specs at least from 6 or 9 month ago.
But we can expect that in near future (maybe few months) Livolo manufacturer make some changes and updates on his designs, because they regulary are doing that in past. So we need keep prepared for that and for make the propper updates to this project to mantain it working with the next version Livolo switchs.

1. -One gang switch I think is most common switch people have installed or at least I think is needed too with this two gang project version.

2.- Maybe will be better use (or make another plate board version) for the RFM69HW (high power), because the pinount on HW not match with RFM69CW you use, and HW type have same working specs but with the plus that HW type have high power possibilities and is most easy to find and both have similar price.

3.- How can be little better documented all the changes (wiring) we need do on the power relay plate board?
At least for me is hard to do correct wiring only seeing the photo and is easy to do some mistake trying to solder the wiring in the correct pins and places.
So I suggest trying to do some more work on that and maybe include some scheme and plan and take all best pic from each wiring bridge is needed to do that we can see without doubts the correct place to solder it.

PD: @tonnerre33 I think one or two way switch function not be affected with this project plate board, because that function only differ on the pic (MCU) Livolo switch firmware and litles changes on power relay plate board to wire the additional pic pin output two way function to the COM connector.
Nothing of that function should be affected by anything that this plate modifies on the Livolo switch.

My best wishes !

[tonnerre33]

@jirm said in Livolo EU switch Mysensors integration:

Nothing of that function should be affected by anything that this plate modifies on the Livolo switch.

I hope because i have many 2 way and the same for my futur home

And what about NRF24L version ?
Many people are using RFM69 but 2.70$ RFM vs 1$ NRF ... Ok for command the sensitives nodes like roller shutters but why for the light ?

[jirm]

@mtiutiu
Of course all we are here to help you make all the proper test needed with all Livolo parts that we speak. One or two gang with/and one or two way, other plate hardware revisions, etc...

That all we need for do that is the avaibilitty to get your plate boards designs working on our houses.

Personally I'm so busy (and so lazy) to make all this project working from scratch trying to find plate board manufacturer, solder all (wooow smd) parts, etc... etc... and need if possible some at least "semi-mounted" plate boards printed and with the smd parts soldered.

[jirm]

@tonnerre33

Personally I think that much better specs from RFM vs NRF and the working radio band (433, 868 Mhz vs 2,4 Ghz) are most than enough reasons to spend that 2 $ more on RFM.

But...more always is better. So we can dream with another plate board version for NRF in the future...who knows

[mtiutiu]

Hi all,

So after some time of testing I give you the final revision of this board which is 0.5. The latest files are available on the openhardware.io page but always the most recent are on my github repo of course(I'll try to keep both in sync anyways). By using good tantalum caps where required I got a good stability for the node. It will be kept for testing for one month still starting from this week(testing already started). So far I'm very satisfied with the results.

As a final note for all of you there: when designing hardware always use good components - especially good filtering capacitors because most of the issues and instability happens because of that.

I will try to design the pcb for the one channel variant of the switch too when I'll get my hands on it so stay tuned. I see that some people asked why RFM69 instead of NRF24L01: this is because the RFM69 module is superior in terms of performance/stability/range compared to the latter(trust me it worth the money difference as I tested both along time). And because it works in the sub-GHz band it's also less prone to interference compared to the 2.4GHz band which is used by all Wi-Fi equipment.

[Cliff Karlsson]

Wow, great. I guess there is no way to be able to buy these boards pre soldered w. components?

[tonnerre33]

Hello, for information, the version has changed Here is the new board : https://fr.aliexpress.com/item/Free-Shipping-EU-Standard-Livolo-AC-110-250V-The-Base-Of-Wall-Light-Touch-Screen-Switch/1898648882.html
But there is good news, 3V and 12V are directly accessible

[Nca78]

@tonnerre33 said in Livolo EU switch Mysensors integration:

Hello, for information, the version has changed Here is the new board : https://fr.aliexpress.com/item/Free-Shipping-EU-Standard-Livolo-AC-110-250V-The-Base-Of-Wall-Light-Touch-Screen-Switch/1898648882.html
But there is good news, 3V and 12V are directly accessible

Might be bad news in the (very) short term, but it will soon be much better as it means after modification to adapt to the new version @mtiutiu board should not require any hardware modification on the main power/relay PCB as 12V is now available.

