Fibaro Dimmer 2 and light goes off after less than 1 asecond

[Alex 31]

I have one problem:
I tried to install the new dimmer 2.

Everything worked fine so far (inclusion, automatic detection).

I have two momentary switches. If I use one of them controlling light works fine.

I am having two light in a row (two lamps at different positions).

 

The dimmer 2 module is connected to L, N. The output of switch 1 is connected to S1 (the other switch is connected to the first switch via

Light 1 is connected to O, than to the second light and back to neutral.

(So phase is connected to L of the module and input to the switches, neutral lead to N and end of second light, output of switch is connected to S1 and this "O" from the module is connected to the first light)

 

 

Now the problem:

If I switch on the lights via app or web page it goes on to the dimming value I want.  but only for very short time (less than a second).

So it switches off directly.

If I repeat this it happens again. But after the third or fourth time the light stays on.

After switching it off again and waiting for a minute (I have not measured this) I can reproduce this behaviour.

 

 

What have I done wrong?

Any more details needed?

[petergebruers 1,261]

The load in the test rig above is a 60w incandescent bulb.

Then something has to be terrible wrong. I have tested dimmer 2 with about 15 different lamp types in 2 and 3 wire setup, with or without bypass.Admittedly, all my dimmers use 3-wire and I have not tested that specific firmware version 2-wire. I also monitor this forum and no one has mentioned this problem (with that test setup with short wires). What did Fibaro support say? Did they suggest exchange?

[redball 0]

I didn't contact support.  I decided to test in a real life environment in the home and when I poped the switch off the switch where I wanted to install the first dimmer2, there was a neutral present, so ran with a 3 wire install - all seems to be OK now.

 

I now have 8 more of these units ready to install.  If/when I get to one with a 2 wire install requirement, Ill post my findings again.

[Guest azevenhuizen]

Just sharing some feedback for a workaround I found.

I ended up adding a small 20nF capicitor between the dimm down (S2) and common (Sx).

This will filter the voltage being picked up by the long cable (appr 20mtr) S2 line just enough.

I used a basic polypropylane 20nF/500VAC, and installed it behind the 'far away' switch.

 

[petergebruers 1,261]

1 hour ago, azevenhuizen said:

I ended up adding a small 20nF capicitor between the dimm down (S2) and common (Sx).

 

I am so pleased with your statement

 

Yes, that might work . Maybe - let me explain.

 

I am one of the first users to test Dimmer 2, and try different setups. I came to the conclusion a capacitor 1 nF up to 22 nF would work in some cases, but not always. Values above or below that range probably won't work.

 

I tried about 10 different cable lengths and configurations and it doesn't always work. If I added a another component, I could make it work in every case. Contact me via PM if you want to know more. I will explain below why I don't want to discuss this in public. It was one of the first dimmers and FW was probably 3.2 so I am unsure if this is still needed. But if you still get the occasional glitch, contact me!

 

I decided not to share this information because adding capacitors and diodes might not be allowed... But now you have crossed that line, I can chime in...

 

Maybe an electrician is allowed to do such a modification in your country. It depends on local regulations.

 

If you do something wrong, this might end badly...

 

You can probably buy a "snubber" of some sort, with insulated wires and proper specifications and approved... I do not have a reference.

 

You probably know what you are doing, otherwise you would not have tried a capacitor in the first place. I am just posting a friendly warning to other forum members.

 

 

[Guest azevenhuizen]

Peter, you are absolutly right.

Being extremly carefull is always important.

 

I am an electrician and I know what I am doing, but even than I always take extra care in live environments.

Therefore, Always make sure all wires and contacts are properly isolated.

Use proper materials and the make sure the capacitor can handle the voltage.

If you feel uncertain, don't even start and ask a pro.

 

Furthermore, I am glad I found a workaround, but nevertheless I still wonder what causes the issue with the dimmer 2.

This same setup used to work with dimmer 1's before, which makes it very strange.

