Exposure is probably the greatest variable and hardest one to control.

Lamps vary in intensity, different brands of transparencies vary in how well they pass UV, and stacks of transparencies can introduce other variables, such as more reflective loses. Some transparencies are specifically designed to block UV. I had to stop using a particular brand, because the manufacturer "improved" it by adding a UV blocking agent. Thus, the only practical solution for home use is experimentation to get the right time and distance. Sometimes, if the PCB I am making doesn't use the entire board available, I will cover the desired circuit after its intended exposure and then continue to expose test areas for additional periods as a confirmation that my exposure time doesn't need to be increased.

Awhile back, I put up a project here describing how to use dry erase markers to darken laserjet transparencies. Here is the link: http://forum.allaboutcircuits.com/showthread.php?t=12474

In sum, I would not recommend layering the transparencies. An LJ4 may be dark enough as it is, since you don't have any very fine traces. If it is not, then you might want to try the dry erase modification. I would recommend being sure the printed side is facing the PCB.

Finally, after the initial dip in developer and removal of most of the exposed resist, I use a small piece of synthetic kitchen sponge dipped in developer to go over areas that look like they have retained resist. The unexposed resist is quite durable. If you rinse your developed board in water, resist will make the water bead up. If you see any beading on areas that should not have resist, you can attack those with the sponge without risking over development of areas where the resist has been adequately removed.

I use the photo method exclusively and can't remember the last failure I had. But, attention to detail is important. I keep records of each time I do it, so I can make small adjustments as needed to adjust for aging of my UV lamps.

John

 

thanks for the tips. what kind of resist are you using? presensitised boards or are you laminating your own?

also: I just played around with eagle3d a bit and made this image.

OK, top let, the simplified wiring schematic. top right, the top side of the un populated board with parts placements and values, and bottom, the board rendered with substitutions for terminals and trimmer. I was hoping to etch some boards tonight, but it looks like thats not gonna happen. anyone know where I can get some fresh FeCl3 at 11:50pm on a thursday?

 

I use Injectorall, pre-sensitized boards. DigiKey carries them: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=PC55P-ND

No idea on the ferric chloride. It is used for a variety of other things, such as root killer and municipal sewage treatment.

John

 

I actually picked up a bottle of hydrochloric acid and some peroxide and Im doing the culpric chloride method. I have a full roll of negative resist film, and now that I know how to print my boards in reverse from within eagle, saved myself the trouble of trying to export it and convert it in another program.

I exposed a pair of boards and had the same problem. the resists just wouldnt wash away, and when it did it took the exposed areas with it too. So I started exposing them for less time. the instructions say to expose it for 12 minutes, so I tried it for ten first with less than great results, but I could tell it was getting better so I tried again at 8 minutes. this worked a LOT better, and I was able to get the two boards I needed out of it. I will try exposing my next boards for even less time next time.

mind you, I am using a proper UV bulb, and not a "daylight" bulb, or blacklight bulb either.

doe anyone have an exposure level test image I could use to experiment with different trace widths and exposure times?

 

I use a little image like this:

I look for a sharp intersection of the two triangles without over etching of the fine lines. My pre-sensitized boards have quite a leeway in exposure times. That is, 12 minutes to 15 minutes doesn't make much difference. Other people will simply make a pattern of straight lines from 4mil to 16 mil or so. You also want to keep a spacing of 8 to 16 mil or so between lines in at least one area. That is easily done with Eagle. If you don't get a good trace at 8 mil, you have a problem.

John

 

I came in late on this thread, so this may not be of any help.

How I make PCBs

 

I came in late on this thread, so this may not be of any help.

How I make PCBs

Thanks for the link Bill. I've done that toner transfer method before and it sucks. I spent the money on masking film and just hadn't perfected it yet. I have to say its a heck of a lot easier then trying to iron on toner and wash off paper. And much much quicker. After doing a bit of reading I am getting the impression that I could do fine with exposure times of as little as two to three minutes. That's a far cry from the 12 minutes I was originally exposing for. Tonight I plan on doing some exposure tests and seeing what I can come up with.

Either way, I've got the proper tools to do it right, I've just got to work out the kinks.

 

Hi, Johnny

Where do you get your photosensitizer film? What brand is it?

As for your etchant, if you consistently have problems with over etching and undercutting, I suggest you go back to ferric chloride. For hobbyists, it is basically the standard that other methods are compared to. I notice you mention using cupric chloride, then say HCl/peroxide. While it is true that HCl + peroxide involves production of cupric chloride, there is more to it than that. Cupric chloride (aka copper (II) chloride) can be used by itself in the presence of excess HCl. An etching bath is not quite as easy as ferric chloride to make, and the endpoint seems a little more sensitive to time, but it works well. I actually switched to it, because it never wears out. The copper it removes is converted to cupric chloride in the presence of air. All you need to do is watch the specific gravity and HCl concentration. You can judge the state of oxidation by color. Any hint of brown/olive drab means that some cuprous chloride (aka copper (I) chloride) is present. It will still work like that, because there is plenty of cupric ion around, but eventually, you will need to re-oxidize it. I just bubble air through my solution (cheap aquarium pump, plastic frit bubblers, not the "stone" ones) to replenish it, if I am using it a lot. If I am not going to be using it for several days, I just let it sit and stir occasionally.

Some resists, like the one I use, are attacked by the HCl/peroxide mixture. Albeit the attack is slow, but I did get pitting on the copper that should have been protected when I tried it a few times.