[mtiutiu]

Well..I was expecting this to happen at some point in time. This is the nasty thing that may happen when you rely on some external board(s) to do your work. It is a good thing that the 12V line is now exposed..but the pcb needs to be recreated.
And who knows in the future when it might change again. So in the end the best solution would be to have all the boards made(including the power/relays one) so that no one needs to rely on the Livolo company. But for the moment I don't have enough knowledge/time - at least not too much regarding power electronics and creating live wire only power supplies(including testing/safety/etc).

[alexus]

@mtiutiu Hello, I wanted to buy 2 devices. For sale? how much is?

[mtiutiu]

This post is deleted!

[mtiutiu]

Hi,

Because some people asked here: I don't have anything for sale unfortunately and I don't have time to produce and sell ready made boards. The project design files are available for free so anyone can build it. But be aware that you need to have the right livolo switches(as seen from the project pictures). Latest ones produced by the Livolo manufacturer use another hardware revision which is not compatible with the pcb that I've designed.

Regarding BOM and total price for producing one board I didn't had time for that either and the thing is that the price varies depending from where you supply the components.

[alexus]

@mtiutiu Thanks for the quick response, can only eat PCBs without components? I can not make a PCB.

[shabba]

Hi @mtiutiu - Firstly nice work. I bought a few and sent on a little gratitude also.

Is there a revision or something on the lenovo switches to make sure that I have the right one? I have quite a few now and a few more on the way so I just want to make sure I am all set.

Thanks!

[jirm]

@alexus

Few weeks ago I uploaded and shared to Oshpark pcb web service this @mtiutiu pcb design (but take care !!! I think this is not last one version @mtiutiu design), so anyone can easily find it there and order build some pcb if like.

You can check it at https://oshpark.com/shared_projects , and easyly find it with only enter "livolo" word in the search case.

Sadly I don't have enough time yet to involve more in this project, but in near future I'm going to that and test that awesome design that this project have.

Regards

[mtiutiu]

@jirm The differences between the hw designs that I made are very small and have no impact regarding the final functionality of the board(the OSHpark version is ok too). The biggest change was to trim down a little bit the round corners of the board so that it fits from the start inside the switch case(otherwise you need to use some abrasive paper to trim it a little bit - but not much). I also added a ferrite bead to filter high frequency noise that may come from the dc-dc switching power supply and replaced the important filtering capacitors with tantalum ones(but the old board revision model can be used also to solder 0805 type tantalum caps as the footprints are large enough).

@shabba VL-C700X-1 Ver: B8 is the hw revision of the Livolo switches that I have. But this is marked on the power/relays board from inside and not on the enclosure. So in other words you can see that revision only if you open the switch. Otherwise it's impossible to say as the Livolo manufacturer sells it using the same model number(like VL-C702 or something like that) but the hw revision can be different and can only be seen inside the switch. You can see the board that I have by inspecting the pictures posted on the project page.

[alexus]

@jirm Thanks!

[shabba]

Crap! I have version C2. I have already ordered all the BOM components! Doh.

Can the old PCB be hacked into working with this? What is the actual difference?

Thanks!

[mtiutiu]

@shabba

Can you post some pictures of the new board? I'm curious how it looks on both sides. I'm referring to the power/relays board underneath. I may need to recreate this project board after the new one because my friend for whom I made this project initially bought the new version of the Livolo switches. So in the end I think I'll work on the new version but only after he receives the switches as I need to compare and see the differences from both a mechanical and electrical point of view.
Don't worry about the BOM as that part stays the same. What could change is only the PCB layout and its size and/or outline.

[shabba]

If you want better close ups etc let me know.

link text

[shabba]

Hmm.. I added some links in last post but they don't seem to show..

[mtiutiu]

@shabba

There's a button called upload image with an arrow inside it and a cloud symbol underneath.

[shabba]

Shared album here: https://goo.gl/photos/H5w7u9cNbiS6SGRD9

[mtiutiu]

@shabba

Can you focus a little bit better on the power/relays board? Those pics are a little bit blurry. Thanks. Oh..and I almost forgot - thank you for your gratitude.

The main change is that they added a 2x7 connector now instead of the old one which was 2x6. But that's not a big problem at all as I can recreate the PCB to accommodate the new one. Still I have to wait for the Livolo switches to come in my hand so that I can identify the changes regarding the new connector(where the relays are connected and the 12-14V line if any). Other than that I don't see any big changes only the rounded corners are not so round now and they have a cut. The MOSFET is placed now at a 45 degree angle and the relays driving IC is a TSSOP one if I see well in the pictures. The 3.3V or 3V regulator is placed somewhere else now but that doesn't matter as I don't need it.

So in short: I need to recreate the PCB to accommodate the new 2x7 connector and to add cuts in the corners. BOM will not change.

[shabba]

Added a couple more pics.

[tonnerre33]

@mtiutiu It's the PCB that i made here : https://github.com/tonnerre33/Linovo_EU_2Gang_1way/tree/develop/

You can take many informations that you want
I'm curious to see how you'll do the pcb for compare with mine and learn

[alexus]

@tonnerre33 Great work. Can design with esp8266, it would be fine

[mtiutiu]

@tonnerre33
Nice work. I will steal maybe the outline of the board(board frame) :simple_smile: . Did you tested it from a mechanical point of view? I mean will the board fit in the enclosure? And I see that you use the existent voltage regulator from the Livolo power/relays board...I didn't wanted to use that because the voltage drop over it is pretty big(12-14V - 3V =~ 9-11V) and with that in mind considering the TX current of the RFM69 which is around 50mA gives a max power dissipation of: 11V x 50mA = 550mW - pretty high(I didn't added here the MCU, LEDS and other stuff). Now this only happens on TX time which should be pretty short in theory so it should handle that(but better check the voltage regulator datasheet though). Rest of the time it will be around 16mA(RX current of the RFM69W) + a few milliamps for the MCU and other components - around 20mA let's say and this translates to: 11V x 20mA = 220mW(half power dissipation compared to TX mode).

Now depending on how much time you're in TX mode compared to RX mode this will be around 300mW maybe if we take the average(but this is just a rough estimation). The onboard regulator that they use is a Holtek 7130-1 device which can whitstand a 24V input voltage BUT only 30mA output current max and it can dissipate using the SOT89 package around 500mW maximum so it may heat.
I don't know if on the current board they changed it or not but still imho it's not a good idea on the long term to overload it. Indeed if the average power dissipation doesn't reach that value(500mW) it's ok and you should stay below that: usually it's a good idea to dissipate half of the max allowed power only.

And I saw that the 3V voltage regulator powers other stuff on the relays board not only the front plate so we should take that into consideration too.

Given all of the above and thinking on the long term that's the main reason I used a DC-DC step down converter powered from the 12-14V line which is also the voltage used to power the onboard 3V regulator that Livolo uses.

Most of the problems(if not all) come from a bad designed power supply. It seems that most people underestimate this part but what they don't realize is that the power supply is the most important part of every circuit and that it plays a very important role regarding the final performances and stability of the whole circuit that it powers.

[tonnerre33]

Thx mtiutiu.
I followed your advises and i converted the project to kicad
For the board frame, i have tested it with paper but it's difficult to check it because paper is flexible..

Same for the position of the 2x7 pins connectors, i think it's ok but i'm not sure until i have tested it with a real pcb.

You are right for the power !! I haven't check the current consumption of the RFM !! OMG 45mA in tranmit mode when NRF24L01 is 11.3mA (but with 0dBm output power for NRF).

For RFM it's 20mA with 0dBm. Can't we set RFM to 0dBm ? I don't know the consequences for the range if we change this parameter...

