Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

LED Video Light Dimmer Not Dimming - Help Me Decode Transistor Names?

Status
Not open for further replies.

EditorInChief

New Member
I have **broken link removed** - a fairly low quality LED light for video. The build quality I would rate as 'indifferent' - but that's another story.

After taking it to a location and back to the office, it quit working, so I tested the leads from the power supply unit that go to the light head. In so doing, I shorted between two of the leads with my probe (yes, I am a dolt.)

Magically, the unit started working again, but now it doesn't dim. Strangely, the 0-100% LED readout on the front of the light still changes from 0-100% as I spin the knob, the light just doesn't dim. So I'm guessing that the pot is fine, but maybe one of the two larger transistors is fried (by yours truly.) I've attached photos of the whole situation and especially the two transistors, which I assume I should just try to buy online and then replace, but I want to make sure to get the right transistor.

One of them seems to be "irf1018e IUR p216d 72RA" and the other seems to start with a snaky squiggle and then have "JM02RP M25701" and then "-12 P+" which I assume refers to polarity.

Can anyone help me find where to order new transistors?
 

Attachments

  • IMG_1434.JPG
    IMG_1434.JPG
    471.6 KB · Views: 324
  • IMG_1435.JPG
    IMG_1435.JPG
    612.5 KB · Views: 378
  • IMG_1436.JPG
    IMG_1436.JPG
    670.1 KB · Views: 312
  • IMG_1437.JPG
    IMG_1437.JPG
    363.5 KB · Views: 336
  • IMG_1438.JPG
    IMG_1438.JPG
    341.3 KB · Views: 298
hi,
They are IRF1018 MOSFET and a LM2570 Switcher IC.
E
 
"irf1018e is the 7.1mΩ Nch MOSFET switch.

The large DC converter may be 36Vdc 3A .. check ratings, model #?

If there is no voltage, measure resistance between Drain and Source should not be shorted out.

Of course it's driver could also be toast among other things.

upload_2015-9-21_10-49-52.png


The large A rating is meaningless here, but the low Rds(On) allows a tiny heatsink.
 
Wow, thanks, that was fast!

Do you order just one thing from alibaba? Should I use those links or just buy the same thing somewhere domestically?

The light is turning on and the fans work, so I think the driver is working normally, Tony. Does it make sense that a faulty capacitor could be causing the dimmer not to work, but still letting some voltage through?
 
ASk your Ohm meter to do some measurements, answer my prev. ??
Quite right, Tony, I missed that. Did your test and got no resistance between drain and source on either component. Either with the ohm meter which I haven't used much (yep, I'm clueless) or with the beep test. So it seems these need to be replaced, have to hope that nothing else is fried as well. I'm just assuming these are the main ways of controlling voltage since they are big and look like transistors to me, but I have no real way of knowing.

I attached photos just to dummy check.

THANKS.
 

Attachments

  • 00000.MTS.21_59_29_27.Still001.jpg
    00000.MTS.21_59_29_27.Still001.jpg
    877.3 KB · Views: 380
  • 00000.MTS.22_00_12_14.Still002.jpg
    00000.MTS.22_00_12_14.Still002.jpg
    928.5 KB · Views: 362
Normally use x1 R scale then short probes and adjust red thumbwheel to 0 Ohms

But looks like a short.


If you unsolder FET, LED will go off and then maybe you can see Gate pulses with AC mode meter or scope. Usually Vgs > 5V to turn on FET more is lower Rds ) or low V is OFF

What PSU is used? in unit?
 
Okay, I did the x1 scale and zeroed it out, and same deal.

The PSU doesn't have any identifying information. It sends 37 volts, which makes sense, i think many LED arrays are constant voltage around 35 and then they are dimmed by varying the amperage. I think you get color shift if you alter voltage. I can't tell you how many amps because I blew the fuse on my tester trying to find out. I guess that what they make the separate 10A thingy for.

I found one of the transistors on digikey, but not the other (LM2570 Switcher IC) should I get it somewhere else or do I just not know how to search for it?
https://www.digikey.com/product-detail/en/IRF1018EPBF/IRF1018EPBF-ND/1894138 is the one.
 
