Oral-History:Robert Mumma

About Robert Mumma

Robert Mumma was born July 28, 1905, in Manila, Philippine Islands, to missionary parents, returning to America in 1914. Mumma became interested in radio and attended the electrical engineering program at Purdue for two years (1922-24), before deciding to leave and become a teacher. He received his teaching certificate from Otterbein College in Ohio, and taught high school in both Florida and Ohio. Mumma returned to work in the electronics field, however, after he received a tutorial from the head of the electrical engineering department at Miami University of Ohio in high frequency AC theory.  He got a position at General Motors Radio, where he worked in Joe Desch’s department making test equipment. After that company folded, Mumma worked for police radio in Dayton, Frigidaire, and finally NCR, where he worked under Desch in the Electrical Research Department. He was involved in many projects there, including an electronic counter using gas thyratrons, tube design, standardized magnetic printing on bank checks, and magnetic tape. He sat on a committee that standardized characters. Mumma was also involved in the war effort at NCR, both under the NDRC and later the Navy which put the whole department onto the Ultra project in 1942.

Mumma talks about the automation trend that occurred at Frigidaire, and the resistance to the trend at the NCR factory. He also discusses the Ultra project, detailing Bombe design, security clearance, the Navy’s involvement in production, and Alan Turing's role. Joe Desch is threaded throughout the interview, with topics including his many patents and designs, his management style, his friendship with Mumma, and his breakdown in the 1940s.

For summary of the U.S. Navy and National Cash Register collaboration on development of World War II code-breaking machines, see Milestones: US Naval Computing Machine Laboratory, 1942-1945. See Vince Gulden Oral History, Roger Hull Oral History, Jack Kern Oral History, Edward de Laet Oral History, Donald Lowden Oral History, and Carl Rench Oral History for further discussion of Joe Desch, NCR, and the Bombe project.

About the Interview

ROBERT MUMMA: An Interview Conducted by Rik Nebeker, IEEE History Center, 15 September 1995

Interview # 274 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, Rutgers - the State University, 39 Union Street, New Brunswick, NJ 08901-8538 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:
Robert Mumma, an oral history conducted in 1995 by Rik Nebeker, IEEE History Center, Rutgers University, New Brunswick, NJ, USA.

Interview

Interview: Robert Mumma
Interviewer: Rik Nebeker
Location: Lebanon, Ohio (Mumma’s home)
Date: 15 September 1995

Childhood, family, and education

Nebeker:

If I could get you first, just for the record, to say where and when you were born.

Mumma:

I was born in Manila, Philippine Islands, in 1905, in fact July 28th to be exact.

Nebeker:

And your family?

Mumma:

My father and mother were missionaries for the then United Brethren Church and they were there on a five-year term and I was born in part of that five year term.

Nebeker:

I see.

Mumma:

They had two terms there. I left the Philippines in 1914 and started in the . . . and came back to the states and went through the school system.

Nebeker:

What education did you get?

Mumma:

Well, I was intensely interested in radio when it came out, of course, and I was still in high school [when] I built that type equipment. A neighbor of mine had a battery-operated radio that required a headset to listen. I built an amplifier for him and a loudspeaker which in those days which was rare. He insisted that I go to Purdue University. He was afraid that if I stayed with my father that I’d be a preacher (my father was a minister in Lewisburg, Ohio, by the way) and he was afraid that I’d be a preacher. He said, “you should not be a preacher” so he practically got me started at Purdue and got me a job with Spots Emrick. He was the director of the band. He gave me odd jobs to do to help defray my expenses and things like that.

Purdue U., electrical engineering

Nebeker:

You were in the Electrical Engineering program there?

Mumma:

Yes. But after two years there, somehow all I could see around Purdue were motors and streetcars and power lines and things like that and that wasn’t the type of Electrical Engineering I was interested in at all.

Nebeker:

And what year was this roughly?

Mumma:

Well, I started in Purdue in the fall of ’22, and I was there between ’22 and ’24.

Nebeker:

And they weren't teaching radio engineering?

Mumma:

No, but I got the basics. I had calculus, chemistry, Spanish and that sort of thing. I had mechanical Drawing and Machine shop work. I had a lot of good things, that’s for sure. After that I was real short of funds so I stayed out a year, and while I was out of school this year I worked at Frigidaire and made some money.

Nebeker:

That was in ‘24, roughly, that you left?

Mumma:

Yes, when I dropped out of Purdue.

Otterbein College; teaching

Mumma:

OK--So then I decided I was going to teach school. And I thought maybe if I taught physics and chemistry and that sort of thing--I would satisfy my cravings for electronics. So I went to Otterbein College up at Westerville, Ohio, and got my education courses so I could get a certificate to teach. So then I did teach school. I went down to Florida and I taught high school one year for the great sum of 800 dollars a year. I couldn’t live on that. I did get married then, and came up to Lewisburg, Ohio, where [I] got a job teaching Physics, Chemistry and Algebra for twice that much money.

Nebeker:

At Frigidaire?

Mumma:

No, at Lewisburg schools. I was still teaching.

Nebeker:

OK.

Mumma:

So I taught a couple of years there, and then I said, “No, this is not my cup of tea I’m not going to do this.”

Nebeker:

Yes.

Miami University, Ohio; high frequency AC theory

Mumma:

So I don’t know how this all happened, but that summer after the second year of teaching I went to Miami University at Oxford, Ohio and I talked to the head of the Electrical Engineering Department and he set up for me that summer a tutorial course in high frequency AC theory. He gave me assignments every morning. It wasn’t exactly electronics--it didn’t involve tubes so much as circuits.

