PDA

View Full Version : PIC chips, stepper motors and precision



Demon
- 5th February 2005, 17:19
I am considering making a small PCB drilling station and I was wondering just how precise we can get using this technology. Are we able to go down to thousands of an inch?

If Pic BASIC Pro is unable to accurately measure distance and we require an external distance-mesuring-tool, I was wondering if there was something we can do with an ordinary optical mouse. Two units could be placed to measure the X and Y displacement.

I'm new to PICn my nose, and even more so when it comes to robotics. I was wondering if it was possible to build such a device with decent accuracy without needing components from NASA subcontractors.

Robert
:)

mister_e
- 5th February 2005, 17:29
Hi Robert,

I think everything is possible using step motor control. IMO the accuracy, precision will be more in the gearing you place than in some IR detection.

I'd already think to do something like that few years ago. Looks like i'm not the only one who's tired to use a stand-alone mini-press drill system and scrap my eyes on drill bit to pad allignement ;)

The PCB software i use (P-CAD), create a x/y mapping .TXT file. i was planning using this file by sending info to the PIC thru COM port.

Once the PCB is drilled, place the film over and etch it. Not sure if it's the best solution but make enough sense to work.

Demon
- 5th February 2005, 18:28
Thanks Steve.

I'd say etch first, then drill. I've often read that drilling creates a small 'lip' around the hole, interfering with the proper etching of pads.

STEP #1:

- use a bolted drilling platform for each PCB format you drill.

- drill 4 corner holes on the platform using the software, they will be your pilot holes.

- place metal rods in the holes tall enough to hold several PCBs in place.


STEP#2:

- etch your PCB, always including those 4 corner holes.

- manually drill the 4 corner holes.

- slip 2 or 3 etched PCBs onto the drilling platform.

Ta daaa!

That's my idea to get the same alignment all the time. You only have to manually drill 4 holes for each new PCB. 2 holes would be sufficient, but 4 adds stability in case you screw up 1 hole. You can drill 6 holes or more for added safety if you suck as much as me.

It might not be perfect, but it would be stable enough for my requirements and low volume.

Robert
:)

OH YEAH, don't forget to include a clamping mechanism on all 4 sides to prevent the PCB from 'lifting' along with the drill bit. One idea is to use 4 bolts instead of rods, so you can bolt the PCBs into place.

mslaney
- 5th February 2005, 20:56
If you're going to use IR as the origin sensor, be sure to shield it from ambient light and keep it clean if you use a coolant.
Use a coolant!- especially with soft metals. Even WD-40 is better than nothing. I have scrap to back me up on this!

Look into an encoder or laser. Lexmark (and other printers) have very nice encoders and strips you can hack and integrate for position feedback.

I presently use the corner of my vise as the origin and get the accuacy I need. Use a $10.00 edge finder
(http://www.mscdirect.com) to find the inner surface of the vice jaw for the Y and the left hand side of the vice jaw for the X then set that as your 0-origin in your program. When you place the part in the vice, the Y is inherently zeroed and the x can be set by aligning the edge of the part with the edge of the vise using a piece of hard smooth material. This is accurate and repeatable.

The smaller the whole, the higher the RPM. Mass production PCB drills run at around 8KRPM. Use coolant.

I use a bridgeport and coolant. An off the shelf drill press will rarely give you .001" in accuracy but you'll get a .040" hole in a .100" solder pad with enough surface for a good connection. Use/make a bigger pad if you can.

If you drill stacked parts, the material from the lower part will be pulled up between the parts and cause the topmost part to distort - giving you an oversized or almost triangular hole in the top part.
Take a piece of PCB material and drill oversize holes in the same position(s) as the holes in the part you want to keep. Put the oversized part in between the keeper parts and you will have little pockets for the chips. This will prevent the distortion and also give you excellent support around the holes you are drilling.

If you can, drill a whole in the middle of the part first (you can use this as a mounting hole in the finished part) and screw all the parts together, this will prevent pullup in and near the center of the part when you drill the holes for the components.

Did I mention using coolant? :-) "Tap Magic" is great but WD-40 will be OK.

If you knew all this, sorry. If you didn't, I have finally contributed *something* to a forum that has been incredibly helpful to me!!!

Demon
- 6th February 2005, 05:05
Thanks Mike, I think I'll add coolant, just a thought... LOL

Yeah, I used to use Tap Magic or some other product whenever I drilled in the workshed. But such a product is not adequate for my needs, it smells and will stink up the place. Water is already an improvement over air, some additive could be included like that yellow stuff we used to have in machineshop back in high school (lubricant I guess).

I hadn't thought of sandwiching a template with oversized holes. That's a great idea, perfect for giving space where the removed matter can deposit. Everyone mentionned that 'pull-up' side-effect, but nobody had any solutions except "drill 1 pcb" d'uh...

Bolting the center is also a good idea. I was wondering about that as I wrote about bolting the corners. I suppose a 5th hole in the middle isn't gonna hurt anyone and will stop larger boards from lifting in the center. Smaller boards are probably not a concern, but it's a good habit to get into.

I have one last comment, carbide. Don't even consider using standard high-speed drill bits, FR-4 eats that for breakfast. I've been fortunate to get my hands on 100 industrial grade carbide bits of various sizes, awesome stuff.

But even carbide could use some coolant...

Robert
:)

mslaney
- 6th February 2005, 15:48
Mmmmmmmm, carbide. I feel like I'm on one of those *other* boards.... ;-)
Yes, that's a good idea - especially if you have a ton on hand. Those ceramic/glass fibers do just as much damage to your lungs too.The coolant will also keep the fibers from floating around too.

I spent half of my life drilling FR4 and other insulators and I am finally learning why. The electronics are a lot more fun than milling the boards!! And it's funny, I can make them BOTH smoke!

good luck with your product. What are you making, by the way?

Demon
- 6th February 2005, 17:28
Thanks Mike, I'll need all the luck I can get. I haven't gotten to making small fires though, so I must be doing something right. I forgot about the airborn particles, yeah FR4 is nasty, liquid will remove 99% of it. And the good thing about liquid is that it's a lot less noisy than the vacuum I was planning on using, easier on the ears.

As for my project, it's a little console to help me with my gaming. I've been playing the NASCAR Racing series by Papyrus (now defunct) since forever. NR 2003 is the latest version and supports up to 43 online drivers, just like the real thing.

You get to do just about every adjustment to your car when you pit as the real drivers do. The problem is that I'm not good at finger acrobatics, watching where I'm going, searching on the keyboard in the dark, so things like 2-key commands are somewhat harder. And then there's the problem of me never remembering what function key is used for what adjustment.

So I'm making myself a little menu-driven console in an attempt to reduce my wrecking as I enter adjustments, in a turn, with other drivers all over me.

I'm gonna post pics once I'm done.

Robert
:)