UV light in scanner project

[deansmith]

Can anybody help me. Iwant to start making my own circuit boards with programs such as eagle. I have made an etching bath out of a small fish tank, and a UV light from a scanner. BUT the UV lighting dont work. I used 6 - 8W - 30cm UV tubes evenly spaced out in a flatbed scanner. I bought a 36 W flurecent tube complete with starter and ballast. I thought i could use the ballast and starter to fire up the UV tubes ,( wired parallel),nothing happened. I then tried just one 8w tube, and it just flickered at both ends. Thought the tube might be faulty ,so tried another it flickered too, and went out. Now it dont work at all. Someone told me i need a T8 starter and not a
T5. I borrowed one from a friends fish tank light. Said T8 on the light housing but had the same information on the starter as the other one i got from the
36W starter.I so much wanted to see them light up, i decided to try without a ballast. It worked for a split second ,now i only have 4 UV tubes.Sombody please help me

 

[Hero999]

What sort of UV tubes are you using?

You need blacklights, the sort used to read security markers and insect killer tubes are perfect.

You can also buy **broken link removed** from Craplin; also don't bother with the incandescent UV bulbs, they're crap.

I think you're problem here is you're not matching the ballast to the tube.

Under no circumstances use an 8W tube with a 36W ballast!

The 8W tube probably flickered, then overheated.

Always check the ballast you're using is designed for the tubes you intend to use, make sure the wattage and diameter is exactly the same!

Never wire the tubes in parallel, you might be able them to connect them in series providing the mains voltage is high enough but if you connect them in parallel only one will light then probably fail, followed by another and another until all the tubes fail.

 

[deansmith]

UV tubes

Thanks for your relpy, I think i have the right tubes. they are...OMNILUX UV-Röhre 8W G5 T5 5000h 300x16mm see also attachment. I dont fully understand how this works. I thought i could add up the wattage of my tubes to match the maximum wattage of the ballest, ,dividing the amps by wiring parallel and keeping to voltage the same. Well,looks like I thought wrong.
So do i need a ballast for each tube. if so do you know where i can buy them from.
Could iI wire the in series with 230V. ???

 

[Hero999]

Thanks for your relpy, I think i have the right tubes. they are...OMNILUX UV-Röhre 8W G5 T5 5000h 300x16mm see also attachment.

No, you need 8W BL/BLB T5, the tube in the attachment it looks like a BLB but the part number you mentioned is a germicidal tube used in sterilisers and EPROM erasers and is totally unsuitable for developing PCBs.

I dont fully understand how this works. I thought i could add up the wattage of my tubes to match the maximum wattage of the ballest, ,dividing the amps by wiring parallel and keeping to voltage the same. Well,looks like I thought wrong.

Yes, it is wrong.

Flourescent tubes can't be wired in parallel, they behave like spark gaps, the one with the lowest voltage will just strike leaving the others unlit, in your case they will all fail one after the other.

So do i need a ballast for each tube. if so do you know where i can buy them from.
Could iI wire the in series with 230V. ???

You could wire two in series and use a 16W, T5 ballast.

You can buy them from RS componants, Farnell, do a Google search.

Read RS's guide to fluroescent tubes.

 

[3v0]

I have been done this road.

Is there some reason you can not do toner transfer via laserprinter or copier instead of photographic. It is faster and much less mess.

They now have a foil that bonds to the toner to seal it. To etch the board you put a bit of Feric Chloride on a sponge and rub the board. Works for .01 inch traces.

See UBW_RS485 where I documented the process.

3v0

 

[Hero999]

The transfer process is easy but isn't 100% reliable and isn't the best quality.

The photographic process is harder but is more reliable and gives a better quality print and is often quicker for larger volumes.

I tend to use the transfer process for small amounts of through hole boards and the photographic process for SMUT boards.

 

[deansmith]

Is there some reason you can not do toner transfer via laserprinter or copier instead of photographic. It is faster and much less mess.

I have just been reading up on this method. at ...**broken link removed**.
seems good although the starter kit is expensive (Price: $79.95) .ill shop arounnd first ,but i would like to give it a go.

Read RS's guide to fluroescent tubes.

useful information. Learning by doing ,now i have most of the parts and a bit more knoledge, I will see it through. At least i will have learnt something from it, even if they are the wrong uv tubes.
Thanks again Much appreciated.

 

[3v0]

The transfer process is easy but isn't 100% reliable and isn't the best quality.

The photographic process is harder but is more reliable and gives a better quality print and is often quicker for larger volumes.

I tend to use the transfer process for small amounts of through hole boards and the photographic process for SMUT boards.

When you do the transfer process are you using the paper and foil from pulsar. It much improves the process.

 

[3v0]

The people at Pulsar made a real improvement over doing the transfer using clay based paper. The use of their system allows you to run 2 lines between IC pads. I have not tried it but I did etch a nice looking .01 line as a test on my first board using the process.

You can get started with just the paper and the green foil.

Mouser numbers
TONER REACTIVE FOIL GREEN 8 x15' 5165-GTRF $8.95
Transfer Paper (10 sheets 8.5x11 sheets) 5165-TTS10 $14.95

You will need to use a clothes iron to do the transfer or buy the laminator for $130. We had other uses for the laminator so we went that route. The pulsar site has info on how to calibrate your clothes iron to get the right heat setting.

You can use any PCB to etch but the 1/2oz copper is best in that it etches faster which reduces undercutting. Mouser has the boards pulsar supplies.

So for about $25 you get enough paper and foil to make a lot of boards. Over 900 square inches.

3v0

 

[Hero999]

When you do the transfer process are you using the paper and foil from pulsar. It much improves the process.

No, I don't see the point, magazine paper does the job, if I want better quality I use the photographic process.

 

[3v0]

This got much longer and maybe seems a bit snippy. Be assured that I mean no disrespect. In fact I take my hat off to you in that you have mastered the more difficult processes. I have always felt like a clutz when attempting them. That is in part why I am so very pleased that there is now a method for clutzes like me that provides good results.

No, I don't see the point, magazine paper does the job, if I want better quality I use the photographic process.

If you are happy with what you are using stick with it. You know all the correct incantations and offerings required by the petty gods of PCB etching. There is no reason for you to be interested in this process.

I like the way the paper floats off the toner when wet.
I like the foil that seals the toner that allows me to rub the copper away using a sponge.
I like the resolution it provides.
I like being able to go from printer to drill in 15 minutes. (If I remembered to turn on the laminator ahead of time)
I like the cost at less then 5 cents a square inch (+ board & etchant which are needed regardless of the resist process).

I like not having to have an exposure lamp.
I like not having to figure out how long to expose the board for.
I like not having to mess with an etching tank.
I like not having to wait for the board to etch in the etchig tank.

The same paper works as the first step in a process for making backing free decals. The toner is transfered to the paper. From the paper to mylar, and from the mylar to the final destination. Somewhere along the way you have the option to add gold or silver colored foil. There is a video of the process on the pulsar site. Simple but very neat/cool.

3v0