I recently acquired this Fowler’s Long Scale Calculator through an eBay auction. The right-hand photo is a front view, the left-hand photo a rear view. This clever little gadget works on the same principle as a slide rule, using logarithmic scales — but by using six circles to hold one scale, they managed to pack a 30-inch long scale into an instrument that’s less than 3 inches in diameter! You can read more about how to use one here if you’re interested.
When I examined it upon its arrival (from Scotland!), I discovered that it was not actually in working order — the black cursor on the front did not rotate when I turned the right-hand knob. Functionally, everything else was ok. Aesthetically there were several issues: corrosion at various places on the metal parts, some “junk” floating around on the inside, a badly scratched cursor disk (more on that in a moment), and various marks and splotches all over the scales on both sides. It wasn’t quite a disaster, but it certainly wasn’t the nice, shiny calculating instrument I was hoping for.
I decided to fix it, and to try to clean it up a bit. The first challenge was how to open the darned thing up — it wasn’t obvious at all. I asked on the sliderule newsgroup if anyone had ever opened up a Fowler’s, but got no response. I turned that thing every which way and could see no screws or fasteners at all — it looked like the two sides were simply pressed onto the “chassis” and held there by tension. So I took a block of hardwood, wedged it against the exposed edge of the front half (which fit tightly over the back half), and pushed. Hard. And it gave just the tiniest bit. I worked my wood block around the circumference, and after about 20 little pushes, the whole front just fell right off. The back half came off the chassis in a similar way. Whew!
When the front fell off, the reason for the black hairline not moving became obvious: the black hairline was etched into a clear plastic (celluloid?) disk (the aforementioned cursor disk), and that disk had come unglued from the brass post in the center (you can see this post on the photo). This is why the cursor disk didn’t rotate. I set that aside for later repair, and took a close look at the scales. To my surprise, the scales are simply printed on thick paper, which is then glued to a metal disk. The markings and dirt that I saw on those disks would not be as easy to clean as an ordinary slide rule, where I could use soapy water — sure don’t want to do that with paper! I’d read that a soft art or drafting eraser was a good way to clean paper scales, but I didn’t happen to have one of them around. But…I did have some Micro-Mesh polishing clothes — basically very fine abrasive (in several grits) embedded into a rubbery compound with a cloth backing. I tried the finest grit I had (12,000), and discovered that it worked very well indeed. In just a few minutes I had nearly all of the marks off the paper scales.
The next challenge was the cursor disk. The old glue was still stuck to both the disk and the post — the first job was to get all that off. I used dental picks to get the big chunks off, then used a small flat jeweler’s screwdriver as a chisel to scrape the rest off. I had to be very careful when scraping the post not to lose control of the screwdriver and tear the paper scale — and equally careful on the disk so as not to poke a hole in it. This was tedious! The cursor disk was very badly scratched, with circular scratchs from the motion of the cursor disk moving over the scale. This Fowler’s must have been used a lot to cause all that scratching! I’ve had some previous success with Novus plastic polishing compound, so I tried that — and got terrific results, albeit with a good deal of elbow grease required. I had to take great care here not to fold up the cursor disk as I bore down hard while polishing. It is very thin and flexible, and threatened to fold up (and almost certainly break) on several occasions. In the end, though, it was well worth while — that cursor disk is now as smooth and shiny as the day it was made. Not a scratch on it. I used Duco cement to glue the disk back onto the post, being very careful with the glue — I didn’t want to get glue on the scales, where it would have been impossible to clean up. So I squeezed glue onto a jeweler’s screwdriver, and then used the screwdriver to carefully “paint” the brass post with glue. This worked well, and now the Fowler’s was back in working order.
I was hoping to find a way to lubricate the gear mechanism, which is in the innards of the chassis. But I could not find any way to gain access to that area. The scales are mounted on metal disks that block all access, and those disks appear to be permanently fastened on. So I gave up on that part.
I used some Nevr-Dull metal polishing cotton to get all the corrosion off the exposed metal parts. While I was handling the back half, the glass fell right out — the glue holding it in had long ago dried completely up and given up on any adhesion. When I discovered that, I pressed gently on the front half’s glass, and it came right out as well. So then I cleaned up all that old glue, thoroughly cleaned both pieces of glass, and reinstalled the glass with Duco cement. That worked great.
After the glue set, I squeezed the back side back onto the chassis, and then the front side onto the back side. And I had a much nicer Fowler’s Long Scale Calculator than I started out with, not the least because now it worked!
In the old blog, Simon said:
ReplyDeleteReading your post caused me to remember a gift I once received from my late father: a Curta. See http://www.webcom.com/calc/Curta_text.html for details of this amazing machine. I was a young boy at the time of the gift and I have no idea what happened to it. I’m also embarrassed to admit that I never spent much time figuring out how to use it. I bet your experience would have been different. My father had a penchant for buying gifts during his numerous overseas travels. Oftentimes he’d buy some electronic gadget that worked fine in Singapore or Hong Kong but not in Australia with our 240V 50 Hz power. The Curta, of course, would have worked fine…
In the old blog, Anonymous said:
ReplyDeleteHi, Simon…
I actually have one of those Curta calculators (a Curta II) — in working order, but in very bad need of a cleanup. And actually I’d like to get one that’s overall in better condition, but they are going for astronomical prices these days (in the neighborhood of $2000 for one in first class condition). I’ll admit to being a little intimidated by the thought of opening my Curta up — there are, I read, some 400 parts inside that small case! There are several people who have made a specialty out of cleaning and repairing Curtas; I may send mine to one of them.
There’s another side to the Curta story that’s kind of interesting. Before about three years ago, the interest in Curtas was confined to a relatively small group of collectors, and their market value was much lower (a few hundred dollars at most). But then Scientific American published a paean to the Curta, and almost overnight the number of collectors jumped; demand suddenly greatly exceeded supply…
Did you know that rally car drivers were one of the primary markets for the Curta? I’ve never seen an explanation for why this was so, but I’d guess that at least half the Curtas I’ve seen for sale were previously owned by rally enthusiasts…
I have learned how to use the Curta — I can add, subtract, multiply, and divide on it. Actually, all but division are nearly trivial operations. Division is only slightly more challenging; basically the human operator is following a simple algorithm. Being a programmer of the ancient and venerable type makes this a bit easier for me — because the Curta uses the exact same algorithms (though in base 10 instead of base 2) to multiply and divide that computer hardware does. In the early days of microcomputers, the chips available (such as the 8080 or Z80) didn’t have hardware multiply and divide capability, so we had to write assembly language code to implement those operations. That code is eerily similar to the multiply and divide methods on the Curta…
In the old blog, Simon said:
ReplyDeleteHi Tom,http://curta.org/wiki/RoadRally and http://www.rallyracingnews.com/manuals/curtaman.html should clear that up for you.Simon