What's the best way to move metal trays with motors

[Spork Schivago]

Hello,

I'm building a UV LED Exposure unit to help me make better traces on PCBs. I have 6 PCBs (2 x 3) for the top exposure and 6 PCBs for the bottom exposure. I was thinking of mounting these PCBs on two sheets of aluminum. The boards are 100mm x 100mm each, so when hooked together, the top and bottom exposure PCBs would measure 600mm x 200mm.

I was thinking of using a sheet of aluminum that gave me an inch (25.4mm) on each side of the PCBs. So I'd have two sheets of aluminum, measuring roughly 650.8mm x 250.8mm. Then I was thinking of drilling some holes on the corners of the aluminum trays and putting long rods through them. I than wanted to hook a belt somehow to the trays and use two motors, one for each tray, to move the trays up and down so I could control how close they are to the PCBs I'm developing.

Does anyone have any suggestions on what type of motors I could use? Would I need to use stepper motors? It'd be nice if there was some sort of sensor that would stop the motors if the boards got to close together or too far away and where about to crash into the enclosure. I think for that though, I'd probably need stepper motors, right?

Also, if anyone has any better suggestions on how I can make these boards adjustable, I'm done for entertaining them. I thought two motors with two momentary switch (on-off-on) would be best. One switch would control the bottom board, one would control the top board. If you push down on the switch, the board would move down. If you pushed up, the board would move up. Maybe it'd be best if I just had one momentary switch to control both trays. Down would move both boards further apart from each other, up would move them closer together. I'm just not sure how to implement the one button setup. What do you guys think?

[ben7]

Re: What's the best way to move metal trays with motors

Hm, kinda difficult, but check out robotics stuff. A lot of motors and mechanical things are used in robots, so you could find some stuff in that department I'd think.

But, from what I can gather .... why not dim the LEDs, instead of moving them?

-Ben

[diif]

Re: What's the best way to move metal trays with motors

Why does it need to be adjustable ?
Just needs to be a box with some LEDs in and a switch/timer.

http://www.instructables.com/id/UV-LED-Exposure-Box/

http://www.instructables.com/id/Ardu...-Exposure-Box/

[mariushm]

Re: What's the best way to move metal trays with motors

I don't understand why you'd need to adjust the distance between the LEDs and your PCB. You don't say why.

Anyway, don't picture your device just as three layers stacked on top of another. For example, try to imagine your pcb on its side locked in the center of your device in some kind of fixture. and the two side walls of your box are with your uv leds.
Now you can have guide channels on the bottom and top of your box on which the two side walls can slide left and right, closer or further from the pcb fixed in the center.
From this point, you can use whatever mechanism you want to control how the two side walls are moved ... i can think of stepper motors twisting those bars with twists on them, stepper motor with wheels with lots of teeth on a strip glued to the bottom... etc

Anyway, you should pay more attention to the led choice, the wavelength is quite important - not all uv leds work in the optimum waelength to work on pcbs. The view angle would also be important, you don't want 15 or 30 degree leds that would act like spotlight, very bright in the center and less bright a few mm around the leds, it would result in non-uniform lightning.
You should experiment with adding a light diffuser (find some broken lcd display and take out the diffuser sheets from the panel), but I'm not sure how well UV goes through those.

See also http://www.electricstuff.co.uk/pcbs.html , may be helpful.

[Spork Schivago]

Re: What's the best way to move metal trays with motors

Sorry for not being clear enough. I ordered a kit that came with the LEDs. Supposedly, they're 160 degree viewing angle, which I don't really think is correct. I don't have the specs on them and the person selling the kit says you have to adjust it until you get the right distance from the PCB that you're designing, you know, a trial and error type thing. Also, over time, I'm under the impression the LEDs will become dimmer. Is this incorrect? I thought it'd be nice instead of having to replace them when they get a bit dimmer, I could just move them closer. If it's not needed though, I won't do the extra work.

The person selling the kit says: The specs of the leds are: 395nm wavelength, 1100mcd, 160°
but won't provide a datasheet on them. It was a 155$ USDs for 12 PCBs and 32 UV LEDs per board, plus 8 150ohm resistors per board. I think maybe the reason he doesn't want to provide the datasheet is because maybe he makes a lot of money on selling them.

I thought about adding a diffuser. I debated it. The pros is it'd give a nice uniform exposure, the downside would be it'd block some of the UV, depending on what I was using.

This link has the case he built. I kinda liked the design and was going to go for something similar I thought. I'm not really artistic at all and if I tried designing something from scratch like that, I'd just be a square box (and not a very well built one!)
http://en.elektronicastynus.be/projects/81/

[stj]

Re: What's the best way to move metal trays with motors

i was wondering about the spread.

the angle will dictate the distance but also how much the led's are spaced apart on the board so they neither overlap or have dark area's between them.

you could use 4 long bolts - one on each corner and raise or lower the upper section by turning the nuts.
you dont need a motor(s) because once you know the distance, it's not going to change.

[stj]

Re: What's the best way to move metal trays with motors

btw, i dont know the total size of this thing, but building it into an old flatbed scanner case would solve a lot, you have a sheet of glass, and a lid!