And you see, I didn’t even know about resonance and impedance and these sorts of things yet, so we began--he gave me a real stiff course in that. He gave me a lot of equipment with which to work. I came out with a pretty good understanding of what was going on in a radio set when it was turned on. I didn’t know what turning these dials meant, you know. (laughter)

Nebeker:

Yes.

General Motors Radio; meeting Joe Desch

Mumma:

Also about that time General Motors Radio opened up in Dayton making household sets they sold like they did cars, in an agency, you know, not in a department store--that didn’t go real well. After a couple of years that was all over. But any way, to start with, I went back and got a job out there--

Nebeker:

I see.

Both:

At General Motors Radio.

Mumma:

Well, they sent me over to see a man named Joe Desch. He was in charge of the department making test equipment--designing and making test equipment for—

Nebeker:

For the radios.

Mumma:

--and boy, that was my meat, you know!

Nebeker:

Yes.

Mumma:

And so we sat down there and I built that stuff and designed it and all.

Nebeker:

Built the test equipment?

Mumma:

Yes, the test equipment. I must have impressed Joe Desch tremendously--I don’t know--and he’s been my friend ever since.

Nebeker:

Police radio operation

Mumma:

Well they folded and they went over to Anderson, Indiana. So I was out, so I had to look for a job. Well the only thing opening up was the police radio that was starting up in Dayton and they needed radio operators.

I had to have a first class radiotelephone license for that, so then I took the exam and passed that and came out with a first class radiotelephone license. So I worked for them for three years. While there I got interested in ham radio. Joe was an amateur radio operator.

So I studied when I was on the night shift and there wasn’t too much traffic going on. I used a tape unit in order to learn the code. I had to learn the code. That was my problem. I knew the theory all right but I needed the code. And I only had to use ten words a minute, which wasn’t too bad. So I got that and passed my examination for amateur radio
That was back in 1935 that I got my amateur radio license. And I was still a good friend of Joe Desch. I'd been there for a couple of years.

Nebeker:

That must have been a fairly early police radio system, the early thirties?

Mumma:

Yes, it was. They didn’t have you talk back from the cars at all. I announced and they had to report back from a box--a police call box, you know.

Nebeker:

I see.

Mumma:

I had to know if they got the call. That’s the way we operated it.

Nebeker:

I see.

Mumma:

We had a state Police Radio system you could check into. We knew their frequency and we could talk to them as well as talk to the cars.

Nebeker:

I see.

Mumma:

That’s the way it worked. I was made the swing shift operator. I’d take the place of the fellows who were off so I’d circulate from day work to afternoon work to night work. That was kind of a tough thing to do but it was all right.

Nebeker:

Wasn’t it in Detroit that they had the first police radio, I think?

Mumma:

They may have. I don’t think we were the first in the country at all.

Nebeker:

Yes, I think it was Detroit that was the leader in Police Radio.

Frigidaire

Automation; device development for moisture testing

Mumma:

So I had a lot of fun at that. But then while I was there, Joe came out to my house, Joe Desch, and he says, “You know,” he says “I’m going to get a job at Frigidaire. They’re really going to automate down there at Frigidaire.”

Nebeker:

In Dayton?

Mumma:

In Dayton. Yes, Frigidaire in Dayton. Moraine City. You came right by it coming down here. You saw the building I worked in. And so, I said “Yes, sure Joe, I'll go with you," so I went over there and I got this assignment of designing a device to check the vacuum inside of refrigeration units to see if there was any moisture in them, before they were charged with refrigerant and also then to charge the correct amount of refrigerant into the unit automatically. You punched in the size of the unit, which then sent the right amount of refrigerant into it and then pumped in the right amount of oil. This took less than a minute for this all to happen and the refrigeration unit would move on down the line. They were all on a conveyor.

Nebeker:

Oh, I see.

Mumma:

Before this time they had people with sight glasses doing this work by hand.

Nebeker:

So you’d--were there standard electrically controlled valves that you could use for that system?

Mumma:

No, we developed our own valves!

Nebeker:

So you developed your own.

Mumma:

Oh yes, we designed and built our own valves with sylphons that were operated by compressed air.

Nebeker:

I was wondering if that was an off-the-shelf item at that time.

Mumma:

No--oh no--not at that time. (laughter) We had to design the whole thing and make it work. No--things weren’t the way they are today--I’ll tell you that. The refrigeration units had a charging fitting over which our unit would fit and then seal with air pressure. In this unit was a wrench that was used to open a valve to check the pressure and then charge the unit and reseal it. The pressure in the unit had to be lower than 1 mm. of Mercury in order to charge the unit.

Nebeker:

I see. So that was the control of the charger?

Mumma:

Yes, to show that it was dry and that it had no moisture in the unit. These units had been pumped down to high vacuum. They had to retain this vacuum after passing the 250° f. Ovens for one hour.

Nebeker:

I see--so this was a test of the seal.

Mumma:

Yes, to see whether it leaked or not, see. Whether there was moisture or leaking we got that checked automatically.

Nebeker:

And if it wasn’t then you’d would charge it?

Mumma:

Yes, then it would charge, yes.

Nebeker:

I see.

Welding

Mumma:

It all worked just fine and, well, the production increased and that’s what happened. Well, then after that had been finished and we were going along and we were thinking about what else to do, I had done some work on automatic spot welders.

Nebeker:

Electric arc welding?

Mumma:

No, it was resistance welding.

Nebeker:

Resistance welding.

NCR job offer for Joe Desch

Mumma:

I went up to Detroit to see the people who built the welders to tell them what we wanted. That was my job at the department. Joe then got a call from Colonel Deeds--I don’t know if you’ve heard of him.

Nebeker:

Sure.

