LED Tape acts up - with solution

Merry Christmas
Status
Not open for further replies.

PetrosA

Senior Member
It's been a long time since I haunted this forum :D

I learned something today I think might be helpful to some of you at some point. About a month ago I install led tape in a kitchen. Two 36" pieces under two cabinets and two 39" pieces vertically in a cabinet along a different wall. It's DC tape with a magnetic driver that has a DC output (all parts are from GM Lighting) and is controlled by a Lutron AYLV600P. It was flickering badly at almost any point in the dimmed down range for the homeowner and she called me back. I confirmed the flicker and started troubleshooting to make sure that the load was in spec and that all connections were good. I confirmed a 56W load based on 4.4W/ft of tape and called tech support. They told me, however, that the 60W driver is actually only good for 80% loading when using a dimmer. I disconnected the two vertical pieces to confirm and sure enough the flickering stopped so I ordered a 100W driver to replace it. I installed that yesterday, came out of the attic, turned the lights on and... sonofa... They still flickered. Even stranger, when I again disconnected the in-cabinet lights the flickering stopped again. At that point I tried calling tech support but got voicemail. I kept trying what I could - checked input and output voltage, output frequency, tried a different dimmer etc but no luck. The flicker wouldn't go away. In the meantime a friend of mine called and I gave him a rundown on the situation peppered with lots of four letter words just to vent. At one point I said, "it's as though I'm *$%#* picking up %&#$@$ interference through the @^*&#$ LED tape when they're all connected." He shot back with "Dude (he's from California), I'll bet your tape is forming a dipole antenna," and suggested that I try and hook up the vertical tape to one driver and the horizontal tape to another driver. Sure enough, they all work perfectly now without a hint of flicker. It cost me a few hours of troubleshooting and a second driver on my dime but the problem is fixed now that the vertical and horizontal runs are isolated. In the future I will be pricing every cabinet job with two drivers just in case.

Obviously, I don't have RF test equipment to determine what kind of interference I was dealing with, but my first guess would be an air handler in the attic not far from the kitchen. It could have a noisy circuit board or transformer. I thought about shutting circuits off to see if I could isolate a culprit, but since the homeowner wasn't there and I didin't know if they had anything plugged in that might get messed up from loss of power, I didn't try that.

I eventually got in touch with the tech support guy later on and asked him if they've had feedback from other installers about the tape acting as an antenna or if using the aluminum channel would help prevent this but he admitted that he hadn't gotten any feedback about outside interference and didn't know whether the aluminum channel would act as shielding. They have had complaints about the LEDs coming on when the driver is installed near a motor, like under the sink near the disposal. All he could really offer in my case was that their product is required (like most others) by the FCC to accept any outside interference.

1508778473516.jpg
 
171122-1245 EST

PetrosA:

First, you need to paragraph your posts.

Second, you need to know how devices you work with work.

I really don't know what your circuit is, but I am guessing:

One possibility:
(1) That the dimmer operates on the 120 V power source to produce some sort of phase shifted modification of the 120 V signal.
(2) This modified sine wave is applied to the input of a transformer that has an output going to a rectifier to provide your DC to the LEDs.

or

Another possibility:
(3) The modified sine wave goes to an electronic converter from 120 V to some lower DC voltage.

or

A different circuit would take the 120 V AC input and produce a constant DC output voltage at a value somewhat above the LED maximum brightness level. Then a DC regulator (dimmer) would adjust the DC voltage to the LEDS.

This approach is unlikely to produce the problem you describe.



I did a quick search on your Lutron dimmer and found no useful technical information on how it works.

Since it is described as a magnetic dimmer my guess is that this is a dimmer designed to work with a transformer or other magnetic load that can be forced into saturation by a DC component in the AC voltage. To prevent saturation of the load's core material these so called magnetic dimmers must be designed to produce an output where the average positive current equals the average negative current, thus, zero average DC. Note that I did not say RMS current.


I conjecture that if you created a DC source consisting of an input Variac from the AC line feeding a stepdown transformer to a bridge rectifier to supply the LEDs that there would be no flickering and you could get a total dimming range from 0 to 100%.


What should you do?

Try replacing the Lutron dimmer with a Variac to power your existing driver. Does the problem exist?

