Also, just as an aside, IBM punch cards are getting harder to find and pricy.

On Sep 4, 2025, at 06:32, Bill Degnan via cctalk <cctalk(a)classiccmp.org> wrote: David, My Operators Guide (to) Electronic Punched Card Accounting Machines put out by IBM in 1951 lists the 513, 514, 519. I agree that the two you're looking for were special units, and it appears not commercially available. To locate the docs you need to track someone down who has documents related to the ENIAC itself. I might try the Hagley Museum in Wilmington, Delaware, sometimes they have rare technical documentation and it might be worth a shot. Bill kennettclassic.com / vintagecomputer.net > On Thu, Sep 4, 2025 at 6:51 AM David Wade via cctalk <cctalk(a)classiccmp.org> > wrote: > > > >> On 03/09/2025 21:48, Brian L. Stuart wrote: >> Apologies in advance to anyone who gets this multiple times. >> I know a number of you are on more that one of these lists. >> >> I'm looking for some information on a couple of the early >> IBM unit record devices, in particular the Type 512A and >> the Type 518. Ideally, I'd like to get the mechanical >> dimensions in enough detail to create a CAD model. > > As I assume you are aware, but perhaps others on here are not, that > these pre-date computing and must come from the era when data processing > involved only punched cards and I feel well before the term "Unit > Record" was coined. Sadly, this seems to be a forgotten era and there is > very little information about it on the web. > >> The background is that I'd like to improve the model that >> I'm using in my ENIAC simulator and I'm involved with a >> school in Arizona that's working on a project to build a >> full-scale model of the ENIAC. There's a letter from >> IBM in the ENIAC archives that suggest the use of the >> 512A and the 518. So I'm pretty sure those are the ones >> they used, but I haven't really been able to find any >> details about them. We've found a fair amount of >> information on the Type 513, and from photographs, the >> punch used on the machine seems similar to the 513, but >> is smaller. > > Looking at the photographs linked from here:- > > https://www.columbia.edu/cu/computinghistory/eniac.html > > in particular > > https://www.columbia.edu/cu/computinghistory/eniac6.jpg > > I think you are correct, but I wonder if the 512A and 518 were > "specials" which is why there is little information on them. > I believe that in Punched Card processing normally you initially punched > cards by hand and any punches without keyboards were some kind of > calculating re-producer. > > So for example on a loan system, you charged interest every 6 months. > For each loan you had a card with the loan amount. > You fed these into a calculating reproducing punch which calculated the > interest and punched a new card with the interest owing on it. (we > actually punched two) > You could use these to produce the interest letters. When the interest > was paid you could use the cards in a tabulator to print a summary of > the days/weeks payments. > After a month the cards you had left could be used to chase payment... > .. when I started work in 1976 one of my first jobs was to replace such > a system with screens.... > > So any punch you used probably had a reader attached and a plug board > that controlled how the cards were copied. The 513 is such a device. So > it has reader, plugboard and a punch. > I feel the devices ENIAC used were simply punches or readers. Looking > through the publicly accessible archives at > > https://www.columbia.edu/cu/computinghistory/ > > its apparent that IBM had already produced custom versions of their > equipment, so it likely that these were specials. > The only other pictures I could find were in this book, 28th page in the > PDF > > > http://www.bitsavers.org/pdf/ibm/punchedCard/Training/IBM_Accounting_Course… > > which shows a reproducing punch so I wonder if that is a 512.. > > >> >> Thanks in advance, >> BLS > I am sorry not to be more helpful > Dave > >

On Sep 4, 2025, at 12:56, Brian L. Stuart via cctalk <cctalk(a)classiccmp.org> wrote: In the interest of not cluttering everyone's inbox, I'm going to try to respond to everyone all at once. This may fail spectacularly. :) Van Snyder wrote: Good suggestion. I hadn't thought to ask him, but he would be a good one to talk to. Wayne S wrote: That is where I found some of the information I was able to find on the Type 513, but as pointed out below, I did overlook the training manual. I tend to steer away from the real-time conversational tools, but this might be a good enough reason to bite the bullet on it. David Wade wrote: I can't remember where I found this, but the best photograph of these units that I've seen is: http://cs.drexel.edu/~bls96/eniac/brl-card.jpg I've wondered that as well. The wording in the IBM letter that I found reads: "Some two or three months ago, Colonel P.N. Gillon called on Mr. Thomas J. Watson, President of International Business Machines Corporation, with a request that we furnish certain of our standard units for use with a special machine for calculating purposes now being built. ... "We have worked out the details of the interconnection of our machines with the remainder of the apparatus and are ready to start construction of the units at our factory. In order to start construction on any of our machines, an order and WPB approval is required. "It is suggested that a new contract be entered into for two items: 1 - Card Reading Unit (Type 512A) 1 - Gang Summary Punch (Type 518)" That would be very consistent with them being special units, probably variations on standard products. If anyone is interested in seeing the letter itself, here's a scan from microfilm: http://cs.drexel.edu/~bls96/eniac/letters/ibm-devs.pdf I hadn't looked that the training manual there. My bad. The one on that page labeled Card Reproducing Punch looks a lot like the one that I'm pretty sure is the punch in the ENIAC pictures. Bill Degnan wrote: I'm inclined to think we have enough informed opinions to say that unless something else comes to the fore, the best conclusion is that they were indeed special units. The only reference I've seen to the 518 outside this letter is: https://en.wikipedia.org/wiki/List_of_IBM_products and that lists as reference a book that doesn't appear to be on archive.org. Good idea. Most of my research has been at the UPenn archives, and they do have some overlap with the Hagley, but I haven't been to Hagley yet and should visit. As a bit of background, some years ago the Hagley, the UPenn archives, and the Charles Babbage institute all realized that among them, they had almost all of the exhibits from the patent trial. So they got a grant to gather them all together and have them microfilmed. Each of the institutions got back their originals and full sets of the microfilm. I've actually been through it (all 210 rolls) and have capture several GB of scans. I'm still processing them an making them available, with priority on the engineering drawings. If you're interested in taking a look at the drawings, here's a link to the ones I've posted so far: http://cs.drexel.edu/~bls96/eniac/drawings/ And anyone who will be at VCFMW next weekend is welcome to stop by my table and see my ENIAC simulaton if you're interested. Thanks for all the tips, BLS

On Thu, 2025-09-04 at 11:35 +0100, David Wade via cctalk wrote: The Computer History Museum in Mountain View, CA has a 519. They might have other models as well, or information about them. Contact Dag Spicer <spicer(a)computerhistory.org>rg>. Haus zur Geschichte der IBM Datenvararbeitung (House of the History of IBM Data Processing) had a large collection in what had been a punch- card factory in Sindelfingen, a suburb of Stuttgart. They had a 650, 1401, and 360\25 in working order. IBM sold the building and moved the collection to the IBM technology center across the autobahn in Böblingen. I don't know how much of the collection they retained. Contact Hans Spengler <hhwsppb(a)t-onlind.de>de>, Werner Seebode <werner(a)familie-seebode.de>de>, or Heinz Oberle <HOberle(a)t-online.de>de>.