Mumma:

--and Chuck--Kettering--they came down to see him and Joe had a reputation of being one of the smartest fellows in electronics at that time and they said “Look. NCR is not going to make it staying with those mechanical machines. We’ve got to get some electronic equipment in there.”

Nebeker:

Yes.

Mumma:

And so you’re the man to do it. So they offered him a job of being the head of what he called the Electrical Research Department.

Electronic controls

Nebeker:

Yes, I’ve read about that. Could I just ask you--were there any electronic controls in these devices, either the charging device or the welding device that you did? At Frigidaire?

Mumma:

Yes, they were electrically controlled. There may have been a few tubes in there as amplifiers but nothing really electronic as we think of it today.

Nebeker:

Yes.

Mumma:

What we had to do to make it operate was to close a little contact to get the current to pull on a solenoid--things like that--but nothing that I’d consider to be electronics.

Radio frequency modulation

Nebeker:

But you knew electronics from your radio?

Mumma:

Oh yes. Oh I knew tubes--I'd built radios--all that kind of stuff by that time. And of course for Joe I’d done all that kind of work. Everything he built was full of all kinds of tubes in the test equipment we designed and built at General Motors Radio. In fact, at that time back at General Motors Radio, do you know we didn’t even have oscillographs, I mean oscilloscopes like now. So we built a big box with a screen on it made of tracing paper. And then inside we had a string galvanometer, you know what that is, with a mirror on it, which reflected a beam of light onto the tracing paper screen. And then we had a motor turning a condenser on a tuned circuit, which generated a frequency modulated signal. We were checking all the IF transformers. We wanted to see what the band pass was on all of them. So we’d hook this thing up to an IF transformer and start this thing going and generated the selectivity curve for the IF transformers.. And we’d see this trace on the screen. This thing worked like a charm. They used it on the production line. I designed and built this device. Of course Joe was impressed by it. (laughter)

Nebeker:

You may know of Harold Wheeler. He was another person who did radio test equipment. I’ve talked with him.

Mumma:

I see. No, I didn’t know him.

Nebeker:

But , Yes--that was an exciting field.

Mumma:

Yes, I know. An exciting time. I think he--Joe gave that device to UD. I don’t know if they still kept that machine or not. It was a monstrosity in a way, but in that day, the only thing they had the Braun tube. The Braun tube didn’t last very long. It was gas focused which destroyed the cathode after a short time. It’s a different world, right now.

Promotion at Frigidaire; automation and development

Nebeker:

So, uh, you were working at Frigidaire and you continued working there when Joe went to NCR?

Mumma:

Yes, for one more year after he left. They gave me Joe’s job and I was in charge of his department.

Nebeker:

I see.

Mumma:

I got promoted and that and got more money at Frigidaire.

Nebeker:

And was that sort of a development department at Frigidaire?

Mumma:

Yes, it was, yes--it was part of the maintenance department in a way but it was development.

Nebeker:

They were trying to improve production techniques?

Mumma:

Yes, we were trying to get them into automation--we automated the paint sprayers, you know, and we had conveyors carrying the units around so they all kept moving.

Nebeker:

Yes I know that period was one of this great transition in management.

Mumma:

Of course, we had to lay off people because of this automation, you see, and things like that happened at that time. But that was just the way it was going, you couldn’t stop it.

Trade publications; AIEE and IEEE

Nebeker:

Do you recall if--I’m interested in that automation trend at that time--do you recall if there were trade journals--industry journals--that you read at that time that told you about what other factories were doing?

Mumma:

Well, I must have seen them because I read all kind of magazines at that time.

Nebeker:

But none that stand out in your mind?

Mumma:

No, nothing stands out. Of course, I was a member of the AIEE, and I got their publication and I was also a member of the IRE.

Nebeker:

Oh you were in both the predecessor societies of IEEE.

Mumma:

Yes.

NCR

Electronic counter using gas thyrations

Nebeker:

I see. OK, and then--so you continued at Frigidaire--

Mumma:

Yes. I was promoted to the position Joe held with a raise in pay. I still was not satisfied, so after a year I asked Joe Desch for a job at NCR. He said OK, but I’ll have to give you the same as you received when you worked for me at Frigidaire.

Nebeker:

But you wanted to get into electronics?

Mumma:

Yes, that was it. So then he gave me an assignment right away. He asked me to develop an electronic counter using gas thyratrons.

Nebeker:

Right.

Mumma:

Gas thyratrons operated in a similar fashion to a Wynn-Williams trigger pair, in that once ignited they would remain conducting until extinguished by a positive pulse applied to the cathode.

Nebeker:

Wynn-Williams got you started?

Mumma:

Wynn-Williams, yes, that’s right. So I read up on that. So he said “Can’t we do something like this and something like this to make a counter?”

Nebeker:

Yes.

Mumma:

Well, about that time we ordered some gas tubes or thyratrons. They were big ones because that was all that were available at that time. I arranged them in a ring of 10 thyratrons, so that a conducting tube would prime the next tube to permit it to fire from a pulse on a common pulse line. Each conducting tube will extinguish any conducting tube.

Nebeker:

Was that the key characteristic of these that they’d stay on once triggered?

Mumma:

Yes, once you lit the gas tube, once you fired it, it would stay on. But then you could put a negative pulse on the plate or a positive pulse on the cathode to turn it off. See you could turn them on and turn them off with pulses. I made a counter using those big tubes.

Nebeker:

The large thyratrons?

Mumma:

I put them in banks of ten--and we had the two denominational orders in our first counter, so as to require a transfer pulse generating tube between denominational orders. This transfer tube was activated after a previously conducting “nine” tube caused a “zero” tube to conduct.

