Jump to content
Light-O-Rama Forums

My fix for LED's Flickering with LOR


Recommended Posts

Posted

Hello all,

I also had an issue with full wave LED's flickering with LOR. I've read all the posts and saw where people were building snubbers. I took the easy way out and simply plugged a 3 way T into the offending LOR circuit. I then plugged the string of LED's into one socket and a simple night light into another socket. Voila, flicker cured! Hope this helps someone else out. It was a cheap and easy solution that required no electrical expertise.

~Doug

Posted

dkh1020 wrote:

Hello all,

I also had an issue with full wave LED's flickering with LOR. I've read all the posts and saw where people were building snubbers. I took the easy way out and simply plugged a 3 way T into the offending LOR circuit. I then plugged the string of LED's into one socket and a simple night light into another socket. Voila, flicker cured! Hope this helps someone else out. It was a cheap and easy solution that required no electrical expertise.

~Doug

Yea, that works well Doug. Several people on the board have used similar things like air fresheners, etc. It's a good simple idea, but also has it's disadvantages. As for me, I didn't want the white light showing, but more importantly the extra wattage. I have to use about 35 of them on my display. If I were to use a night light on each of them, it would cost me $90 more than the snubbers and an extra 225 watts of constant wasted electricity (those night lights draw about 7 watts each). A snubber is less than 1/2 watt.
Posted

Both are good solutions. The night light is quick and easy and totally non-technical. The snubber is more efficent.

Posted

I can vouch for the benefit of snubbers. I was having all sorts of problems with my boards (and I am on 240VAC as well which means I am even closer to the TRIAC avalanche voltage) and found that adding a 100nF and 100ohm carbon composite resistor in series across the TRIAC solved my problems.

LightORamaDan, with more and more of us using LED based gear with lower load currents, has there been any thoughts of actually incorporating the snubber components into the PCB design?

Posted

As I understand it and I may be all wrong, the current triacs are "snubberless" so I am assuming that there are triacs out there that have snubbers built in? If so, do they cost a lot more and could they be used in future LOR boards?

Posted

Richard Hamilton wrote:

*snip*
If I were to use a night light on each of them, it would cost me $90 more than the snubbers and an extra 225 watts of constant wasted electricity (those night lights draw about 7 watts each). A snubber is less than 1/2 watt.

What are these 1/2 watt snubbers you're talking about?
Posted

Duke,

A snubber is the name of a circuit designed to filter high frequency spikes and prevent them from reaching the TRIAC in the first place. I have never seen TRIACs with them built in. The circuit is a simple resistor (say 100ohm carbon composite type) and a high voltage capacitor (around 100nF) placed across the TRIAC from input to output to stop the voltage spikes that result from a TRIAC mis-fire or just general switching hash getting back to the TRIAC causing it to "avalanche fire" - which will cause it to "turn on" by itself. This occurs because once you get above the TRIAC breakdown voltage, regardless of the control signal on the gate, the TRIAC will conduct anyway. If you have a 600V rated TRIAC, apply a voltage greater than 600V and it will conduct anyway - at least up until the point it fails completely due to excessive voltage.

You can generate impulse voltages greater than 600V easily in a phase controller by changing voltages very quickly (i.e. switching at a time other than a zero crossing of the AC waveform or by dimming which means you are switching mid cycle) with inductances in the circuit. The inductances can simply come from your power cords connected to the controller and the combination of an inductance in series plus rapid voltage change causes voltages to spike as the magnetic fields are varied over the inductance.

Low loads can make the phenomena worst because the TRIAC may momentarily fire but if the load current is too low the TRIAC will drop out again very quickly - creating a rapidly changing voltage (far shorter than one cycle) and hence the potential for these spikes to occur.

In Australia where we run 240V these problems are exascerbated because with the higher voltage, the current is lower for the same power lights. We also end up closer to the avalanche voltage of the TRIACs more easily.

Placing "dummy loads" or additional loads on each channel isnt really being a snubber - but is simply reducing the chance of mis-fire due to reaching the TRIAC dropout current. I think the 16A TRIACs used in the LOR CTB16 boards need nearly 80-100mA to stay conducting. At 110V that is not too bad but at 240V (remembering as volts go up, amps come down for the same power) it is very easy for us to reach the mis-fire current limit - causing the controllers to mis-behave more readily.

Hopefully that explains it a little better.

Regards,
Grant

P.S. This is the theory as I understand it. My power electronics theory is a little rusty as I haven't used it for 10 years until this year starting up with my first LOR attempt - hopefully if I have got something wrong here Dan or others can correct me.

