Electronics > Projects, Designs, and Technical Stuff
Assistance with LED Driver pulsing lights when load placed on same circuit
(1/1)
neo8222:
Hi all, I was hoping I could get some help with some overhead LED work lights I have that have a strange issue I would like to either fix or mitigate. The drivers are "Flicker free or low ripple" in the since that the they don't pulse at a frequency that would cause problems with cameras. If I run the high frame rate recording from my phone at 960fps only then can the light pulses be seen. This normally wouldn't be an issue but there is a noticeable flash when my window AC kicks on and recently when I got my cheapo hot air station which regulates the temperature of the air by pulsing the heating element and this causes a constant pulsing in the light output of the work lights. I don't mean a minor pulse, i mean like someones turning the light switch on and off really fast. I tried a few other high power draw appliances and have discovered that any load placed onto the same circuit that the lights are plugged into will disrupt the light output. If I turn on a heat gun for instance the lights will blink once but recover since the heat gun is mostly constant power draw while heating up. This tells me that the problem is with regulation of the LED driver and the voltage drop from loads on the circuit are causing an interruption in output (Please correct me if I'm wrong). The problem is I'm not entirely sure how to fix it. I've posted a picture of the driver (both sides, with some major components labeled) and would like to know if anyone has any input on a way to improve the driver. Let me know if specific close up pictures or part numbers of items on the board are needed. I've already tried some monster caps on the DC side of the circuit just to see what would happen but there was no change in the light output.
tpowell1830:
This is related to a recent EEVBlog video, perhaps?



There is a second video as well:



Hope this helps...
neo8222:
I appreciate the reply but these videos don't seem to address the problem with my LED drivers. I did some more investigating on the driver board and have found there are 10 3VDC LEDs wired in parallel and then 24 of these 10 LED circuits are series together and the driver is supplying ~74VDC to the LEDs using a controller and N channel mosfet. The driver I have is running the mains AC through a bridge rectifier to provide 120VDC direct to the LED string positive, the LED negative is then routed to the drain on the mosfet and the controller is then pulsing the gate to convert from 120VDC to 74VDC in a DC to DC converter setup. The problem I'm running into now is that the little 6 pin chip doing the magic for the mosfet is not labeled with a part number and the flicker I'm seeing is coming from the controller failing to maintain stable voltage out of the mosfet when a load is placed on the same circuit that the lights are on.

I'm not opposed to replacing the driver and mounting it to the top of the light as they are hanging from a drop ceiling. If I were able to calculate the power requirements on the LED strings with bench supply and replace the constant voltage driver with a constant current driver would that prevent rapid changes in load on the AC circuit from causing flickering in the output?
tpowell1830:

--- Quote from: neo8222 on October 19, 2019, 08:49:59 am ---I appreciate the reply but these videos don't seem to address the problem with my LED drivers. I did some more investigating on the driver board and have found there are 10 3VDC LEDs wired in parallel and then 24 of these 10 LED circuits are series together and the driver is supplying ~74VDC to the LEDs using a controller and N channel mosfet. The driver I have is running the mains AC through a bridge rectifier to provide 120VDC direct to the LED string positive, the LED negative is then routed to the drain on the mosfet and the controller is then pulsing the gate to convert from 120VDC to 74VDC in a DC to DC converter setup. The problem I'm running into now is that the little 6 pin chip doing the magic for the mosfet is not labeled with a part number and the flicker I'm seeing is coming from the controller failing to maintain stable voltage out of the mosfet when a load is placed on the same circuit that the lights are on.

I'm not opposed to replacing the driver and mounting it to the top of the light as they are hanging from a drop ceiling. If I were able to calculate the power requirements on the LED strings with bench supply and replace the constant voltage driver with a constant current driver would that prevent rapid changes in load on the AC circuit from causing flickering in the output?

--- End quote ---
First of all, welcome to the forum, I see you only have 2 posts. There are some very knowledgeable people on this forum.

I am a little confused regarding your comment "the flicker I'm seeing is coming from the controller failing to maintain stable voltage out of the mosfet when a load is placed on the same circuit that the lights are on.". Do you mean the same AC circuit? This unit should not in any way load the AC line coming in and vice-versus, another load on the AC line should not affect this circuit at all. If this is what you are talking about, you may have other problems on that AC line.

A step back may be in order. If you have an o'scope, maybe a before and after shot of the waveform that you see from the output of this lighting circuit before and after the change condition is encountered. Also, if you could clarify your statement, it would be helpful. My first thought after you described this problem in the quote above was that it sounds exactly like what the videos are discussing, a lot of ripple on the output. A current measurement would also help as well as a reverse traced circuit diagram (see BigClivedotcom YT channel), this should be easy enough with that circuit/PCB, and some close shots of the entire board, front and back.

You may just have a poorly designed power supply, however,  a constant current power supply may be better for LED light regulation. This type of problem is very difficult to examine over a forum, especially without the waveforms, measurements, etc. to see what actually is going on. I gave you the links to the videos for seeing the process that Dave used for determining his problem on his power supply, thought it may help.

Hope this helps...
thinkfat:
The LED driver probably has insufficient line regulation. And your AC and hot air station are not very friendly towards the mains circuit. I probably have a similar hot air station around, the ubiquitous "858D" with its many clones. It sucks. When it reaches temperature and starts pulsing the heater, all lights on the same mains circuit flicker. You could try increasing the filter caps on the primary side (after the rectifier), but this has its limits, of course, and will quite likely impact the power factor. You can also rewire your house so that light and power outlets are on different mains lines (only works if customary in your country to have 3-phase mains).

EDIT: looked at your photos a bit closer now, this looks like a very, very basic driver circuit. No secondary feedback path, very little primary side filtering, just a common-mode choke and two 100nF caps. A single switching transistor and a discrete bridge rectifier together with a big capacitor as output filter. Very similar to the one in Daves' video. You should probably get a better LED driver.
Navigation
Message Index
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod