[SOLVED] Fibaro Dimmer 2 alternate wiring help needed

Hardware Home Automation
1

Hi
According to the Fibaro manual, the wall switch should be connected to the Dimmer’s Sx connector for power. However, I’m not able to use an extra cable to accomplish this in my installation. I’m wondering if anyone has tried or know if it’s possible to use the circuit Live lead directly to the switch instead. I’ve seen this schematic as an option for the Fibaro Switch, but not for the dimmer. I don’t see why this shouldn’t work, but, I’m not an circuit expert. I have attached an image to illustrate what I’m after. Any advise would be appreciated.

Fibaro%20dimmer%202%20alternate%20wiring

2

I have read that sx is internally connected to live, however I have not tested this myself. You could mail and ask fibaro themselves.

Regards S

3

Hi
Just in case someone is searching for an answer to the same question. I registered on the Fibaro forum to seek an answer and it turned out that others were facing the same wiring problem and wanted to know if the above setup is working. Short answer: Yes it does. However, someone linked to a post where Vesternet had issued a warning using this solution with the reference to that Sx is not just Live. It’s connected through an internal fuse and potentially have other functions as well. They also said it could potentially be dangerous using this setup.
I did some tests with a multimeter and could not find anything that would prevent me from wiring the switch directly to Live. Worst case scenario I would probably damage the dimmer. Nothing worse than that. So, I connected my dimmers like this and they are working fine.

1 Like
4

Continuing the discussion from [SOLVED] Fibaro Dimmer 2 alternate wiring help needed:

Hi sorry to drag up an old topic. I just wanted to check your wiring still worked.

Have you discovered which features magically do not work wiring this way?

It seems Fibaro dimmers specify more wiring than all the others (aeotech, qubino etc).

I have a situation where I am writing dimmers remotely, I cannot fit 3 dimmers into a shallow back box. The switches currently share a common live, I have 2 choices I either wire like you or I steal an earth conductor as my 6th conductor (2 per switch)!

5

Thanks to this article I can learn more. Expand my knowledge and abilities. Actually the article is very real.

6

Hi
Yes, the dimmers are still working fine. No issues. In my setup I have two dimmers sharing a common Live.

7

Thank you Marten

I cannot quite understand why they won’t approve the wiring as per your diagram.

It is an acceptable wiring for almost all the other modules out there, qubino, aeotec, Shelly etc etc.

I e-mailed fibaro and they point blank refuse to say that your solution is acceptable. I cannot run 2 wires per switch into a 4 gang switch, it would be ridiculous. Yet I now have 2 options, either stick with fibaro dimmers and wire them outside of the approved methods OR just go with say qubino dimmers.

Any thoughts?

Nick

8

If I’m not mistaken, the first version of the Fibaro wall-plug never got world-wide approval due to some internal electrical design not being sufficient from a safety perspective. Maybe this is something similar. Internally, maybe some wire is a bit too close to another one preventing it from being an officially approved design. That would also explain why Fibaro doesn’t say it’s OK. I haven’t really given it much thought. It has been working well for me for quite some time now and I use my dimmers daily. I can’t really advise you to use a non approved wiring, but I can say that it is working.

9

Hi,
The reason is simple - the L phase connection connects directly to what looks like a polyfuse, or similar protection device which disconnects the dimmer should a fault occur. The protected L phase then connects to the Sx terminal. Measuring between the L and Sx terminals does indeed show a low resistance, but this is misleading.

Should the dimmer develop a fault (like the power triac welding itself a dead short, or the single insulated aerial wire being chafed and connecting to the circuit protective conductor), connecting Sx directly to phase bypasses the FGD protective fuse and increases the risk of bad things happening.

I am an electrical engineer, but still managed to blow a FGD-212 when screwing the face plate on a 25mm UK metal back box, which is connected to the CPC. The FGD uses a capacitive dropper power supply, so is still referenced to mains voltage, so if the aerial cable shorts to the CPC it lets the magic smoke out violently. The aerial cable chafed after having to reset and re-pair the FGD-212 several times when diagnosing what turned out to be firmware issues in pre-v3.5 software.

Here’s a picture of the topsides of the dead FGD-212 after stripping the board for parts - the blue disc is likely a polyfuse or similar protective component. The other side of the PCB (with the Sigma module) is very black due to the energy of the sudden failure. N.B. this is a multi-layer PCB, so you are seeing several traces together:
fibaro fgd-212 pcb topsides

If you connect to L or phase, rather than Sx, you are defeating a safety fuse when the switch is closed. Like Fibaro, I’d advise against this. Just shorting the aerial created a loud bang, and an arc flash big enough to be seen through the plastic switch plate before the MCB tripped. Mains plus capacitive lighting loads can be very, very nasty!

At the very least consider additional protection such as quick blow fuse rated for mA, and not A - the signalling current is unlikely to be high as S1 and S2 seem to connect to 470k resistors.

Please don’t increase the risk of short-circuits by struggling with a shallow back box - use a multi-tool to deepen the wall opening and fit an appropriately sized enclosure.
In the UK there are also spacers to raise the face-plate which also give space to safely fit components. Depending on your wiring system, there also may be other location options (e.g. for UK loop-in ceiling roses, install in an insulated enclosure above a standard rose).

Best regards,

James

1 Like
10

Hi James

Thank you so much for the detailed but technically accessible reply, I genuinely appreciate it.

I have no reservations hollowing out walls but in this instance I just don’t think it will help.

I am wiring a 4 gang switch all with dimmers.

