I'm trying to create a circuit board via the toner transfer method (print artwork on transparency then expose via UV). This time, instead of a homemade box, I use a professional UV exposure box. It does much better at producing PCB's for the most part, but I tried a large one (about 8 inches by 4 inches) but after developing the PCB, I tried to align an IC to the holes and the alignment is off on the PCB, but the alignment is correct on the transparency, so the software and the printer itself isn't the culprit. The IC pin spacing on the PCB seems to be about 25% smaller which is no good because then I have to force the legs to be closer together if I'm going to successfully fit the IC in the holes.

I tape two sheets of transparency together to create a dark image which seems to be sufficient.

Now I think my problem is either of the following...

1. Bad taping.

2. Insufficient sanding.

The reason why I say #2 is because I used a utility knife to cut the board from a larger board as this method is cheaper than to purchase multiple smaller boards, and every time I do this the copper is raised at the edges that came in contact with the utility knife.

I also think its possible #1 is a culprit but I don't think by 100% because I only put tape on at the corners.

Correct me if I'm wrong, but I think my answer is to look for some specialized sand paper (or equivalent) that can nicely sand an edge of a PCB so it looks like it was sanded by the manufacturer. What sand paper grit should I use or if there's a better way to solve my problem without wasting money on smaller boards, let me know.

 

I'm trying to create a circuit board via the toner transfer method (print artwork on transparency then expose via UV). This time, instead of a homemade box, I use a professional UV exposure box. It does much better at producing PCB's for the most part, but I tried a large one (about 8 inches by 4 inches) but after developing the PCB, I tried to align an IC to the holes and the alignment is off on the PCB, but the alignment is correct on the transparency, so the software and the printer itself isn't the culprit. The IC pin spacing on the PCB seems to be about 25% smaller which is no good because then I have to force the legs to be closer together if I'm going to successfully fit the IC in the holes.

I tape two sheets of transparency together to create a dark image which seems to be sufficient.

Now I think my problem is either of the following...

1. Bad taping.

2. Insufficient sanding.

The reason why I say #2 is because I used a utility knife to cut the board from a larger board as this method is cheaper than to purchase multiple smaller boards, and every time I do this the copper is raised at the edges that came in contact with the utility knife.

I also think its possible #1 is a culprit but I don't think by 100% because I only put tape on at the corners.

Correct me if I'm wrong, but I think my answer is to look for some specialized sand paper (or equivalent) that can nicely sand an edge of a PCB so it looks like it was sanded by the manufacturer. What sand paper grit should I use or if there's a better way to solve my problem without wasting money on smaller boards, let me know.

I simply run a utility knife at 90-degrees to the raised edge and pull back to scrape it away. Copper is soft and doesn't sand well.

You said it is off by 25%? That is a lot. That means a DIP8 layout is the length of three pins of the actual part (instead of four). Is that correct?

 

I simply run a utility knife at 90-degrees to the raised edge and pull back to scrape it away. Copper is soft and doesn't sand well.

You said it is off by 25%? That is a lot. That means a DIP8 layout is the length of three pins of the actual part (instead of four). Is that correct?

I might of got the percent wrong but the length of a DIP40 chip is more like a DIP36 length on the PCB. and what do you mean by 90 degrees to the edge?

 

I might of got the percent wrong but the length of a DIP40 chip is more like a DIP36 length on the PCB. and what do you mean by 90 degrees to the edge?

That is a lot of shrinkage.

For the rough edge,
1) hold the pcb in front of you (with your left hand) with copper side facing up and rough edge on the right side. You should be holding the far side and near (not left and right side).

2) with utility knife in right hand, position the blade parallel with the far side of the board, just touching the righ edge.

3) lower the handle until the blade makes a 45-degree angle with the copper (the right edge should be the fulcrum for the blade as you lowered the handle).

4) put some pressure on the board and draw the blade towards your torso.

5) remind me to say something about utility knife safety as step one next time.

You are not slicing, you are scraping. A nice curl of copper should come up if you do it perfectly and if you had a perfect ridge of copper. Otherwise little shards or hairs of copper are also good.

 

That is a lot of shrinkage.

For the rough edge,
1) hold the pcb in front of you (with your left hand) with copper side facing up and rough edge on the right side. You should be holding the far side and near (not left and right side).

2) with utility knife in right hand, position the blade parallel with the far side of the board, just touching the righ edge.

3) lower the handle until the blade makes a 45-degree angle with the copper (the right edge should be the fulcrum for the blade as you lowered the handle).

4) put some pressure on the board and draw the blade towards your torso.

5) remind me to say something about utility knife safety as step one next time.

You are not slicing, you are scraping. A nice curl of copper should come up if you do it perfectly and if you had a perfect ridge of copper. Otherwise little shards or hairs of copper are also good.

I think I did that before once. but it takes alot of copper away. I'm wondering if the transparency is the problem. I'm using one compatible with inkjet and lazer printers but I print with lazer

 

Agree. Utility knife or a file. If you do a lot, try a hand deburring tool: https://www.mscdirect.com/browse/tn...-Sets/Hand-Deburring-Tool-Sets?navid=12100123

The usual method is to use a sheet of glass on each side for exposure. I've found that a drop of water works with polyester paper, anyway.

 

Agree. Utility knife or a file. If you do a lot, try a hand deburring tool: https://www.mscdirect.com/browse/tn...-Sets/Hand-Deburring-Tool-Sets?navid=12100123

The usual method is to use a sheet of glass on each side for exposure. I've found that a drop of water works with polyester paper, anyway.

ok i'll try a file. Is there another name for deburring? and I'm doing single-sided boards and the uv exposure unit I have already has glass in it.

 

Also, what else can I do to prevent this problem from happening again?

 

There is another technique, described in detail in another thread (no, I am not looking for it. It's 5am and I fell asleep an hour ago. And I have to be awake in an hour for meeting with government officials)

After printing your transparency (yes, I mean one), go over it completely with a dry erase marker. Then, gently rub it off. The pigments get trapped in the toner, obviating the need for a second transparency.

 

This http://www.megauk.com/artwork_films.php is the "sort of" film I used. For some reason, there was some like it in the lab.

Looks like this http://www.ladyada.net/library/pcb/inhouseetch.html us a good tutorial.

 

Also, what else can I do to prevent this problem from happening again?

For the best registration, you will also want the ink side toward the PCB.

 

I'm trying to create a circuit board via the toner transfer method (print artwork on transparency then expose via UV).

Surely with the toner transfer method no UV exposure is necessary?

 

I do not understand how the image could possibly come out 25% smaller than the transparency.

Can you sow us a picture of the transparency placed over the developed board?

Bob

 

Surely with the toner transfer method no UV exposure is necessary?

I do not understand how the image could possibly come out 25% smaller than the transparency.

Can you sow us a picture of the transparency placed over the developed board?

Bob

I think we have a troll.