PCB exposure time based on light

I own the following UV exposure unit:



I bought it a year ago and now some lights no longer work. So I took out the non-working lights and tried to identify them, but the bulbs have no label and the metal does not either.

I then went to a few hardware stores, and the only thing that's available to me is the following:



The nice news is that I looked at this one in store and its exactly the same size and appearance as the bulbs in my unit (minus the black marks mine have from wear-and-tear), but according to someone who responded on my other thread, they suggested I needed a 4000K color? these ones are 3000K color.

Let's just say I replaced every light bulb in my unit (6 of them total) with six of these lights instead, just so I have light. If these new lights can work at all for PCB exposure, how should I factor the timing? Should I double or triple the exposure time because they're specs are somewhat off the required specs for pcb exposure?

No, if a regular bulb at all is going to have a chance at working at all then it would have to be at least 6500K. The specs aren't somewhat off. They're really off. It's not safe to look at UV bulbs but it's perfectly safe to look at 6500K bulbs.

It would be kind of like trying to expose your boards outside on a sunny day. That might actually be more effective since you can actually get a sunburn on a sunny day. When was the last time you got a sunburn from indoors fluorescent lights?

Since it's so slow and safe to look at with the box open monitor that way, I'm guessing it will be an unreasonably long amount of time (hours).

RS Components sell replacement UV tubes for exposure units.
https://uk.rs-online.com/web/p/uv-exposure-units-replacement-tube-starter-kits/0556250/

ditto dknguyen.
Also, a longer exposure time is more likely to cause undercutting.

Sinedup said:

ditto dknguyen.
a longer exposure time is more likely to cause undercutting.

Click to expand...

How so if the effectiveness if the light is greatly reduced?

I can handle undercutting to a certain degree because I make my traces and clearances larger than todays standards. My min clearance is 10mils and My min trace width is 12mils. Today's traces are 8mils or even lower I think.

But is there any way I can get the correct bulb in person without traveling a total of 140km just for a bulb? Because right now, I'm in a process of switching credit card companies so I won't be able to order online anytime soon.

mik3ca said:

I can handle undercutting to a certain degree because I make my traces and clearances larger than todays standards. My min clearance is 10mils and My min trace width is 12mils. Today's traces are 8mils or even lower I think.

But is there any way I can get the correct bulb in person without traveling a total of 140km just for a bulb? Because right now, I'm in a process of switching credit card companies so I won't be able to order online anytime soon.

Click to expand...

Do you have pet shops near you? Especially ones that sell stuff for reptiles? Some reptiles need UV bulbs.

dknguyen said:

How so if the effectiveness if the light is greatly reduced?

Click to expand...

What I've read on another forum, that long exposure allows light to progressively react with more of the masked sensitiser edges.
According to this guy, a strong, directional UV source for shorter periods creates a sharper trace.

dknguyen said:

Do you have pet shops near you? Especially ones that sell stuff for reptiles? Some reptiles need UV bulbs.

Click to expand...

Although not as long as 12 inches, a hardware store near us sells replacement fluorescent tubes for handheld UV lamps.
At only about 8 inches or so, this would need serious rewiring or rebuild of your exposure head.
Offhand, I don't have my UV torch nearby (buried in a box somewhere), otherwise I could furnish wattage output per tube.
At only 6VDC supply, maybe too weak for UV exposure.

there are a couple of pet stores. I checked petsmart on their online site and their tubes are too long and too pricey ($35 each)

I look here since this store is nearby (at least it was nearby as of last check):

**broken link removed**

and they can possibly order and sell me this:

**broken link removed**

They state:

The Power-GLO linear fluorescent normal output bulb with a colour temperature of 18000K fits all suitable standard and electronic ballasts. This fluorescent bulb generates super bright light with a strong blue spectral component adding to the vibrant white/bluish illumination,...

Click to expand...

Would it be suitable for PCB work or would that high colour number kill the presensitized layer in no time?

18000K? I never even knew such a bulb existed.

Did you not get the answer you wanted to hear in the previous post about this?
https://www.electro-tech-online.com...x-bulb-burnt-options-for-replacements.153999/

I told you the type of bulb to get (F15T8/BL), which I have used for years to make photo PCBs, but my reply was the final one in that thread; ie. you didn't even acknowledge if the bulbs are 18 inches (457mm) in length.

The bulbs measure 12 inches end to end. I'm sorry I forgot to measure.

I mean its easy to provide model numbers but I really would like to buy them from a store. I guess that means I'll have to order from the store I bought the box from then since I haven't found a light from a store that the electronics community approves for PCB use yet.

The simple answer is try it using a range of exposure times.

I think aquarium bulb specifications are very specific (to fish keeping). The majority of the spectrum is in the normal range and the "special bulbs" emit a very small amount at the stated temperature to make certain colours fluoresce.
I don't think they'll be suitable for this application.

Mike.