Here's a jig I made up for my dremel. I use it mainly for making manhatten style IC pads. The idea is not originally mine. I based it off of this one, modified a bit for my purposes. It's not the safest tool in the world, but use cautiously it seems to produce these pads fairly well (read: I am not to be held responsible if you hurt yourself). All of the images can be clicked on to see them in a larger size, but without the colorful markup. If you have a slower internet connection, be warned, a few of the full-size images are a bit over 1MB. Most are around 200k.
Here we have the jig in it's normal operating position (left), and it's open position (right). With the jig open, I have pointed out the hinge which lets it open, three countersunk nuts, and a shim. The hinge is very small, and a bit wobbly. Really, it's only purpose in life is to let it open up fully like this. It generally doens't have to be very substantial. The nuts are sunk into particle board by drilling holes slightly smaller than the OD of the nuts, then using a bolt and washer to draw them in. The particle board is pretty soft stuff, and they draw in quite easily. Once in, they seem to do a good job at staying. The shim is just a scrap piece of wood that let's the depth adjuster screw raise the dremel up high enough. The shim isn't attached in any fashion.. it just sits there, held by gravity.

The base is made from a scrap piece of red oak. The U-shaped piece that the dremel rests in is made from scrap particle board. The bolts and nuts are from the junk drawer of my toolbox. The hinge I found floating around in a tool box. The carriage is made from two pieces of 4" wide tonge-and-groove oak flooring glued together. The tongue side of those boards has been cut off so that edge is flush. The shim is also a scrap piece of oak flooring. The rails that the carriage rides on, as well as the fence, are made from some scrap acrylic. The index wheel was made with casting plastic poured into an empty fruit cup (the bottom of the fruit cup had those nice indentations which looked perfect for index marks). When I cast the index wheel, I embedded a 2" (or so) bolt from the junk box into it, head first. The index wheel rides in a piece of scrap steel.

With the carriage removed, you can see the rails it rides on. The rails do a pretty good job of keeping the carriage square as it moves perpendicular to the cutting wheel. You can also see the depth adjuster (long bolt out of the junk drawer) resting on it's shim. I probably could have done a better job designing and found a way around having to use a shim, but I put this thing together in about an hour, and cared more that it worked than looked good. It works well enough it's not been worth my time to re-engineer.

Although you cannot see it clearly on the right carriage rail, both of them do extend nearly the full length of the base board. The right rail stops right against the hinge (which you can see a small nub of in the upper right). This length is completely overkill. The rails really only need to go as far back as the dark line that runs between them. This again came down to building without much design work. The rails are simple pieces of scrap lexan about 1/2" wide by whatever length the base board is. Again, length only *needed* to go to that black line. To install the rails, I cut a ~1/4" deep channel with the circular saw (I'm not lucky enough to own a nice table saw... just this old inherited circular saw. It does the trick, and I'm greatful for having it). The thickness of the blade happened to be a good match to the thickness of the lexan. Oh, and it might not be real lexan, it be some cheaper form of acrylic. Anyway... with the slots cut, I put a bit of wood glue in the slots, then pushed the plastic strips into them. The plastic rails are a bit thinner than the slots, but once the glue set up they stopped wobbling. They've been stable ever since.

Looking at the carriage alone, you can see the plastic fence in the image on the left. I installed the fence the same way as the carriage rails. The fence is embedded a little bit deeper than the carriage rails were, and only sticks up about 1/4-3/16". Because the slot for the fence went so deep, I stacked two of these scrap pieces on top of each other for extra strength.

In the image on the right, you can see some slots I cut in the underside of the carriage. These slots are what the rails in the base ride in. When I first cut them, everything fit together very nicely, and the carriage slid smoothly on the rails. Now, the rails are very tight in the carriage. I believe this is because the grain of the wood in the carriage is running perpendicular to the grain in the base. Wood tends to grow and shrink a bit over time as it absorbs and releases moisture from the air. Since the carriage and base are growing and shrinking in different directions, they're not happy with each other all the time. Some day I'll probably try to widen these slots a little bit to make things move more smoothly all the time, but for now it's only a minor annoyance. If I ever come across a bunch of UHMW plastic for cheap or free, I might be inclined to rebuild this thing out of it. It would be much more dimensionally stable, and about as easy to work with as the wood. If you've never heard of UHMW plastic, go find your XYL's thick plastic cutting board. Good chance it's UHMW. Neat stuff.

The left image shows the index wheel head-on. If you look closely, you should find a bit of blue marker next to one of the orange index marks. I use this as my "zero" setting, lining it up with the metal plate (where it is lined up in the piture). I calculated the distance that the carriage moves for each turn based on the TPI of the bolt I cast into the index wheel. 1 turn should move the carriage about 0.036". It might seem weird that I also have values for 1/6 and 1/18 of a turn, until you count the number of marks on the wheel. In theory, the wheel should let me get down to a 0.002" carriage move. In reality, I do not expect it to do this well. In practice, I figure out what the spacing needs to be, and then figure out the number of turns needed per pass. SOIC dimensions for a PIC16F627A (something I might use) specify about 0.033" between pins. That's about 1 turn of the wheel (or 1 1/6 turn, if you want to try for high precision), which seems to be an accuracy that this jig is capable of obtaining, at least for smaller pin counts. 16 pin SOIC packages seem to do ok. I haven't tried bigger. Yes, I know you want to see pictures of a pad to prove this! The real limiting factor, for me, has been the thickness of the cutoff wheel. The Dremel branded diamond cutoff wheel I'm using measures about 0.020" thick. It also wobbles a bit. I am not sure if this is due to my neglect, or if they tend to come this way from the factory. I tried using some other thin abrasive cutoff wheels before moving to the diamond wheel you see in these images. The abrasive wheels wear away as you use them, their diameter getting ever smaller. This produces a constantly changing cutting depth. Since I usually make my pads out of pretty thin PCB stock to begin with, this is doubly annoying. I don't have much extra FR4 thickness to play with!

The image on the right shows the plate and notch that the index wheel rides in. It is made from a scrap piece of 1/16" thick mild steel that I had in the garage. After trimming it down to size, I marked a horizontal line about 3/4" up from one edge (you can still see a bit of the line from the marker). Once marked, I knew the area it would be safe to drill holes in for mounting to the board. Once drilled, I countersunk them a bit, mostly because I had found wood screws with countersink heads in the junk box.

The notch in the plate matches up with a groove I ground in the bolt shank on the index wheel. This keeps the index wheel in one spot, and lets it pull the carriage in and out. The carriage has a complimentary nut embedded in it. This should make more sense once I get some picures of the carriage and wheel separately. It would probably be more functional if I put a bit of allthread in the carriage, and fixed a nut to the wheel, but that would also require something like a second plate with the wheel/nut sandwiched between.

Looking at the back side of the index wheel, you can see where I ground a groove in it. This groove sets into the notch in the steel plate. It does an OK job of holding the index wheel steady while you dial in a new carriage position. I still like to add a little downward pressure on the carriage while I'm adjusting the wheel, just to be sure that the groove in the index wheel doesn't leave the notch in the steel plate. I originally cut the groove by putting the threaded end of the wheel into the chuck of my drill press, then cutting into the shank of the bolt with a triangle file. I know, drill press bearings aren't designed for lateral load, it's a drill press, not a lathe. I really don't generally abuse my drill press like that. In fact, this is the only time I can remember doing so.
Here's a look at the nut I've embedded into the carriage. It's held in with JB Weld (2 part "quick steel" epoxy). The carriage is drilled out behind the nut deep enough for the index wheel to be threaded completely in.
Ready to see the jig in action?