Tube design and production; patents

Nebeker:

Yes, that part is described, and then--was it Joe who had done tube design before or had you done any tube design?

Mumma:

No I hadn’t. He was a glass blower. He had done glass blowing. So he started out making small thyratrons. We had those big tubes you know--and we knew we had to do something better than that.

Nebeker:

It was just too expensive

Mumma:

No, the small hand-made thyratrons were more expensive. So we built these little miniature tubes, about as big around as your finger and brought the 5 necessary leads out of the tube base. We soldered these tubes on to a unit containing 11 tubes, or one denominational order.

Nebeker:

Yes.

Mumma:

And we made a bunch of these tubes and Joe did the first work on it. Then we hired a glass blower to make these tubes. We just made the tubes right there in our laboratory.

Nebeker:

I was wondering about how Joe got into tube design and...I’d read about this miniature thyratron that he designed...I guess in those days there weren’t enough standard tubes out there that you could get something to do your job?

Mumma:

That is right. Later on we built a 4 denominational order multiplier using commercial miniature thyratrons.

Nebeker:

Did Joe do other tube design that you know of later in his career?

Mumma:

Oh, Yes--a lot. In fact he and I collaborated on certain patents--on some of the patents. Here let me show you the pictures of some. [Looking at patents]

Nebeker:

They’re tube designs, is that right? I see the second one is an electron tube. Is that the miniature thyratron?

Mumma:

Yes. That might be it . . .

Nebeker:

Yes, I remember reading about this multi-anode tube.

Mumma:

See there’s a tube that he made, right there. And it had the extra electrode in here--we used it for the control element to set it whether it was primed or not. You see--if this wasn't at the right voltage this tube wouldn’t fire. And I think it was an effort to speed up the operation of these counting rings. Of course we controlled that somewhat by the pressure of the gas. We used just enough gas in them to make them stay on.

Nebeker:

Yes.

Mumma:

And we got up to ten thousand counts per second. That was about as fast as we could go with a gas tube.

Nebeker:

Yes. And what was the reason for preferring the gas tubes over the vacuum tubes?

Mumma:

Less tubes and simpler circuits. High vacuum tube counters required two tubes per digit.

Nebeker:

Yes.

Mumma:

See we didn’t have the miniature vacuum tubes yet. Now you can get a double tube in one envelope. They were not available at that time. In any case, gas tube counting rings were simpler and more compact. We did have one problem with gas tubes, in that they would self extinguish because of certain inductance and capacity in the circuit. In introduced a resistor in the circuit to solve the problem.

Nebeker:

So it would stay on.

Mumma:

Yes, it would stay on and it wouldn’t got into self-oscillation--we’ll say it that way.

Nebeker:

OK.

Mumma:

We put a loss in. We put a loss in the circuit so it wouldn’t do that.

Nebeker:

I’m just interested in this question of--you know--it seems remarkable to me that you and Joe would many times decide to design a tube rather than just working with what’s available.

Mumma:

Well, they didn’t have what we wanted . . . (laughter)

Nebeker:

And you had the capability there to turn out a tube.

Mumma:

Yes, so we had our own tube lab and we started making our own tubes.

Nebeker:

Yes.

Mumma:

This first calculator I made used all hand-made tubes.

Nebeker:

And you could get reliable tubes out of your little factory there?

Mumma:

Yes, we knew how to make them. We used a separate cathode from the filament heater. We had a heated cathode. We knew how to make these tubes and how to process them so that they would work. They had to go through a special process.

Nebeker:

So they’d emit well.

Mumma:

Yes. I think Joe was able to buy these cathodes already coated. I don’t think we had to spray the coat on them. I’m not sure about that.

Nebeker:

Jack Kern--he was telling me about spraying, spraying the cathodes.

Mumma:

The cathodes?

Nebeker:

Yes, at least some times he did that.

Mumma:

He worked with them some.

Automation, commercial applications

Nebeker:

OK. Now at the time when you first started with Joe, and I know this very important work that you and he were doing, was the thought that--ah, what was the commercial application in mind at that time?

Mumma:

Well, the unfortunate part was that NCR had a lot of equipment made to do various things and so we wanted to add electronics to the machines they had, and to automate them. One of the first things we got into was picking up information on a magnetic coating on the back of paper. We coated the back of the paper used in our accounting machines with iron oxide particles on which we could store data entered on the keyboard of the accounting machine as well as data resulting from its operation.

Nebeker:

Right.

Mumma:

When making several entries on the same sheet of paper, you could automatically pick up the last balance from the magnetic coating on the back of the paper. So you didn’t have to pick it up manually off of the paper and type it in. That’s the kind of thing we did.

Nebeker:

Right. Well that’s certainly--that’s electronics for a kind of data storage.

Mumma:

That’s right.

Nebeker:

But this calculating ability...

Mumma:

We never did get into the big computer business at that time.

Nebeker:

Well, the main business of NCR, of course, is the cash registers. They didn’t multiply, they just added, right?

Mumma:

Yes.

Nebeker:

And there wasn’t much need--I mean if they had sales tax, you’d have this little card on a table.

Nebeker:

Yes. The first thing we did on the cash register was to use a punch tape device that actually would punch a tape along with the amounts that were already on the printed tape. So you’d run that through a tape reader and that’s the way you kept your inventory just like you’d do with a scanner.

Nebeker:

I can see the value of that. What I’m really trying to get at is: the calculating ability--this electronic calculating--where did you think that would be applied?

Mumma:

Of course that came later on when the transistor showed up. And we wanted to get into that and do this calculating with computers...I didn’t have that expertise, and I guess Joe didn’t either, so we bought a company called Computer Research out in California.