Well, for some reason, I couldn't get the interface to work with fewer than 1000 items (had to get the total price over $1) so apparently I'm going to have 999 left over transistors in case anyone needs some.
 
hi Ed,
It is possible that last digit could be an '8', please look and let me know. ie: LM2578
E
 

Attachments

  • IMG_1438.JPG.jpeg
    IMG_1438.JPG.jpeg
    80.6 KB · Views: 242
It is possible that last digit could be an '8', please look and let me know. ie: LM2578

Wow, I'm glad you made me look at that again. It's actually a '6' - LM2576. And on closer inspection, it has 5 leads, not the 3 I saw - it's got a dab of hot glue or something that makes it difficult to see.

On digikey, there are a bunch of LM2576 components - https://www.digikey.com/product-sea...e=0&rohs=0&quantity=&ptm=0&fid=0&pageSize=500

Do you know how to select the right one? Does the line of text above LM2576 - "JM02RP" - mean anything to you? It isn't listed on digikey, maybe manufacturer? Also, I thought it only had three leads, so I didn't test it correctly and maybe it isn't fried and I could just replace the other.
 
hi Ed.
Just use the LM2576 ident for searching, make sure the one you buy has the same package type ie: basically the same shape and pin configuration.

I have traced the LM2576T and it comes in two types ADJustable and fixed.!!!!


Check your board for the 'T' suffix on the LM2576.

E

EDIT:
Reading your post #9, ie: 37V would suggest that the device is a LM2576T ADJ

The fact that you measure 37V would indicate that the LM2576T is OK.???
 

Attachments

  • datasheet.pdf
    576.9 KB · Views: 333
Last edited:
Thanks for that, Eric.

I am starting to think that one is okay since I wasn't seeing two of the pins that were buried in hot glue on the board and tested incorrectly. I thought it was three pin, but it's five. Testing first and second pins now, Vin and Output, doesn't seem to short. The 37 volts is coming out of the silver PSU at the right in the second picture on the first post. I think that whole unit is fine.

Here's my guess (and I don't pretend to know what I'm doing with electronics, but...) - the board has to do two things - dim the light variably from the main voltage, and run the fan, which would run at a lower steady voltage. This LM2576 unit drops the voltage for the cooling fans, and the other mosfet, the 1018, is connected to the (potentiometer?) knob which dims the main light. I think that knob is fine because the digital readout still changes from 0-100%, just the light doesn't dim.

So I replace the shorted 1018, and bob's my uncle, the thing works again, right? Haha. Buying parts will let you know.
 
For a 100Watt LED array running at 37V , the current should not exceed, 2.7A
FET ON resistance is rated at 8.4mΩ which results in 23 mW. So it seems unlikely to have overheated unless a short circuit was imposed.

If the White LED turns blueish, it is overheating and being over driven.
 
So it seems unlikely to have overheated unless a short circuit was imposed.

Tony,
As quoted, I suspect the MOSFET has died, due to Ed's shorting.
After that the LED's run at an uncontrollable maximum intensity.
E
Ed said: After taking it to a location and back to the office, it quit working, so I tested the leads from the power supply unit that go to the light head. In so doing, I shorted between two of the leads with my probe
 
I got the MOSFETs delivered (i had to spend a few days in austin) and am trying to get this straightened out. My plan was to clip off the legs of the old burnt out one and solder the new one onto it so I could avoid direct soldering onto the board and overheating, etc.

I am bad at soldering.

Unfortunately, in trying to clip off the leads, I broke one of the supporting solder joints, which seems to meet right up with what I think is a tiny resistor. I've attached a picture of the board with the broken MOSFET lead as well as the leg I clipped and broke off. I think it makes sense to solder that leg back on since the joint will match up neatly and then solder my new MOSFET on top like it has stilts. But if anyone has any wisdom, I'd take it, as I'm not very experienced at soldering and I don't always get things perfect right away. I can't quite see the lead on the end of that resistor, and it makes me wonder what exactly I'm trying to connect to.
 

Attachments

  • SOLDER1.jpg
    SOLDER1.jpg
    1.2 MB · Views: 263
  • solder2.jpg
    solder2.jpg
    441.7 KB · Views: 308
One other thing - I assume that thing labeled '200' is a resistor because 'R9' is printed on the board next to it... so if it's a 200 micro-ohm resistor, I could just connect further back up the circuit rather than trying to solder onto it maybe?
 
hi Ed,
It will not be 200micro-ohm, it looks like a 200 ohm to me.
With an Ohmmeter it will still be possible to measure the resistance by placing the probes on the ends of the resistor.
E
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top