John

 

A bit late, but Radio Shack sells ferric chloride in 16oz bottles as "Etchant Solution".
http://www.radioshack.com/product/index.jsp?productId=2102868
You have to ask for "Etchant Solution" at the store, if you ask for "Ferric Chloride" they'll look at you as if you had two heads.

 

Hi, Johnny

Where do you get your photosensitizer film? What brand is it?

As for your etchant, if you consistently have problems with over etching and undercutting, I suggest you go back to ferric chloride. For hobbyists, it is basically the standard that other methods are compared to. I notice you mention using cupric chloride, then say HCl/peroxide. While it is true that HCl + peroxide involves production of cupric chloride, there is more to it than that. Cupric chloride (aka copper (II) chloride) can be used by itself in the presence of excess HCl. An etching bath is not quite as easy as ferric chloride to make, and the endpoint seems a little more sensitive to time, but it works well. I actually switched to it, because it never wears out. The copper it removes is converted to cupric chloride in the presence of air. All you need to do is watch the specific gravity and HCl concentration. You can judge the state of oxidation by color. Any hint of brown/olive drab means that some cuprous chloride (aka copper (I) chloride) is present. It will still work like that, because there is plenty of cupric ion around, but eventually, you will need to re-oxidize it. I just bubble air through my solution (cheap aquarium pump, plastic frit bubblers, not the "stone" ones) to replenish it, if I am using it a lot. If I am not going to be using it for several days, I just let it sit and stir occasionally.

Some resists, like the one I use, are attacked by the HCl/peroxide mixture. Albeit the attack is slow, but I did get pitting on the copper that should have been protected when I tried it a few times.

John

I am using MG Chemicals Dry Film (Negative) resist which I got from mouser.com. http://www.mouser.com/ProductDetail...=sGAEpiMZZMtyU1cDF2RqUOAVFWNYOnG7rfrmOLuWvvI=

its reccomended to be applied to the board with a laminator, but I've been able to stick it on with a regular iron instead. I just peel and stick it to the board, then the the board in a phone book and press down hard with a hot iron for a while. - Ive only made a few boards so far, so I've just been working out the kinks. -

the first issue with this film is that it is a negative exposre type, meaning that you need a negative of your image to burn it on, and the exposed areas harden up while the unexposed areas stay soft and disolve under the developer solution - which is the other issue. the developer solution is Potasium Carbonate, and its nasty stuff when you get it in your eyes, as I mentioned earlier.

the GOOD thing about this stuff is that if you screw up the mask somehow you can just remove it with alchohol or acetone and just reapply. - whereas with a presensitized board you are stuck with an otherwise useless copper clad piece, or then forced to use the toner transfer trick.

either way, since my success with it with these speedo converter boards the other night, I am really quite pleased with it.

as for etchant, Ive etched quite a bit with ferric chloride and it works well as long as its fresh, but my stuff is spent. I did get mine from radioshack originally, and had every intention of just getting more until someone suggested the cupric chloride trick. - meaning: taking Hydrochloric acid, and adding Hydrogen Peroxide to oxidize it, then adding copper which then disolves and creates cupric chloride. - so now I have curic chloride and Ive etched a couple boards in it. seems to work OK, and I didnt have any over etching happening, certainly nothing I would attribute to the etchant. Conveniently, I do have an aquarium pump that I can use when my green stuff turns icky brown.

I still intend to run some test exposures on my film and see what kind of trace widths I can get out of it.

Back on topic tho, the speedometer converter works! My friend installed it in his bus today and was able to set it and test it out. He took the time to post on facebook that it was working, but unfortunately he didnt do any real extensive testing on the linearity of it and whatnot. so Ill have to get all the details and post back soon.

 

the film I use for making my boards is the MG Chemicals negative dry film resist.
http://www.mouser.com/ProductDetail/MG-Chemicals/416DFR-5/?qs=sGAEpiMZZMtyU1cDF2RqUOAVFWNYOnG7rfrmOLuWvvI%3d datasheet: http://www.mgchemicals.com/download...df?PHPSESSID=a21dd3060783a8513ae7d7bcdc268537

the developer for it is the "MG Chemicals negative developer" - also known as potassium carbonate. http://www.mouser.com/ProductDetail/MG-Chemicals/4170-500ML/?qs=fFKnDw9uGJYeMAI9tnxFsQ== - MSDS here: http://www.mgchemicals.com/msds/english/4170.pdf

 

That's great news about the project. Congratulations. Please be sure to post the YouTube link when it is up.

I plan on giving the negative resist a try. Thanks for the info. The last time I looked for resist film, it was only available in commercial quantities (full rolls) and was quite expensive like that.

John

 

your welcome. for under 30$ you can be set up with 5 square feet of film and all the developer you should ever need.

 

Hello, I was goggling about adjusting the speedo and I found this. I have printed out the data-sheet for the 4151 chip and I am also designing a speed calibrator for my Motorcycle. I have some time on my hands and I am excited to build a similar circuit.

I am looking over your equations and I don't understand.

For example:
Rs=100k
Rb=50k
Ro=36k
Co=0.1uf
Hz= 200hz (I suspect this is my max)
V= 12 (standard for Bike/Car batteries)

Is this correct?

K^-1= 0.486 [100k/(50k*36K*0.1uf)] (200hz/12)
K^-1 =2250
K= 0.000444
Vo=Fi*0.00044

I don't think this is correct?
Please help hope your still active.

 

Welcome to AAC

I am afraid the OP has long gone.

The rules have changed since this thread was created, AAC no longer support automotive modifications, and that includes motorcycles. Please review our Terms of Service .

This thread will be closed, but feel free to participate elsewhere.