I think we need to know whether the LED+ is connected to drain or whether both are just connected to the positive rail. There could be some lower power mosfets on that board.

 

When seen through a digital camera, the clear diode doesn't emit light. I replaced it with another IR diode, which I know works, and it didn't emit any light again. At this point, I've given up on the safety interlock. We probably won't be able to find the fault with the mainboard.
Replaying to post #37: I also think that the LEDs are arranged in series in a group of 3 with a resistor, because when I short an LED(connect the + and -) the other two light up and the shorted one doesn't. This happens for all of the LEDs on all panels.
I found the specifications of the LEDs: http://www.smdledchip.com/sale-1159...5050-led-chips-uv-led-chip-for-nail-lamp.html

I just want the lamp to turn on when I plug it in. I want to remove the mainboard, solder the power supply to the big LED panel, and then the LEDs to start working. What should I do in order to achieve that?

 

Just connect the input power to the LED + and LED - as you proposed above. You can incorporate a switch in the positive lead to make it more convenient.

I looked up the chemistry of the photocuring gels and polish used. The UV light needed is relatively long wavelength (i.e., not UVC) as a photosensitizer (aka photoinitiator), such as benzophenone or related compound, is incorporated in the gel/polish. That photosensitizer is not destroyed, and the one potential problem mentioned is that the final product remains sensitive to too much UV, which can weaken and/or soften the coating.

Thus, running the UV LEDs at full power may reduce the time needed and make the exposure time more critical. Just timing with a watch should be sufficient.

 

Some digital cameras have an infra red filter fitted to them so they do not see infra red light. You can check if your camera can see infra red by pointing a known working TV remote control at it and pressing buttons on the remote control. If it can see infra red then you will see the IR LED on it flickering.
Before connecting the 12 volt supply directly to the LED panels we need to understand the apparent polarity reversal in the wiring. I only noticed this when I was checking to see if just bridging the source and drain of the mosfet would involve less work than swapping wires around. Can yo check the resistance readings between the red power supply wire where it is soldered to the control board and the red and the black wires where they are soldered to the LED panels. (Just testing to the connections on the large LED panel is enough as all the panels are connected in parallel.)

Les.

 

I believe the TS has already tested the UV panels using the polarity as marked.

 

I wonder if the TS unsoldered the wires to the LEDs at the control board end when doing that test ? They then may have been soldered back the wrong way round as the markings for polarity on the component side of the board are confusing. (See the second picture in post #1)

Les.

 

Well, anything is possible, but as I understand the TS, he/she is a student and just wants the lamp to light. Reversed polarity to an LED (within reason) isn't doing any harm. If plus to plus and minus to minus doesn't work, we can cross that bridge later.

A Sunone nail lamp is about $12 on Amazon w/Prime. Even cheaper on Alibaba. A BigMac meal is $6.00; a Whopper meal is $6.50, but that takes two hands. I think she (?) clearly indicated a desire for fresh nails. We are coming into the holidays, afterall, and if it doesn't last until graduation, so what?

I just don't want the cured nails to be overdone and sticky.

 

So connected the power inputs to the LED+ and LED-. Nothing happened, the lights didn't turn on. Could some of the LEDs be broken or could the paths between them be damaged?

 

Please explain: Well back in this thread, you said the UV LEDs all tested good. How did you test them?

If LED+ and LED- are mislabeled, as another member has suggested, that would also explain the failure, but it is inconsistent with what I assumed you did to test them.

 

I wonder if the TS unsoldered the wires to the LEDs at the control board end when doing that test ? They then may have been soldered back the wrong way round as the markings for polarity on the component side of the board are confusing. (See the second picture in post #1)

Les.

In that picture, it looks like there is a track connecting V+ to LED+, but the black wire to the LEDs is connected to that pad.
So MAYBE it needs the wires to the LEDs to be reversed where they connect to the board.

[Edit] Would reversing the 12V power to the LEDs damage them?

 

Well, I took a jump wire like the one in the picture and connected the + and - of a single LED. Then that LED didn't light up, but the other two in the series did. Then I connected the jump wire to the other two LEDs and the same thing happened: the connected one didn't light up, but the other two did

 

Well, I took a jump wire like the one in the picture and connected the + and - of a single LED. Then that LED didn't light up, but the other two in the series did. Then I connected the jump wire to the other two LEDs and the same thing happened: the connected one didn't light up, but the other two did

Was this with the LEDs connected to the control board?
Did you, at any point, unsolder the wires to the LEDS from the control board?

 

"Power inputs" I thought you meant the wires from the wall supply,

That is still strange, if the LEDs are in series, and the first one doesn't light, maybe that one is bad, but the other still turned on. Now, as I understand it, you are attaching the supply the same way and none are turning on.

Please clarify.

EDIT: Maybe they are reversed? Schematic being made now.

 

I also desoldered one of the panels and connected it to a 12V 6A 80W LED power supply. The LEDs didn't light up again

 

In this sketch, V+ and V- are the polarities of the leads you attached:



That would not light the top LED, if connected that way, but it would light the other two. That schematic makes no sense. If connected properly across the top LED, then only the top one would light. (edited)

At this point, words seem to be failing us. Please draw, scan/photograph, and post what you did.

Edited again.

 

I've connected the minus from the power supply to LED- and the plus from the supply to LED+. Here is a video to see that when it's connected like this, and when I connect the + and - of the LED, the other two light up: https://drive.google.com/file/d/15ajg_sCis460ysED6O6vj6N4o3PlAIjo/view?usp=sharing

 

Here is a video to see that when it's connected like this, and when I connect the + and - of the LED, the other two light up: https://drive.google.com/file/d/15ajg_sCis460ysED6O6vj6N4o3PlAIjo/view?usp=sharing

I can't see the video I get only a forever whirling circle.

 

Here's what I see in the video:


That is, when you short across one LED in a string, the other two in the same string light. Now, if you short across the middle LED, do the other two light?

Can you attach your voltmeter across the supply leads (LED+ and LED-). It should read about 12V. What does it read? When you then short across an LED and the other LEDs light, what does it then read?

EDIT: There is presumably a 39Ω resistor somewhere in that string. What happens if you BRIEFLY short across that resistor? If you have a small resistor (10 to 50 Ω) you should add that in series with your shorting lead.

 

I've connected the minus from the power supply to LED- and the plus from the supply to LED+. Here is a video to see that when it's connected like this, and when I connect the + and - of the LED, the other two light up: https://drive.google.com/file/d/15ajg_sCis460ysED6O6vj6N4o3PlAIjo/view?usp=sharing

Have you isolated the LEDs from the remainder of the circuit before testing. In the attached picture, the blue cloud represents the remainder of the circuitry. There may be a sneak path through the circuit that is affecting your tests. The red/blue line represents the trace that needs to be cut.

Have you run a test by connecting LED- to the power supply plus and LED+ to the minus of the power supply? Try at a lower voltage first.

In this picture, the top circuit is what you expect. Note the sneak path I have drawn. The bottom circuit is what you are observing. It represents what would happen if the LED+- connections are reversed and there is a sneak path.


UPDATE: You can cut the trace with a knife. After your tests are complete, it is a simple matter to scrape any coating off either side of the cut and repair it with a solder bridge.

 

That is, when you short across one LED in a string, the other two in the same string light.

Without one of them shorted, do any of them light?