Random Thoughts - Part Deux [NO POLITICS]

[cjd]

Originally Posted by Ashikabi

That looks just like them. How would I test the specs on the existing bulbs? There are a ton of options at the bottom of the page to choose from

I'm not sure if you can test (I know enough to stop when I don't know any more...) you really need to see if there's a part number *anywhere* on the lens casting, or find a schematic that says what the parts are. Or just find somewhere someone's already done the work. Maybe one of the companies that does replacement LED parts would already have an easy way to check/tell...

What part is this? Easy to get to?

You should be able to read voltage across the leads at the very least, which is the most important part. Also check lead spacing, if it's not 0.20" these aren't quite the right part.

[Ashikabi]

Originally Posted by cjd

I'm not sure if you can test (I know enough to stop when I don't know any more...) you really need to see if there's a part number *anywhere* on the lens casting, or find a schematic that says what the parts are. Or just find somewhere someone's already done the work. Maybe one of the companies that does replacement LED parts would already have an easy way to check/tell...

What part is this? Easy to get to?

It's in an aftermarket headlight for a 350z. Retailer doesn't know anything. I'm confident those are the right bulbs but I can't find a place that sells them in small quantities

[Ultramaroon]

Originally Posted by Ashikabi

That looks just like them. How would I test the specs on the existing bulbs? There are a ton of options at the bottom of the page to choose from

Uh oh. You're gonna need a guide for that rabbit hole. Start by downloading the spec sheet. They don't look the same. Can you get a couple closeups of one component at different angles?

edit: nvm, I just clicked the image. Duh.

[Ashikabi]

Ultra might be right. I missed the nipple on the top

[cjd]

This might help.

mouser.com has a lot more they sell by single unit, and they have image search... I dug up a few possibilities in a minute, but they're probably still wrong.

closer?... http://www.mouser.com/Optoelectronic...-b1bc8Z1ymgz9n

But closer isn't close enough.

Hole spacing and voltage may be all you need to find a suitable replacement.

[Ultramaroon]

Originally Posted by Ashikabi

Ultra might be right. I missed the nipple on the top

I think it's functionally close. Headlight LED helps define it. The original component was designed to have the widest possible dispersal pattern. That's why it was molded into that spherical shape with a specific offset from the PCB.

I think we need to know the current at which those are driven.

edit: Shit, I think you were right. I studied your photo more closely. I thought there was no material except on either side of the sphere around the diode. That packaging is deceptively clear.

[Ashikabi]

Originally Posted by Ultramaroon

I think it's functionally close. Headlight LED helps define it. The original component was designed to have the widest possible dispersal pattern. That's why it was molded into that spherical shape with a specific offset from the PCB.

I think we need to know the current at which those are driven.

They sit inside a reflective bowl so I'm not too worried about the nipple. The for is more important. How would I go about testing their current? Especially when there are 4 posts?

[Ultramaroon]

Originally Posted by Ashikabi

They sit inside a reflective bowl so I'm not too worried about the nipple. The for is more important. How would I go about testing their current? Especially when there are 4 posts?

You'd have to disconnect posts 1 and 3, or 2 and 4. Then stick an ammeter in between one of the disconnected leads and its respective solder run. You will read something slightly less than the rated current of the device.

Assuming the packaging is standard, you can see the marking that identifies the #3 lead, like on the SunLEDs.

[Ultramaroon]

Also often overlooked with large semiconductor devices, these are just as sensitive to ESD as their microscopic counterparts. Improperly handling them will significantly shorten their life.

[Ashikabi]

Originally Posted by Ultramaroon

You'd have to disconnect posts 1 and 3, or 2 and 4. Then stick an ammeter in between one of the disconnected leads and its respective solder run. You will read something slightly less than the rated current of the device.

Assuming the packaging is standard, you can see the marking that identifies the #3 lead, like on the SunLEDs.

Well with the way they are attached I'm confident that won't be possible. Why disconnect 2 posts? I assume they need to be lit for this correct?

[cjd]

Originally Posted by Ashikabi

They sit inside a reflective bowl so I'm not too worried about the nipple. The for is more important. How would I go about testing their current? Especially when there are 4 posts?

The lens is important to getting the right dispersion, regardless. The 70° dispersion LEDs in my initial link have the 'nipple' and the 110° dispersion versions did not.

[Ultramaroon]

Originally Posted by Ashikabi

Well with the way they are attached I'm confident that won't be possible. Why disconnect 2 posts? I assume they need to be lit for this correct?

You have to desolder them to replace them anyhow. Just remove one, bend the leads and stick one back in place.

I can see by the schematic that posts 1 and 3 are electrically the same. Same goes for 2 and 4.

I assume mounting position and tolerance for mechanical shock are the main reason four leads are used in this application.

[Ashikabi]

OK so I sorted on Mouser by led emitters>standard led through hole>green, 7.6mm(caliper read 7.67but that could be tolerance), water clear. Found a handful of ones that look like it from Cree. They all have the same 30mA at 3.6V

[Ashikabi]

Originally Posted by Ultramaroon

You have to desolder them to replace them anyhow. Just remove one, bend the leads and stick one back in place.

I can see by the schematic that posts 1 and 3 are electrically the same. Same goes for 2 and 4.

I assume mounting position and tolerance for mechanical shock are the main reason four leads are used in this application.

That makes sense. Why bend 2 leads though?