Nebeker:

Yes.

Mumma:

And they had a computer mostly for mathematicians. It wasn’t designed for use in commercial work. But they adapted it, then, they became part of our company and we built the big 304. It was the first computer that was all transistorized.

Mechanical cash registers; electronic calculator

Nebeker:

Yes, that was important--I’m just wondering what was in it for NCR back before the war. Did they produce a desk calculator? NCR? You know, like a Marchand, or--did they have their--

Mumma:

No we used--they used Marchands for that part--they produced an adding machine. But nothing that multiplied and divided.

Nebeker:

And mechanical methods were fine for that.

Mumma:

And the cash registers were all mechanical except for the motors that Kettering had put in them.

Nebeker:

So it wasn’t that you were thinking of making electronic these adding machines or these cash registers?

Mumma:

No, that was--that was never in our assignment and it bothers me that we didn’t get those concepts earlier and could branch right into that but we didn’t...

Nebeker:

Well one might think that since mechanical multiplication and division is difficult that you had in mind “Let’s do an electronic calculator that can do all these things” but that wasn’t your objective in the beginning?

Mumma:

No. Not at the beginning. But it wasn’t very long after--I was there less than a year before we had our first counter that added and subtracted.

Nebeker:

Yes, that was very impressive.

Mumma:

They were big tubes, you see, and we only put three denominational orders in this calculator--because of the space involved. It was built on a vertical rack. Next we ran into IBM and we had a patent infringement. We went to court and we won on that one. We went to New York for the hearing. Later I built a machine that would add, subtract and multiply. That original model is still in storage at NCR. It was used in a patent infringement case between Minneapolis Honeywell and Sperry Rand. That machine is still around. I don’t know if you’ve seen it or not--

Nebeker:

No, I haven’t.

Mumma:

It’s a huge machine, you know. It doesn’t do as much as the little hand-held calculators used today.

Nebeker:

Right.

Mumma:

But at any rate, we made it using the little miniature gas tubes and we got them to run at about ten thousand pulses per second, but even at that slow speed we could do accounting in just a very short time. But for anybody to see the result of the calculation--you just couldn’t look at the glow on the tubes--so we had to put an indicator on the machine. It would take longer to hunt the correct position than it took to make the calculation.

Nebeker:

Yes.

Mumma:

So we were able to multiply four digit numbers by four digit numbers and read the answer on the machine.

Patents

Nebeker:

And all the while whenever you did design a new tube or get one of these circuits--accounting circuits to work you’d file for a patent?

Mumma:

Oh yes.

Nebeker:

Right. And was the thought at the time that these would be incorporated in products by NCR or other people?

Mumma:

Well, yes. We did make a multiplier attachment for our accounting machines based on this multiplier. I forget the number for it now. It would make automatic extensions, printing out the product in the correct location on the document.

Multipliers; paper tape punch

Nebeker:

Was that a large part of NCR’s business, the accounting machines?

Mumma:

Oh yes.

Nebeker:

And I can see that having a multiplier would be a great improvement.

Mumma:

It was too large to use on a cash register. We finally ended up automating the cash register by adding a paper tape punch, to record the data. These tapes could be read later, to use in generating an inventory of products sold.

Nebeker:

Yes. Now that was--on the tape was just the sales price, right? Not inventory Control?

Mumma:

Yes.

Nebeker:

Wasn’t there some ticket system for inventory control early on? Machine readable ticket?

Mumma:

Punch card?

Nebeker:

No I’m not thinking of--my memory on this is very hazy but I thought there was a ticket system maybe with magnetic characters.

Mumma:

No, if that came out I didn’t know about it.

Nebeker:

OK. So one application then of a multiplier would be for these accounting machines?

Mumma:

Yes.

Resistance to automation

Nebeker:

Did you have the feeling that NCR was really interested in going electronic with some of these devices?

Mumma:

Well, not yet. We were kind of ostracized. The factory was not ready yet for automation.

Nebeker:

This is Joe’s group of the twenty or so--

Mumma:

Yes, and we were kind of fought by the factory people. They didn’t want anything to happen to their present program. (laughter)

Nebeker:

Yes, well they had a big business going.

Mumma:

Yes, I know (laughter). The only person interested in us was Colonel Deeds. He’d come down and see us but we never got an assignment to do something for a particular machine.

Nebeker:

I see.

Mumma:

We were just initiating things on our own.

Nebeker:

I see. That’s very interesting. Did you ever have a case where you had something that you thought could be commercialized, and tried to do it and...?

NCR projects, WWII

Shell velocity counter ====

Mumma:

Well, see, our development work was stopped by the war effort. The NDRC group got hold of us and we got all types of assignments from them. We built a device to measure the velocity of a shell as it left a cannon.

Nebeker:

Yes, I’ve read about that. There you had to get very high speed.

Mumma:

Yes, [we] developed a counter that would operate at one million counts per second.

Nebeker:

That was with a vacuum tube?

Mumma:

Well we used vacuum tubes in the first denominational order of the counter. We had developed gas tubes to count at 100,000 pulses/second.

Nebeker:

I see, but then the higher ones didn’t need to be--

Mumma:

We got the higher ones after that.

Nebeker:

I see. And why not all of them vacuum because the other ones were expensive--oh, they were smaller--

Mumma:

Well it took too much space, you see. Our first vacuum tubes--even the metal tubes we had by that time--took up too much space.

Nebeker:

I see--mainly a matter of space--

Mumma:

Yes, the thing just got out of hand with the size of those big tubes.

Nebeker:

Yes.

Aberdeen; counting signal pulses

Mumma:

So we built the whole box with our high vacuum tubes in the first denominational order with gas tubes in the rest of the counter and took it over to Aberdeen.