Very likely the problem is a combination of the Lutron dimmer with the driver.

.
 
Gar,

I believe your possibility #2 is the correct one. The drivers from this manufacturer are definitely magnetic ones with a rectified DC output. The dimmer is the one recommended by the manufacturer and also the one I have been using with their drivers for years. I don't think it's a question of incompatibility. It would be nice to have a variac, but there are other expenses I have to put first.

I have been installing this brand of LED tape for about four years and they have been, by far, the most stable when it comes to flicker. I have never had flicker at any other installation locations.
 
171122-1658 EST

PetrosA:

I believe your LED strip is simply individual LED chips and resistors. This is not a linear resistive load, but a very predictable current vs voltage curve with virtually no hysteresis. Because it has current limiting resistors within the strip that contribute about 1/2 the voltage drop the curve is not as nonlinear as just the LEDs alone would be.

If your driver is simply an ordinary iron core 60 Hz transformer with an output bridge rectifier, then this alone in combination with the LED strip is not the problem. I can apply any AC waveform into this driver that does not have a DC component, does not over-drive the core, and does not over-power any component, and there will be no flicker other than related to the base 60 Hz waveform. Further assuming there is a constant steady state waveform into the transformer primary.

If the above is what you have, then the dimmer is the problem.

If you have some transformers with 2 and 4 to 1 ratios, then you can create some specific lower voltages that produce a sine wave, even without a Variac. Also you can use various combinations of incandescent bulbs or resistors as a series impedance to the LED driver.

.
 
All he could really offer in my case was that their product is required (like most others) by the FCC to accept any outside interference.

More precisely their product is required to accept interference from any licensed service or product or any unlicensed service or product conforming to the applicable rules for emissions.
And, of course, there is nothing preventing them from designing their product such that is it not susceptible to such interference. The FCC rules do not give their designers an excuse for shoddy work.
 
171122-2047 EST

PetrosA:

Following is some data from a 12 V DC 20 W LED light strip.

Logic and some knowledge of LEDs tells you there will be no flicker at any steady DC current within the power rating of the strip. Basically the same conclusion can be drawn if the excitation is from a full wave rectified 60 Hz sine wave.

For my light strip:

The DC data is:
09 V 0.6 A 6.4 W not quite fully dimmed
10 V 0.9 A 9.0 W
11 V 1.3 A 14.3 W
12 V 1.7 A 20.4 W
13 V 2.0 A 26 W

The full wave rectified data is: (ave DC V, line AC V, line I, input power, input VA)
4.8 V 53.4 V 0.03 A 1.6 W 2.0 VA just glows
5.0 V 56.1 V 0.04 A 1.7 W 2.2 VA
5.5 V 62.0 V 0.04 A 2.5 W 3.0 VA
6.0 V 67.9 V 0.06 A 3.7 W 4.2 VA
7.0 V 79.1 V 0.09 A 7.0 W 7.8 VA
8.0 V 90.3 V 0.14 A 11.7 W 19.0 VA
9.0 V 101.5 V 0.18 A 17.7 W 19.0 VA
10.0 V 112.8 V 0.23 A 24.9 W 26.9 VA
DC voltage measured with a Fluke 27, other measurements using a Kill-A-Watt EZ at transformer input.

I do not have a phase shift dimmer to perform the same test.

It may not be real feasible to build an inexpensive phase shift dimmer with near zero average DC output current that will be stable with all loads and various dimming settings.

Note: the on time of the current pulse with full wave excitation is moderately short because of the moderately high threshold voltage of the LED strip when adequate current starts to flow. 1 to 2 milliseconds of 8 mS. However, at low dimming levels because of this thresholding there can be light flicker from source line voltage variations.

A much better dimmer design would be to build a filtered DC power supply, and on its output use a a fast pulsing on-off modulation of the DC supply. This can provide 0 to 100% dimming range.

Further, note, that LED illumination sources are built with a UV LED that excites a phosphor to generate visible light. The UV LED has a very fast response and decay time. While phosphors have relative long decay times. This tends to reduce the magnitude of 60 or 120 Hz visible ripple.

.
 
If you want to have less problems with DC Led tape, don't use line side dimmers. Use a DC load side dimmer that's meant for 12-24 volts. They work wonders and I've never had a problem with flickering with them. I have however had many problems like you have described with line side dimmers.


