[SOLVED] issues in pcb making

i made a 4X15w UV exposure unit using GE F15T8-BL . my pcb sits 2 inches above the bulbs . I cant seem to get the thin track lines. I even tried low exposure times of just 4 mins but the ground area came out good but the thin lines didnt get developed.
I am not sure i am exposing it right . i use continous exposure but i was reading online that you have to give bursts of exposure .can some one explain how?
also i made the print out on inkjet transparency. should i put the coarse face (Where the ink print is ) on the glass or the other side?
I also noticed that the sensitized pcb are sticky once i remove the protective cover. should i press the tranparency against the pcb so it snugs around the pcb tight ?
what are the best practices in this pcb making area if someone can guide me.

thanks

Why not make a test PCB, a piece of PCB with tracks of decreasing widths across it. Expose the whole board for your 4 minutes, mask of a strip going across the tracks with a piece of thin cardboard, re-expose the board for another 2 minutes, slide the masking further across the tracks, repeat. . . Etch the board, you now have a board with strips of tracks with variable exposure, you should be able to figure out the optimum time.
Frank

good idea i will do that .
what about exposure method ? why every one is mentioning about short burst exposures or can i get away with continous exposure ?
also there are talks about which side of the positive image be placed on the pcb.

I can't think of a good reason why burst should be any different to a continuous period, the effect of the UV is cumulative regardless of how it is presented.

Where you may be going wrong is with the artwork, you should print it mirrored and then place the sensitized board on to it touching the inked side. Assuming the lights are below the board, you would have the inked side facing upwards and the board facing downward on top of it. This stops (or greatly reduces) the effect of UV scattering within the thickness of the film. I can reliably produce 10 mil tracks using that method. I use a laser printer rather than ink but that shouldn't make any difference. Also keep the board and film pressed together so no air bubbles can trap between them, my light box has a foam pad n the lid so when it is closed it squeezes the board and film against the glass top.

A test strip at different exposure times is a good idea, and of course, keep the lighting as dim as possible when developing the board. Especially avoid fluorescent lighting which itself tends to have a high UV content. Red LEDs are a good source of light to work under until the board is ready to rinse.

Brian.