Nebeker:

Aberdeen?

Mumma:

Yes, Aberdeen.

Nebeker:

In that one there was a crystal that was generating the high frequencies, right?

Mumma:

Yes, that’s right. Oh yes.

Nebeker:

Was that a difficult task?

Mumma:

No, it was just an ordinary oscillator.

Nebeker:

That was something that was well understood at the time?

Mumma:

Oh yes.

Nebeker:

And so it was a matter of counting the pulses between these two signals.

Mumma:

And we had a gate, you see, fixed up here that operated very fast and when it went through the first loop up here we got a signal and that opened the gate and the pulses started flowing out.

Nebeker:

And you’re counting away! (laughter)

Mumma:

Oh yes, then it went to the next loop where the counter was stopped then we could read the output of the counter, which read in microseconds.

Nebeker:

And this was an application of the kind of counting circuit that you’d been working on before?

Mumma:

Yes. So on that distance we’d know how many microseconds it took.

Nebeker:

Yes.

Mumma:

After that we’d know the velocity of that shell.

Communications system

Nebeker:

Were there any other applications of that kind of a high speed counting circuit?

Mumma:

Oh yes, we had a communications system we worked on too. I got a patent on that. We would send out a message in code groups. We did that by numbers, of course, five digit numbers in code groups that could be interpreted. And then we had storage tubes--Joe made some gas storage tubes. There were 10 positions in the tube. We stored the information we received in these gas storage tubes. Each storage tube represented a denominational order. We sent messages in five number groups. We transmitted our messages on a frequency of 7 megahertz. Our receiver would pick up and count these pulses putting them in order in a whole bank of these storage tubes so that we could read the number out.

Nebeker:

Yes.

Mumma:

We actually demonstrated that to the NDRC but I never heard any more about it. There must have been a better, easier way to do it.

Early electronic calculating

Nebeker:

Was there any effort to interest other people in these patents when--

Mumma:

No not at that time, no. That I knew about anyway.

Nebeker:

But I suppose that happens, perhaps, by itself once the patent is issued and other people look at it and they could approach you if there were interested but . . .

Mumma:

See ENIAC was being worked on by this time too. I do not know when it came out.

Nebeker:

It was finished just after the war. ‘46.

Mumma:

And that was all great big tubes--oh that was a massive--I would have liked to see that thing. It had a big special air conditioner to cool it. It was done by the brute force method too, if you know what I mean.

Nebeker:

Right.

Mumma:

The transistor changed all that, I tell you.

Nebeker:

What were the major hurdles in getting this early electronic calculating to work--the computing? Was it the tubes themselves--to get them to have the right characteristics?

Mumma:

The use of tubes was too bulky for practical use. We never had anything good enough, let’s say, to get the company willing to drop what they were doing and get into this sort of thing. That didn’t come ‘til after the war because this NDRC thing took up all our time when that started and it wasn’t long after that that they heard about us over in England and Alan Turing came over. They wanted us to build this big machine we talked about.

Joe Desch as manager and collaborator

Nebeker:

Yes, I certainly have some questions about that. Maybe this is a good point, though, to get you to say something about Joe Desch: how he influenced you and what sorts of things either in design or management style you learned from him.

Mumma:

Well, he’s the one who told us what he wanted us to do. That was his function. And he was interested in how it was done, too. He was a good electrical engineer.

Nebeker:

So he followed all of your work?

Mumma:

Oh, very closely. And he got patents. You’ll find we’re co-inventors on quite a few of these things. Joe Desch and I were co-inventors if you go through these patents you’ll find he’s on many of them. We had other people co-invent. I wasn’t the sole inventor on many patents. Here’s one there--that’s one of our counting circuits. I’m on that patent along with Joe.

Nebeker:

Yes, it says right here.

Mumma:

These are the little gas tubes, this is the counting ring. Oh this one here was a--we made some of these in groups of five, but I think this is kind of a trigger pair. These tubes operated in a ring with opposite tubes connected for a trigger operation. We went through all kinds of combinations and got patents on them because we didn’t know which one would be practical. Oh here’s that storage tube we used for that communicating system here. We had the ten positions and by putting a charge on one position we could make it glow. Each position represented a digit in a denominational order.

Nebeker:

And once on it would stay on?

Mumma:

Yes, then you could read it out with some mechanical device and we made that too in our lab.

Nebeker:

What was Joe’s management style? Was he somebody who sort of micromanaged?

Mumma:

Oh, no, he did not.

Nebeker:

He gave you more independence?

Mumma:

He was one of the boys and he’d told us what he wanted us to do and if he had any ideas he would suggest that, too, but he wasn’t a man that made you do it “my way or not”. He wasn’t that difficult. He was a wonderful man to work for. He and I got along real well, and the other fellows, too, that worked for him.

Navy Ultra project, Bombes

Nebeker:

And then in the beginning of ‘42 suddenly while you were doing NDRC work the Navy came in and suddenly you’re on this Ultra project.

Mumma:

We had the Navy come in and that was the end of the NDRC work.

Nebeker:

How did you feel about changing, you know, or suddenly going onto that project?

Mumma:

Well, we didn’t like it very well, I didn’t--but anyway, we were at war. And they of course came and gave us quite a pitch on we had to save the country; we had to win the war! And if we did this we could win the war! And we got quite a sales pitch I can tell you that.

Nebeker:

Did they tell you exactly what it was you’d be doing?

Mumma:

Oh yes, at least I was told.

Nebeker:

And the application of it?

Mumma:

Yes, at least I knew about the Enigmas and, by God, of course we were sworn to secrecy.

Nebeker:

How many of you were in on it to that extent?