[mariushm]

Re: What's the best way to move metal trays with motors

I didn't do extensive research but it seems the optimum wavelength would be around 360-380nm

Your seller probably sells you LEDs with a PEAK output of 1100mcd , as in when you run the led for a few milliseconds at a current a few times higher than normal - for example see this VCC LED that's usually designed for 30mA but can run at 100mA with a 0.1/1kHz duty cycle (i guess 0.1 ms every second, like a flash)

If I were to make something, I'd buy an old flatbed scanner and replace the light with a strip made with leds like these : http://uk.farnell.com/vcc-visual-com...mcd/dp/1712756 and when i want to process a pcb, i'd just hit scan document at 600-1200dpi several times (or hit batch scan to automate it)

Alternatively, I'd just fill the surface of the glass with leds like these, as they're cheap and tiny yet still easy to solder: http://uk.farnell.com/vishay-semicon...c-2/dp/2395945

But I admit, didn't do any research on this subject so I don't know how well the whole thing would work..

[stj]

Re: What's the best way to move metal trays with motors

it would take more than a few "scans" or you would have to handle the pcb's in a darkroom!

you could automate the scanner motor-drive with a microcontroller to just go forward and back for a programable length of time though.

[Spork Schivago]

Re: What's the best way to move metal trays with motors

it would take more than a few "scans" or you would have to handle the pcb's in a darkroom!

you could automate the scanner motor-drive with a microcontroller to just go forward and back for a programable length of time though.

The UV light doesn't have to be constant? I figured if I underexposed a board, I wouldn't be able to put it back in and turn the UV LEDs on again. Figured it was some sort fo chemical reaction with the photoresist that was causing the picture to develop. I figured the UV LEDs where the catalyst.

[Spork Schivago]

Re: What's the best way to move metal trays with motors

btw, i dont know the total size of this thing, but building it into an old flatbed scanner case would solve a lot, you have a sheet of glass, and a lid!

The two boards are each 600mm x 200mm. I didn't want to use the flat bed scanner idea. I've seen it, and although it looks neat, I kinda wanted to work on building my own case. Once I get it built the way I want out of wood, I think I'm going to buy some 6061 T6 Aluminum and have it bent and milled and finished to look real nice.

The case that the original guy did opens up like a drawer. I was thinking maybe having it open up like a cupboard instead. It'd have a glass window so you could see inside and it'd have some shelf you could pull out to put your boards in and take them out. It'd look like one of those mini-refrigerators, a little smaller though, with a glass window! That might look kinda cool.

[stj]

Re: What's the best way to move metal trays with motors

sort-of, but it is down to total exposure time.
not sure about putting them back in, you probably couldnt line up the mask again.

btw, you know that UV is BAD right! you need a safety switch and no light leakage.
ideally you should be able to turn out the light in the room and not see any blue glow at all.

[Spork Schivago]

Re: What's the best way to move metal trays with motors

I didn't do extensive research but it seems the optimum wavelength would be around 360-380nm

Your seller probably sells you LEDs with a PEAK output of 1100mcd , as in when you run the led for a few milliseconds at a current a few times higher than normal - for example see this VCC LED that's usually designed for 30mA but can run at 100mA with a 0.1/1kHz duty cycle (i guess 0.1 ms every second, like a flash)

If I were to make something, I'd buy an old flatbed scanner and replace the light with a strip made with leds like these : http://uk.farnell.com/vcc-visual-com...mcd/dp/1712756 and when i want to process a pcb, i'd just hit scan document at 600-1200dpi several times (or hit batch scan to automate it)

Alternatively, I'd just fill the surface of the glass with leds like these, as they're cheap and tiny yet still easy to solder: http://uk.farnell.com/vishay-semicon...c-2/dp/2395945

But I admit, didn't do any research on this subject so I don't know how well the whole thing would work..

Do you know of any cheap, easy ways to measure wavelength of the a light source? I almost remember seeing some lab where you used a laser and a ruler and measured some angles and used those angels to calculate it or something. Maybe that was just in a movie though. Can't seem to remember where I remember that from.

[ben7]

Re: What's the best way to move metal trays with motors

btw, i dont know the total size of this thing, but building it into an old flatbed scanner case would solve a lot, you have a sheet of glass, and a lid!

And, you'd have the mechanical assembly that moves the scanner thing back and forth! A good idea! :P

-Ben

[stj]

Re: What's the best way to move metal trays with motors

Do you know of any cheap, easy ways to measure wavelength of the a light source? I almost remember seeing some lab where you used a laser and a ruler and measured some angles and used those angels to calculate it or something. Maybe that was just in a movie though. Can't seem to remember where I remember that from.

a defraction grate.
not sure they can be used in the uv spectrum though.

[Spork Schivago]

Re: What's the best way to move metal trays with motors

I thought the diffraction grates were just for separating lights of different wavelengths, not for measuring the wavelengths. Thanks for letting me know about that Stj. I would've never thought of using one for measuring the wavelengths!