The installed regulator is 7130-1 = 30mA for Vin=5V. Maybe it's more current for 3V ? I didn't see the information in the doc

[mtiutiu]

@tonnerre33

If you keep the average power dissipation under 250mW(half of the max rated value from the datasheet) it should be ok. Reducing the Tx power helps but will decrease coverage - but that's not a problem if you set your nodes as repeaters too as this will increase coverage(not the battery powered ones of course). So yes there are solutions don't worry.

Oh and I used paper too when I replicated the original board outline, touchpads and main connector placement. Then I scanned it, imported it into Inkscape, a little bit of QCAD...and in the end got the final results. There may be an easier way of doing it but I didn't had any better idea(s) at that time.

[MiKa]

Hi,
What about change 7130-1 to for example: MCP1703T-3302E/MB ?
MIKa

[mtiutiu]

@MiKa

That device is rated at 16V max input voltage...I won't use that because the input varies from 12 to 14V approx. So it's like living on the edge. Why are people afraid of including a DC-DC converter into their design when high voltage drops are involved and a little bit more output current is required - actually we can speak in terms of power here as it's more appropriate. So we all know that when it comes to more power a DC-DC converter is more efficient than a classic LDO. Is it because of noise? That can be filtered not a problem and reduced to an acceptable level. In terms of volume - indeed there are more components involved and a little bit more care is needed when designing the PCB but hey we need to make some compromise in the end...and I for example like to stay on the safe side on the long term.

[alexus]

Is there a connection example with ESP?

[mtiutiu]

@alexus

Hi, I don't know of any. There was @DJONvl who claims that he did it on this thread: https://forum.mysensors.org/topic/2775/livolo-glass-panel-touch-light-wall-switch-arduino-433mhz/75. But it didn't got my attention as ESP or any WiFi module is too power hungry for this project where simple sensors/actuators are involved and very light in terms of power usage(including radio transport). But as I said this is only my personal preference.

[mtiutiu]

@tonnerre33

Nice work regarding Kicad. You'll see that on the long term it's a real benefit to switch to Kicad and I strongly adhere to it even though some people would say the contrary. It is indeed a little bit hard maybe to start with it and the learning curve is not the easiest one but after you master it then it will be a joy to work with. And more features/bug fixes are made as we speak because it's a very active developed project now. And on top of that that it's FREE - no limitations or whatsoever.

[shabba]

I got my PCBs today in the post. Is there a way to hack them to work with new circuit or should I just bin them?

[mtiutiu]

@shabba

Until I get my hands on the new switches I cannot say. The thing is that they use now a 2x7 connector instead of a 2x6 one. I don't know how the relays are mapped to the new one and the 3V line. But I think that for sure a new pcb needs to be created. But until then maybe @tonnerre33 can shed some light here with some pictures? Is the new connector mapped ok in your schematics?

If I see well in the pictures they moved the connector in the opposite direction so yeah...a new pcb design is needed.

And to be honest I cannot keep up with the Lenovo manufacturer and create every time a new pcb because they maybe didn't slept very well over night and decided to change hw revision of the power/relays board on which this project relies. This is overkill and it's a downside for this project as I need to rely on their power supply board.

Now as I promised I will create a new pcb for the new hw revision but ONLY for that and not the future ones. The new hw revision that they use now is the one that @tonnerre33 and @shabba has which is: VL-C702X-2 VER:C1

But @shabba there's @tonnerre33 who works on this and started ahead of me so maybe you two can sync up on this?

[alexus]

@mtiutiu Very much I ask, prompt as to connect esp-01 instead of RFM

[mtiutiu]

@alexus

Creating a new board with new hardware/new stuff like the ESP-01 is a thing that requires time and testing which is a luxury that I really don't have now. So I'm sorry but I won't invest time/money in creating/testing boards with ESP modules. The only thing that I'm going to do is to re-design the RFM board only as it doesn't require too much changes/work/time/money and it was tested beforehand.

[alexus]

@mtiutiu I'm not asking you to design the board again. If you can tell how to connect the ESP module to your board?

thanks for the advice

[tonnerre33]

@mtiutiu Yes the new connector is mapped for the new version. I have checked the pins with the multimeter.