Mumma:

Our whole department was in on it. The Electrical Research Department. I don’t know how many knew all of the details that [I] was told.

Nebeker:

But Jack was telling me that he wasn’t ever told what these devices really were.

Mumma:

Well, that may be true. Maybe he just knew parts of it. I knew all about it because I was working with Joe and helped design the original machine.

Nebeker:

I think they were careful that only a few people knew the whole thing.

Mumma:

I’m not sure--we had our designing section, the drawing boards, we brought them over to the unit. A fellow named Bob Goebel was in charge of that. And we told them what they needed to know to design the machine, even though they may not have known the end use.

Nebeker:

Well, I’ve read in Colin Burke’s book, and I guess some other places, about that work then on those machines and I guess there was a hope early on that electronic means could do the job.

Mumma:

Yes, but they were in a hurry and we did not have time to develop more exotic electronic devices.

Nebeker:

Yes. Were you and Joe in constant touch in deciding, for example, that it won’t work to do it electronically?

Mumma:

Well, yes, as far as I’m concerned we never even did consider it. When these Navy people came in to talk to us--it was all adding up to rotating commutators. I did build a little electronic device to remember the position of the commutators when the “hit” occurred. We stopped the machine as quickly as we could and rolled the commutators back at slow speed to the exact position where the “hit” occurred. I used some of my previous knowledge of electronics to design this electronic memory.

Nebeker:

I see, an electronic storage device.

Mumma:

Yes, we caused this electronic memory to record the exact position of the commutator wheels where the “hit” occurred. As I mentioned above, we rolled the commutators back to where the “hit” occurred.

Nebeker:

I see, and was that used?

Mumma:

Oh yes, every Bombe had that block of tubes for the control of the final position of all the commutators.

Nebeker:

I see.

Mumma:

And it wasn’t all mechanical. Because anything mechanical is heavy, with a lot of inertia and you just don’t stop instantly. There is a lot of inertia in a big 5 Horsepower motor and the equipment that it was driving. We couldn’t stop it instantly.

Nebeker:

Yes, sure. I see. That was for registering the hit locations--

Mumma:

Yes, I might take this opportunity to describe, as I remember it, how the Bombe was given its final test. The check out of each Bombe was performed in my presence by Navy personnel. The WAVES mounted the commutators on the Bombe spindles according to a plan developed by Alan Turing, the mathematician, who invented the Bombe and developed the theory for its operation.

Naval supervision, both male and female, supervised the WAVES who installed the commutators on the Bombe spindles and plugged the cables into the plug boards on the end of the Bombe. Electricity had to flow in sequence through contacts on each commutator and through the wires in each cable. As the commutators rotated, the electricity switched from wire to wire. This high speed switching pattern was developed by Alan Turing to identify which four commutators were used in the Enigma code machines that day.

When all of these wheels were lined up on the Bombe to permit electricity to flow through all of the commutators for the instant that “hit” was registered, the Bombe was stopped and returned to the “hit” position.

From the position and location of the commutators on the Bombe when the “hit” occurred they were able to determine the commutator settings for that day in the Enigma machines. It was my understanding that the WAVES manually recorded the commutator identifications, locations and positions at that time. A printer for this function may have been added later in Washington. At least this is my understanding of how the Bombe worked.

The same commutators and settings were used for testing all of the Bombes for any error in the wiring of the Bombes.
The WAVES can give you a better description of how the “hits” were recorded and the Bombes were tested.

Nebeker:

What were other challenges there? Registering this hit position was certainly one. What were the other real difficult tasks?

Mumma:

Well, one of our big problems was getting a brush with low enough resistance to permit the electricity to flow through hundreds of brushes without too much loss of power. As a result we used a mixture of carbon and silver in the brushes.

Another problem was wiring the hundreds of commutators used on the Bombes. The WAVES had to solder both ends of 52 wires to the 104 terminals on each commutator. The pattern of wiring was the same as the original wiring on the 8 commutators used on the Enigma machines. I didn’t have anything to do with it except that I ordered the wire--we ordered 28 different color codes--I didn’t dare say 26 color codes--I thought that would give something away (laughter). We had an awful time making these commutators.

We did have to put 26 segments in each ring of the commutator. We ordered these commutators with a solid ring. Then we had in-house people slot the commutators. Again we though 26 segments would be a give-away of what we were up to.

Nebeker:

I see.

Mumma:

We did use 26 terminals on the back of the commutator rings as terminals to connect the wires.

Nebeker:

I see, so getting good low resistance contacts was a major problem.

Mumma:

Yes, as I mentioned before, we used silver in the carbon brushes, instead of copper, which we’d normally use. It dropped the resistance of the brushes.

Nebeker:

Was this something that NCR had experience with, these wheels?

Mumma:

Yes, but I never knew them to use silver in their brushes.

Nebeker:

This was all new?

Mumma:

We also had to use diodes to isolate these circuits to keep the electricity from backing up in any of the circuits. Sylvania tube Company made special glass tubes for us with 4 diodes in each tube. These were used in the Bombe to isolate the circuits. You can see this panel of diodes on the rear end of the Bombe.

Nebeker:

Yes.

Nebeker:

What, if any particularly difficult technical problems do you recall from the Bombes?

Mumma:

Well, another big problem we had was the matter of these line shafts driving the commutators. We used a spiral beveled gear to get the 90° take-off on the line shaft. They were specially ground to a minimum backlash, because they had to read the same position going forward and backward because, when we went back to find that “hit” we were going backwards.

Any play in those gears and you might miss a “hit.” So those were all specially ground. They paid a good price for these gears.

Nebeker:

That sounds like something maybe NCR had experience with.