One gang is a standard overhead live ceiling rose circuit, that is fine I am going to wire the fibaro in the ceiling rose.

The other 3 gangs are zoned wall lights, these are simple switched lives with a common feed at the switch, this common feed then goes off to another switch behind the current back box as a spur. All done before my ownership.

The only way of wiring this space wise is to have remote dimmers.

I have toyed with the idea of dragging an extra 3 core cable up the channel, giving me enough cables (8) to have 6 cables for the sx/s1 loops per gang plus a L & N for the hall light.

Alternatively I can simply elect to have an alternate wiring system, which I was going to protect each dimmer with fuses. Or my other option is to go for Qubino probably DIN rail dimmers in a remote box. These have approved wiring schematics for a switched live straight into them.

My issue is, I’ve only had Fibaro dimmers, I like them and I don’t know if I’m just making life hard or would the qubino dimmers satisfy me? In which case no real re-wiring needs to be done.

11

Hi,
Just wiring a 4-gang switch with ‘normal’ SPST switches is a pain, so adding dimmers sounds like a nighmare. :frowning:
In the UK, I’ve used MK Grid modular plates with momentary ‘Press’ switches for my Fibaro kit and although these do go up to dinner-plate sized face plates with lots of surface area, they would look better in school hall rather than a living room!

If DIN rail kit would work (with centralised spur load / switch wires back to a separate DB), then you can mount Fibaro modules in euFIX D212 DIN rail enclosures. Screw the Fibaro module to PCB terminals, then clip into a DIN rail case. They include S1 and S2 switches, but no a means of pressing button B (unless you drill a hole).

I bought the euFIX D212 when the range was being sold off cheap and it’s handy for my test-bed (created after blowing up the production FGD-212 seen above).

The challenge here is the Fibaro switch wiring is mains-referenced so using 6-8 core alarm cable to save space isn’t an option as it’s only rated for SELV applications. I think the Shelly range might have a low-voltage control contact with a mains rated relay, but haven’t checked.

For me, wall switches are needed as a backup but shouldn’t be needed in normal operation as the whole point of automation as the lights should just work. The Aeotec NanoMote Quad is an excellent Z-wave RF switch for scenes or any openHAB script, although the IKEA Tradfri Zigbee remote might be a cheaper alternative if you can mix RF tech.

With the Matter IoT spec delayed until 2022, products aren’t out yet but I’d not rewire to a specific product at the moment. Whatever the future brings, it’ll likely fit in a double DIN-rail space.

If I were rebuilding, I’d use separate cables from switches and loads back to one or more DBs and suffer the 2+CPC / 3+CPC cable cost to gain the flexibility to use any control technology. If you can pull-in extra cables, a small enclosure under the floor above might well future proof your install (hint - search for ‘Solid Board Access Cutter’).

Unless Fibaro change (e.g. admit failures and share fixed firmware), the open development of ESP8266 / Tasmota/ ESPhome, Shelly, Sonoff, and a load more will eat their lunch - just look at hardware manufacturers working directly with HASS developers.
TTFN,

James

12

Again James thank you so much.

The plan was to spur from the existing lighting circuit into a DIN box, protect this with a 1A breaker and install the dimmers in there. From there run cable to the switch plate and to the existing wall lights.

Maybe I should be reading between the lines but you still didn’t reflect on Qubino dimmers? The only reason I ask is that a fibaro - £45/50 and an eutonomy adaptor at £33 eclipses the price of the qubino DIN module (£45) with the added benefit that I can safely say to myself it is wired in accordance with the manual.

Thank you again.

Nick

13

I should probably have made clear the din rail enclosure was just a way to keep everything neat and tidy, not a full electrical install.

This could in effect be any form of enclosure.

14

Hi,
Sorry - I’ve not looked at Qubino kit, but any device that supports your preferred network and wiring topology sounds good.
For a perspective, I’m a grey beard tech who started with self-built triac controllers, upgraded to X-10 (US kit modified for 240V!), and currently use Z-Wave.
Wiring X-10 was hard, but as I used DIN rail enclosures near ceiling roses, the upgrade to Z-Wave was pretty simple.
On the same basis, I’d expect an upgrade from Qubino to future Matter / magic stuff will be easier if you use centralised wiring and DIN rail enclosures.
Don’t just think of DIN rail as a full ‘fuse box’ enclosure.
CPC used to sell cheap 2 gang DIN rail boxes without bus bars, but ideal for 3G+ CPC (triple and earth) connections to a ceiling rose whilst maintaining secondary insulation for building regs.
I’ve not read the recently published IET code of practice for building automation, but the mix of mains voltage and network protocols is starting to be less surprising to a jobbing sparks as long as good practice of two layers of insulation and isolation are maintained.
All the best,

James
P.S. I’d not buy Fibaro if starting again. Their formerly unique feature of not needing a neutral wire at a switch isn’t worth it - just install a competitor module in a plaster circular box behind the ceiling rose and use a momentary switch. Sadly, Sonoff glass wall switches are very cheap and lovely, but need a deep box and a neutral - not going to work in the UK.
PPS Buy expensive LED bulbs and test, test test dimming. Some work, some explode!

15

Hi @FloatingBoater

Sorry to bother you regarding these fibaro dimmers. Is there any way of PM’ing you?

Kind Regards

Nick

16

Very old thread, I know.

Does anybody tried to measure between S1/S2 and N if wired just with N and L 230v AC?
I get between 40 and 120V DC on S1/S2..
And 12,5V DC if tested with load.
I used a Duspol Digital.

Can anybody confirm this?