 

[petergebruers 1,261]

10 minutes ago, azevenhuizen said:

Furthermore, I am glad I found a workaround, but nevertheless I still wonder what causes the issue with the dimmer 2.

 

Sure! I also hit the "like" button on both of your posts!

 

It is capacitive coupling. I bet in your case, the output of the dimmer runs alongside the S wire. So the dimmer output feeds back into S. It is only a few nF of coupling between the wires, but the phase output of the dimmer has steep edges so is a high frequency component and it can "pass through the insulation" so to speak. The input can handle a certain level of coupling, but beyond that it is going to detect a "push" of the switch. By adding your 22 nF capacitor, you have made an AC (capacitive) divider which brings back the input level within a range the S can handle... This also explains why it does not happen with cabling systems up to about 3 maybe 5 meter... it usually starts at about 10 m. I also rarely happens if S wires are not near O wire.

 

Does this help?

[hcke 2]

Problem was solved by upgrading to the latest firmware

[petergebruers 1,261]

27 minutes ago, hcke said:

Problem was solved by upgrading to the latest firmware

 

 

Thank you for chiming in!

 

I am not sure it was solved in 3.5... The original 3.2 had it with relatively short lengths... I know it improved on 3.3 but I was still able to find cable lengths that did not work. I do not remember the details because those lengths where... "very long" (I think > 20 m).

 

Did you mean to say: "3.5 solved it *for me*"? If this is the  case, what was your previous firmware version? And what cable length? You still remember when you did the upgrade?

 

I'm sorry to bother you, I am an inquisitive person.

 

Edit: I am curious, because 3.3 was a milestone but it was released not so long after the introduction of the dimmer. I wonder if you can still find a boxed dimmer with 3.2...

Edited January 28, 2018 by petergebruers

[hcke 2]

Yes 3.5 solved the problem for me. I guess my cable is near 20 meters

[Guest azevenhuizen]

I got 3.5 here, but it did not fix the issue for me.

On the other hand, I estimate the cables 25-30 mtrs.

[petergebruers 1,261]

12 minutes ago, hcke said:

Yes 3.5 solved the problem for me. I guess my cable is near 20 meters

 

8 minutes ago, azevenhuizen said:

I got 3.5 here, but it did not fix the issue for me.

On the other hand, I estimate the cables 25-30 mtrs.

 

Thanks guys... That is exactly how I remember it, 3.2 was OK < 5 m and had trouble > 10 m but 3.3 bumped that to > 15-20 m. So I guess 3.5 has actually the same performance as 3.3. @azevenhuizen you got unlucky, probably just exceed the limit. It also depends on cable system ( individual wires or cable bundle...).

 

But 3.5 has fixes for countries that use "ripple control" (for street lights or tariff purposes) so it is worth installing it if you experience flicker at certain (usually fixed) times. It introduced parameter 38: Brightness level correction for flickering loads.

[Guest JBP]

Hi,

 

I believe there is a difference between Dimmers that have been upgraded from older firmware versions to 3.5 vs. Dimmers that have been on 3.5 from the start (different hardware version?).

 

The new ones I bought which were 3.5 out of the box have no issues with long cables at least in my house, while the upgraded ones (initially v3.2) still caused occasional issues after the update to 3.5.

 

In the end I replaced the old ones...

 

cheers,

JBP

 

[petergebruers 1,261]

14 hours ago, JBP said:

I believe there is a difference between Dimmers that have been upgraded from older firmware versions to 3.5 vs. Dimmers that have been on 3.5 from the start (different hardware version?).

 

Thank you for chiming in. A hardware revision is quite possible, I do not own a dimmer that came from factory with 3.5 installed, so I cannot test it. I can tell, Fibaro engineers were able to change the impedance of the input by updating the firmware (I was able to measure it). It is possible, however, they further improved the input at some point in time...

 

So if someone experiences the issues mentioned in this topic, I suggest they send a quick mail to [email protected] to find out what to do.

[k2aab 1]

I have the exact issue, with long cables. Just throwing this into the mix.