Sent from my iPhone using Tapatalk
 
Peak repetitive inrush current ?

Peak repetitive inrush current ?

The dimmer you provided- when plugged in at Lutron, site says not recommended for LED loads?

In some cases the peak repetitive inrush current of an LED driver requires an even greater de-rating of the dimmer. Especially if the dimmer is not designed for that type of load.
 
171123-1111 EST

ELA:

I believe the circuit PetrosA is working with is:

(1) An LED strip light designed to work from a 12 V DC supply (a car battery, or equivalent, also a rectified AC voltage). This LED strip consists of LED chips and series resistors, and probably no reverse polarity protective diode. The IV curve would look like the data I measured above. Never will this load exhibit a peak inrush current.

(2) Apparently there is a driver consisting of a transformer with an output bridge rectifier. No capacitors mentioned. So this would look like a series impedance of inductance and resistance.

If there is a filter capacitor, then it looks like any other transformer, rectifier, and capacitor input filter looks like. There are current pulses, but somewhat moderated by the series inductance and resistance. Nothing here to create dynamic instability.

When an LED is driven from a rectified AC voltage, then there are current spikes sort of like the spikes into a transformer. rectifier, and capacitor power supply.

Much larger current spikes occur for a bridge rectifier connected directly to a capacitor. These rather large current spikes exist near the voltage peak every half cycle because there is little series impedance.

(3) I believe the dimmer is the problem and that under certain loading conditions that it has poor dynamic stability. But this is not a peak inrush current problem.

.
 
Last edited:
171123-2146 EST

strawburner:

The dimmer you referenced is a Lutron MRF2-6ND-120 is slightly better described at
http://www.lutron.com/TechnicalDocumentLibrary/369-143.pdf
This is what I call a 3 terminal dimmer, meaning a neutral is required. This type of dimmer can provide a fairly consistent firing angle because power for the electronics is derived from the hot input and neutral rather than using current thru the load for power to the electronics. Also provides good phase information.

What Lutron does not discuss is what the average DC output current is that adds to the AC output.

You are possibly correct that this dimmer would solve the flicker problem.

.
 
It's been a long time since I haunted this forum :D

I learned something today I think might be helpful to some of you at some point. About a month ago I install led tape in a kitchen. Two 36" pieces under two cabinets and two 39" pieces vertically in a cabinet along a different wall. It's DC tape with a magnetic driver that has a DC output (all parts are from GM Lighting) and is controlled by a Lutron AYLV600P. It was flickering badly at almost any point in the dimmed down range for the homeowner and she called me back. I confirmed the flicker and started troubleshooting to make sure that the load was in spec and that all connections were good. I confirmed a 56W load based on 4.4W/ft of tape and called tech support. They told me, however, that the 60W driver is actually only good for 80% loading when using a dimmer. I disconnected the two vertical pieces to confirm and sure enough the flickering stopped so I ordered a 100W driver to replace it. I installed that yesterday, came out of the attic, turned the lights on and... sonofa... They still flickered. Even stranger, when I again disconnected the in-cabinet lights the flickering stopped again. At that point I tried calling tech support but got voicemail. I kept trying what I could - checked input and output voltage, output frequency, tried a different dimmer etc but no luck. The flicker wouldn't go away. In the meantime a friend of mine called and I gave him a rundown on the situation peppered with lots of four letter words just to vent. At one point I said, "it's as though I'm *$%#* picking up %&#$@$ interference through the @^*&#$ LED tape when they're all connected." He shot back with "Dude (he's from California), I'll bet your tape is forming a dipole antenna," and suggested that I try and hook up the vertical tape to one driver and the horizontal tape to another driver. Sure enough, they all work perfectly now without a hint of flicker. It cost me a few hours of troubleshooting and a second driver on my dime but the problem is fixed now that the vertical and horizontal runs are isolated. In the future I will be pricing every cabinet job with two drivers just in case.

Obviously, I don't have RF test equipment to determine what kind of interference I was dealing with, but my first guess would be an air handler in the attic not far from the kitchen. It could have a noisy circuit board or transformer. I thought about shutting circuits off to see if I could isolate a culprit, but since the homeowner wasn't there and I didin't know if they had anything plugged in that might get messed up from loss of power, I didn't try that.

