UV LED PCB Lithography Exposure Unit

[chopnhack]

chopnhack submitted a new Showcase Item:

UV LED PCB Lithography Exposure Unit

Received some goodies in the mail from our brethren in the Far East today

I have never tried photolithography for producing PCB's, so I figured for a few bucks, I could give it a go. I used twenty 1w 395nM UV emitting LED's mounted on aluminum PCB which I then mounted with a heat sink epoxy to some aluminum strips. I think I will encase those in a wooden box to prevent stray UV rays from exiting the targeted area. I am waiting on my driver to arrive to do a test before building the box.

Read more about this showcase item here...

 

[chopnhack]

chopnhack updated UV LED PCB Lithography Exposure Unit

Driver came in.

I found it interesting that the drive was not a stable voltage source, it seemed to bounce around between 58vdc and 80v! Even when taxed with twenty 1 watt LED's is seemed capable of providing the power. I haven't inlined my multimeter yet to see just how many milliamps are running through the circuit. I have some more fabrication to work on, adding a power switch and a fuse as well as finishing the top.

View updates to this showcase item...

 

[Bluejets]

I tried it donkeys years ago but had anything but success.

Used a gang of 8w fluro tubes but found out later that the pre-coated board which came from Dick Smith could have been sitting on the shelf for a year or more. One would expose to the uv for a given time(experiment) and then spray with a developer.

At the time ( I found out later) all these items had a shelf life of around 1 to 2 months at best.

Did some repairs on Nail curing units a while back. They were 365-370nM...took a lot of care not to have the uv hit my eyeline.
I think it could do an amount of damage if not aware.

 

[(*steve*)]

A colleague of mine is working on a similar device. Based on the datasheets of the material he is using, the LEDs are a better fit to the sensitivity of the film than UV florescent tubes.

 

[chopnhack]

Did some repairs on Nail curing units a while back. They were 365-370nM...took a lot of care not to have the uv hit my eyeline.
I think it could do an amount of damage if not aware.

Wish me better luck!
I am going to use the diy dry film that gets laminated to the blank pcb. This sandwich is then exposed to the uv lights. Interestingly enough, the LED's I got are predominantly 390-395NM wavelength, barely UVA. Although longwave radiation of this type is harmful to the skin (causes aging) its the UVB short wave that causes cancer. Of course, there is no reason to needlessly expose yourself to this type of radiation, but sunglasses were used when I took that photo

A colleague of mine is working on a similar device.

Cool, how many LED's and of what type will he use? I had initially looked at the 3w, but they were too expensive for the qty desired. I also noted that as the wavelength got shorter the price went up tremendously.

 

[(*steve*)]

Cool, how many LED's and of what type will he use?

I will enquire and get back to you.

 

[Bluejets]

A colleague of mine is working on a similar device. Based on the datasheets of the material he is using, the LEDs are a better fit to the sensitivity of the film than UV florescent tubes.

Yes. With the nail cure there tends to be more mobs in on the act so-to-speak these days, all with differing requirements for cure wavelength. Hence more are producing the cure machines with fluros.

mmmm...wonder if we can find somewhere to use nail polish as glue, still have a few of those leds around. Costly also at around $5.00 each especially when one gets to the 365-370 range.

Good luck with that anyhow, chopnhack, like to see your results when you get there.
Results I get with toner transfer are iffy, sometimes it works, others not so good.

 

[chopnhack]

I will enquire and get back to you.

Thanks Steve, keen on knowing what wavelength he will get on with.

Good luck with that anyhow, chopnhack, like to see your results when you get there.

I will certainly post them, perhaps later this week if I get some time.... found my water tank to be leaking while I was in the shop - if it leaks it takes precedence!

 

[(*steve*)]

This is where my colleague obtained his LEDs.

He notes that the price has come down in the last 5 years

 

[chopnhack]

This is where my colleague obtained his LEDs.

He notes that the price has come down in the last 5 years

Yes, they have, but the 380 and lower wavelength are still pretty pricey as far as LED's go. The jump is amazing, that is why I went with the 390nm. I hope its not a bad decision!

 

[(*steve*)]

Have you checked the spectral sensitivity of your material?

 

[chopnhack]

Have you checked the spectral sensitivity of your material?

I wasn't aware of it at first. Most of the resources I had read stated to expose to UV light and didn't get into specifics. My source for the film was e-bay so no spec sheet. I guess it will be 'determined empirically' LOL

 

[Canobi]

I made this one out of an aluminium briefcase, a flatbed scanner and a fluro nail curing unit:



It was just a warm up exercise and to acquaint myself with the resist as I intend to make a second one using LEDs for something a little more low profile.

For practicality reasons, I'm putting the UV LEDs in the lid facing down and plan to use the underneath compartment as a regular lightbox (I got a small lightbox for post etch inspection and realised I could also use it for lining up and securing solder resist masks prior to exposing, work a treat) but figured I could combine the two, that way I won't have to risk anything shifting by moving it more than necessary.

 

[chopnhack]

I made this one out of an aluminium briefcase, a flatbed scanner and a fluro nail curing unit:

Wow! Really nice job Canobi! You put my scrap box to shame here . Have you used it yet? What was your time of exposure like?

using LEDs for something a little more low profile.

I wondered about height as well. I ended up going about 3" distance from LED to pcb. My thought on that was giving enough room for the beams to spread and produce a more uniform pattern across the face of the pcb. Too close and there may be spots with poorer coverage. No calculations done, just gut..

combine the two,

That is a great idea! Maybe I can get a smaller driver and wire in a separate switch, install the LED's on the side walls or something.

 

[Canobi]

Thanks chopnhack I've used it a fair bit actually, the nail unit had an inbuilt 2 min timer but exposure times are usually well under that.

 

[kellys_eye]

I've got a similar unit but using UV LED strips and installed in an old flatbed scanner case. The glass and lid are perfect for the job and the original external PSU is re-used for power.
I intend to fit a digital timer but that's a winter project on the back burner for now.