I have done many picture that you can see here :

http://www.photorapide.com/albums/jordan/kqadrk/livolo-vl-c702x-2-ver-c1.html

I hope you will find the informations that you want.

Caution if you look my schema for see the mapping of the pin because on my board the bottom is your top and the top your bottom

Edit : I have added the mapping in the picture album

Did you know why the pin in red surrounded is connected to the VIN ?

[mtiutiu]

@alexus

Ok...my bad. I didn't understand your initial question. But now the next question is: what do you want to accomplish exactly? Can you be more specific?

[mtiutiu]

@tonnerre33

Thanks for the pics and for your time. Well that's how the SOT-89 package is designed. If I'm not mistaking the big metallic pad was made like that for better thermal dissipation if you make the copper area which it connects to big enough. Now why it was specifically connected to Vin...that I'm afraid I don't know.

[alexus]

@mtiutiu I ask you to help me connect ESP-01 to your board

[bjornhallberg]

Just want to confirm that I also received a "VL-C702X-2 VER:C1" today in the mail so at least they haven't made yet another version.

[tonnerre33]

Hello @mtiutiu i think i have finished my first version.
I have modified the pcb with your suggestions (power the node with 12VCC).
I have corrected a big mistake too, the AT25DF512C footprint was wrong (convertor mistake i think) ...

I have to put 2 resistors on the prohibided face ^^ but normaly it should be ok

[shabba]

So besides the header being bigger what else has changed? I have received the current PCBs and wondering what I might need to do to make them still work?

[tonnerre33]

@shabba said in Livolo 2 channels 1 way EU switch(VL-C700X-1 Ver: B8):

The position of the header has changed too. Now it's on the right part.
I think you can use your pcb but you'll need some wires between the femal header and the new mal header.

[shabba]

@tonnerre33 Thanks - do you happen to know the pin out?

[tonnerre33]

Yes, you can find the new pint out here :

http://www.photorapide.com/images.php?photoName=dm0l6i.jpg&photoId=1002306

[shabba]

@tonnerre33 Thanks! Do you know the old one?

[hhalibo]

@tonnerre33 I just browsed your Github repo, found nothing there today, you deleted the contents？

[hhalibo]

That's my fool. I saw it in the develop branch.

[tonnerre33]

@shabba You can see the old one on the picture of this project

@hhalibo Yes i'll push this project in the master branch only when i'll have test it. @mtiutiu is working on it too, i did this node for training me because i'll develop other livolo EU
I think the mtiutiu node will be more optimised. I learned and i hope continu to learn many things with him.

[shabba]

hi @mtiutiu - I see https://github.com/mtiutiu/kicad/tree/master/mysensors/node/livolo_EU_switch/VL-C702X-2_Ver_C1 - Is this good to be ordered? If you give me a link on where you get a % I would like to order from there.

Thanks.

[mtiutiu]

@shabba

It's not finished yet. That's just a draft for the upcoming new hardware. When it's finished I'll publish it. As I said all of my work is done using my free time. For now I'm on vacation...will start working again on this project next week.

[shabba]

@mtiutiu I was not meaning to sound pushy! Enjoy your vacation.

[dzungpham0703]

How about this project now?

[freynder]

@dzungpham0703

It is abandoned, have a look at the description.

[dzungpham0703]

@freynder Yes, but he start new project here https://github.com/mtiutiu/KiCAD_Playground/tree/master/mysensors/node/livolo_EU_switch/VL-C700X-1_Ver_C2/CYBLE-012012-10

[mtiutiu]

@dzungpham0703

I replied in the nrf51822 thread. Yes I managed to make it work like the original with BLE stack. And using the wonderful Cypress Proc BLE solution including capsense and lots of goodies . Cypress really rocks! I love them.

The current consumption was lowered to 200uA without being connected and 700uA connected. Those numbers include LEDs current consumption and other blocks like touch sensing.

I'll get back with more updates when it's finished.