I eventually got in touch with the tech support guy later on and asked him if they've had feedback from other installers about the tape acting as an antenna or if using the aluminum channel would help prevent this but he admitted that he hadn't gotten any feedback about outside interference and didn't know whether the aluminum channel would act as shielding. They have had complaints about the LEDs coming on when the driver is installed near a motor, like under the sink near the disposal. All he could really offer in my case was that their product is required (like most others) by the FCC to accept any outside interference.

View attachment 18990
I've had the same issues with the tape light and the ariadni dimmers. Even though they are LED dimmers they are horrible and don't always work right. Try a diva dimmer I find they have been working they best

Sent from my SM-G955U using Tapatalk
 
Welcome back Pete

I have never seen an led tape that was allowed to have more than 16' in any run. If you have more than 16' then you need 2 runs from the trany. I am not sure what you have there
 
171125-1254 EST

What is an LED tape strip? I believe this can be defined as:

(1) a dc unidirectional load that must have the correct polarity applied.

(2) consists of a + and a - bus bar running the entire length of the the strip. Shorter strips can be connected in series, but the maximum series length is determined by the current carrying capacity of the bus bars. This length will be dependent upon brightness (power per foot) of the strip.

(3) for each LED chip in the strip there is a series resistor that provides some degree of current limiting, but not very good.

(4) several (like 3 or 4) LED chips and their associated resistors are connected in series and that series combination is connected between the + and - bus bars. The 3 or 4 is about correct for a 12 VDC source. This is a balance between wasted power in the resistors, and an attempt to work toward a current source to the LEDs.

(5) there is nothing inherent within this strip that will cause flashing.

(6) by associating a resistor with each LED chip power dissipation in the strip is more uniformly distributed than if one resistor was used with each 3 or 4 LED chips.

.
 
why does a LED strip need ± bars? some stuff is built for AC or DC operation where diodes can be downstream on the strip, or even built into the LED itself (hah, LEDs are diodes).

"dimming" a diode really needs PWM.
 
171127-1150 EST

With an LED light strip you must provide the correct input polarity.

If the wrong polarity is applied, then:

(1) the LEDs won't light.
and/or
(2) the PIV rating of the LEDs might be exceeded.

At the LED strip input you could build in a bridge rectifier, but that would add two diode drops. Bad idea at the 12 V level.

.
 
Obviously, I don't have RF test equipment to determine what kind of interference I was dealing with
Not knowing much about the physics involved, your idea that a vertical and horizontal tape together produce an antenna that picks up interference is plausible.

However, looking at the configurations that you list testing, it seems like your data is also consistent with the theory that tech support was wrong, and that the drivers and dimmers you are using are only good for about 50% of nominal driver rating when used together. If you were to test just one horizontal tape and one vertical tape together on one driver, and see the flickering upon dimming, then that would be good evidence that the issue is the physical orientation of tapes on a single driver, and hence some sort of antenna/interference issue.

Cheers, Wayne
 
171127-1422 EST

wwhitney:

Have you ever worked with radio transmitters? Do you know how power density varies as you move away from an RF radiating source? Is RF power of a low level going to affect a simple DC power supply? Is low level RF power coupled into LED strings going to have sufficient energy to propagate thru the DC supply and get into the dimmer and affect the dimmer? How close to the LED light strings is any high power RF transmitter? Do bed springs in the house talk?

The LED strings acting as an antenna are not a plausible
cause of the flickering problem.

With my 25 W 2 meter transmitter connected to a 1/4 wave antenna I had a tough time to get partial excitation of a circular fluorescent tube within a couple inches of the antenna.

It makes no sense to suggest or propagate ideas that have virtually no likelihood of being the cause when there are much better explanations.

The problem is dimmer design with respect to the load.

The people using and working with dimmers need to understand how they work, and why the devices may not work well.

.
 
The LED strings acting as an antenna are not a plausible
cause of the flickering problem.
OK, I'll accept your statement to that regard. The point of my post was that a simple test would distinguish the two proposed causes.

Cheers, Wayne

P.S. (Edit) If that was already pointed out, I missed it.
 
Status
Not open for further replies.
Top