I brought a load of dimmer 2 when they first came out, i.e. 3.2, 4 of them have been installed since I brought them and have the issue with long wires even though they have all been upgraded to 3.5.

The other week I installed another 2, upgraded them to 3.5 and they are working fine. Now I though this may be down to the fact that these 2 new ones have short wires, and that may well be the case. However I have noticed the "Device kind:" in HC2 is also different.

The ones with the issue are listed as "Dimmer" while the ones that work perfectly are "com.fibaro.FGD212". Does this mean they are using different templates? Does this mean if I remove these devices and re add them they may come in with the other template and work? I'll test this but what do others think?

Andrew

Please login or register to see this attachment.

Please login or register to see this attachment.

[Al3x 2]

So the solution to this was that the long cable and the lower firmware version has issues which cause the lights to turn on and then off. The solution is to update the firmware to 3.5?

[Alex 31]

34 minutes ago, Al3x said:

So the solution to this was that the long cable and the lower firmware version has issues which cause the lights to turn on and then off. The solution is to update the firmware to 3.5?

Yes. Or using another electric part between cable and dimmer. But for me firmware 3.5 did help.

[XiteHosting 0]

Hello,

Just wanted to let you know that I was also facing the same problem.

 

Having a momentary switch connected between L and S1 or between SX and S1 made the light go up to 100% immediately.

Turning on the light through web would turn it off almost immediately.

 

I used momentary switches with a light in them (on the switch there is a L and N for the light, L is then connected to C (contact)). C is connected to S1.

 

Lots of trial and error later the "simplest" setup that would go wrong was:

Place 1: dimmer connected to load (no local pushbutton)

Place 2: couple of meters away no pushbutton but 2 wires. One connected to SX and one to S1. Touching the wires together for a short time made the light come up to 100% and not the previous diming state. It's as if the wires / pushbutton is pressed twice.

 

It was even so that when I touched the wire going to S1 (not touching SX) made the light turn off

 

After reading the whole 13 pages I was going to try to use a capacitor, or maybe even a dimmer between every pushbutton, linked to the main dimmer at the light. All my dimmers came from my previous home, so I thought there was no chance of having a newer 3.5 version.

Looked everywhere in my boxes, then remembered I had bought one new dimmer, went through all my wall switches (I don't use the Fibaro controller, and the controller I use has no option to show the firmware version) and found a 3.5 version. Replaced the older one, and everything worked perfectly!

 

Maybe this could be added to the opening post, would have saved me reading through the 13 pages. Although maybe the reading the whole topic helped me to get ready to open every wall switch

 

thanks for all the useful information in this topic!

 

regards,

Stijn

[k2aab 1]

On 1/29/2018 at 7:40 AM, petergebruers said:

 

Thank you for chiming in. A hardware revision is quite possible, I do not own a dimmer that came from factory with 3.5 installed, so I cannot test it. I can tell, Fibaro engineers were able to change the impedance of the input by updating the firmware (I was able to measure it). It is possible, however, they further improved the input at some point in time...

 

So if someone experiences the issues mentioned in this topic, I suggest they send a quick mail to [email protected] to find out what to do.

Bit of an old post but thought I'd give my update. 

 

Having updated all of my dimmer 2s to firmware 3.5 with no luck, the other day I replaced an older hardware version with a newer hardware version (FGD-212) also running 3.5. Changed nothing else and the problem is gone. I'm using the older dimmer 2 where I have no need for a physically attached switch, and in this situation it also works fine.

 

So in summary, if you have older hardware versions of the dimmer 2 (mine were the originals), and have switches attached with long cable runs, you are likely to have this problem. The only solution seems to be, replace the dimmer with newer hardware version of the dimmer 2.

 

Andrew

[oolon 1]

I recently updated one of my FGD-212 dimmers to 3.5 and it started having this problem, which it did not previously have. I have it wired in a 3 wire setup and the module is by the light not the switch so not a cross talk issue. I noticed in auto calbration other lights in the kitchen on different dimmers auto calibrated to leading edge however this one was trailing edge. Switch the mode fixed the problem so worth a try.