Posted

The "snubbers" that people are referring to is actually just a "load." I think we all should stay away from the word "snubber" from now on, as it is confusing.

Stan

Posted

Sorry Stan, but snubber is the proper term to use. A resistor in series with a cap in parallel with the input and output is correct if the input and output is the source side of the Triac and the output side of the Triac. Now if you put it across the output of the Triac and netural terminals it is a load. See the difference? We at work will put an R.C. circuit across the contacts of a relay to snub out the CEMF of an inductive load so that the contact do not burn out as fast from the CEMF induced ARC. The purpose of the snubber is to reduce the ARC.

And in sense that is what Grant is talking about.

Max

Posted

Ahhhh...... yes, I do see the difference. Also, the zip cord or any extension cord does have parasitic capacitance as well.



Max-Paul wrote:

Sorry Stan, but snubber is the proper term to use. A resistor in series with a cap in parallel with the input and output is correct if the input and output is the source side of the Triac and the output side of the Triac. Now if you put it across the output of the Triac and netural terminals it is a load. See the difference? We at work will put an R.C. circuit across the contacts of a relay to snub out the CEMF of an inductive load so that the contact do not burn out as fast from the CEMF induced ARC. The purpose of the snubber is to reduce the ARC.

And in sense that is what Grant is talking about.

Max
Posted

stanward wrote:

Ahhhh...... yes, I do see the difference. Also, the zip cord or any extension cord does have parasitic capacitance as well.

Thats true and it is also true of any close parallel running pair of wires. But usually the inductance is greater that the capacitance.

Max

Stan, good discussion. Are you by chance an Amateur Radio Op? KF0OX


Max-Paul wrote:
Sorry Stan, but snubber is the proper term to use. A resistor in series with a cap in parallel with the input and output is correct if the input and output is the source side of the Triac and the output side of the Triac. Now if you put it across the output of the Triac and netural terminals it is a load. See the difference? We at work will put an R.C. circuit across the contacts of a relay to snub out the CEMF of an inductive load so that the contact do not burn out as fast from the CEMF induced ARC. The purpose of the snubber is to reduce the ARC.

And in sense that is what Grant is talking about.

Max

Posted

Hi Max,

Yes, I am a HAM, WH6CMB. Only a tech, haven't tried for my General yet (been a Tech for 20 years now). I haven't used my radio in ages........... Also am an EE too, but heck, what do I know?

Where are you located?????

Stan

Max-Paul wrote:

stanward wrote:
Ahhhh...... yes, I do see the difference. Also, the zip cord or any extension cord does have parasitic capacitance as well.

Thats true and it is also true of any close parallel running pair of wires. But usually the inductance is greater that the capacitance.

Max

Stan, good discussion. Are you by chance an Amateur Radio Op? KF0OX




Posted

Hi Stan,

Shame to say also, I have not fired up a radio for the past 8 yrs, and then it was only 2mtrs. QTH is about 60 miles west of the Gateway Arch in St. Louis, off of I-70.

Not and EE myself, but I used to do componet level repair and reverse engineering. No expert, but my father-in-law used to say "you know some shite". Man had no EE, but was involved with the Astronauts and the Mercury capsule. He too was a ham W0NSV but SK now 10 yrs.

Good to meet you.

73
De Max

BTW the internet is the bane of Amateur Radio. Look at us chatting and not having to wait for skip to come in or light up our neibhors with 1KW.

Posted

You seem to know a heck of a lot for an electrician! Yeah, the internet really changed a whole lot of stuff! I guess no motivation for me as well to use the radio to chat. No need for QSL cards anymore? Do they have "online" QSL cards we could send each other?

I also heard, there are programs that would interpret CW as well. I also heard that the future licensing may remove the CW aspect of it?



Max-Paul wrote:

Hi Stan,

Shame to say also, I have not fired up a radio for the past 8 yrs, and then it was only 2mtrs. QTH is about 60 miles west of the Gateway Arch in St. Louis, off of I-70.

Not and EE myself, but I used to do componet level repair and reverse engineering. No expert, but my father-in-law used to say "you know some shite". Man had no EE, but was involved with the Astronauts and the Mercury capsule. He too was a ham W0NSV but SK now 10 yrs.

Good to meet you.

73
De Max

BTW the internet is the bane of Amateur Radio. Look at us chatting and not having to wait for skip to come in or light up our neibhors with 1KW.
Guest
This topic is now closed to further replies.
×
×
  • Create New...