Mumma:

Well, yes, they knew about gearing and so forth. But we had a company up in New York state make them for us--make those spiral beveled gears. I forget the name of that particular company.

NCR electronics applications

Nebeker:

Was this work on these machines--how valuable was it to NCR?

Mumma:

I don’t think it really helped them much.

Nebeker:

There weren’t ways they could draw on it?

Mumma:

No, they really didn’t. We really did not get into electronics at NCR until we took over Computer Research in California. Before that we had a paper tape punch and a paper tape reader. After NCR took over Computer Research they got into electronics in earnest. Joe Desch was very influential in talking Colonel Deeds and S. C. Allyn into making this acquisition. Immediately after taking over Computer Research Corporation, NCR contracted with General Electric Company in Phoenix to build our first Electronic Computer, the 304 model, with Computer Research Corporation to be responsible for Engineering and Design. Both NCR Dayton and Computer Research were responsible for making their assigned peripherals.

I became Head of Peripheral Equipment Engineering Department in Dayton. At first, we purchased the magnetic tape data storage units. Later we developed a high speed printer for use with this computer. Later, with the help of Pitney Bowes Corporation, we developed a check sorter for banks, which read magnetic characters printed along the bottom of all bank checks. I was involved in standardizing this magnetic printing on bank checks.

Navy project stress, security

Nebeker:

If we could return to the war work. I understand that Joe had a breakdown in--maybe it was ‘44?

Mumma:

Yes, he did in 1945.

Nebeker:

I can imagine there was great pressure on all of you.

Mumma:

I couldn’t have lived the way he lived. During that war work, Commander Meader, or Captain Meader I guess, he came to be--lived in his home. See Joe had some cousins and relatives over in Germany, and they were so sure that something might leak at this point (I don’t know why, Joe wouldn’t have said a word--Joe was just super on that--he put the fear of God into us, I tell you), Meader stayed in his home the whole time.

Nebeker:

Was that for that reason, for security?

Mumma:

Yes, that’s what I was told.

Nebeker:

My goodness.

Mumma:

And Joe hated him with a passion.

Nebeker:

And he was living in his house? My goodness.

Mumma:

But apparently he could not do anything about it.

Nebeker:

My goodness.

Mumma:

And it was terrible. But you know, just a little aside here, I was told that after the war, Meader spilled his “guts” to a security man someplace along the line on an airplane and he lost his commission in the Navy right away. He may have been sent to jail. I can’t vouch for this story, but Joe told me the story with great glee.

Nebeker:

My goodness, I didn’t know that. So that certainly didn’t help things.

Nebeker:

That was after the war, of course.

Nebeker:

No, I mean having Meader living in his house.

Mumma:

Oh he was under a terrible strain. So he--that was about the time we were closing down, he had to quit. And so then he said ‘You take over, you run the operation.’ And so most of my job was getting the operation shut down.

Nebeker:

Oh, this was right at the end.

Mumma:

Pretty close to the end. Maybe we made a few machines after that, but it was pretty close to the end before he finally broke and had to go.

Nebeker:

Was there a particular event that precipitated it?

Mumma:

Oh, I don’t know. I think things just got worse and worse. And he probably went to a doctor and they said ‘Look, you’ve got to give up this job or you aren’t going to live’ you know. His daughter could probably tell you more about this situation.

Nebeker:

How were you coping with that?

Mumma:

Well, it wasn’t too bad. We worked seven days a week, you know, and pretty long hours. One Easter Sunday, I know, I laid off and didn’t come to work and Joe was awful mad about it! (laughter)

Nebeker:

Is that right? (laughter)

Mumma:

But it was a strain, no doubt about that.

Nebeker:

Was there a lot of pressure from the Navy?

Mumma:

Oh yes, they were all over us, of course. And, see, they supervised the testing. I was there representing NCR but their WAVES came down and put the wheels on the machines to test them and they plugged the machines up to get this hit, you know . . . they personally saw that every machine was right before they went out and then they loaded them on baggage cars and took them at night up to Washington, D. C., under Marine guard.

Nebeker:

How was it for you then, you know, as a relatively young person to suddenly be in charge of this huge operation?

Mumma:

(Laughter) I don’t know. I was so--I just took it in my stride. I didn’t worry about it. See, Joe had it pretty well set up and everybody knew what they were supposed to do. I didn’t have much managing to do at that time.. We weren’t hiring anybody anymore and were firing or sending people back to their old jobs.

Nebeker:

Yes, but you were still trying to meet some Navy demands.

Mumma:

Things were pretty much routine by this time. My biggest job was disposing of all the material that would reveal what we had been doing.

Nebeker:

Did you travel back to Washington, D.C.?

Mumma:

At that time?

Nebeker:

During the war?

Mumma:

No I didn’t. Joe did all the traveling.

Nebeker:

You had, of course, to get a security clearance at the beginning of the project.

Mumma:

I sure did.

Nebeker:

Was that an ordeal?

Mumma:

Nobody seemed to complain about it. My folks, I guess, were clean and all that and I never heard any more. Joe took care of getting it all done. We filled out the forms and I guess he told us we got the security clearance. He had no more trouble than anyone else did.

Alan Turing

Nebeker:

Yes, that’s interesting. I hadn’t heard about that. What about contact with...you were seeing Navy people all the time here in Dayton. Did you meet any people from England?

Mumma:

Yes, this--oh, I mentioned his name a bit ago, this mathematician--Alan Turing, yes, he was over with us. He came in and told us what we wanted to do and how to do it. I worked with him on that.

Nebeker:

Do you recall when that was?

Mumma:

Oh it was very early in the project. Very early. While we were designing the machines. Before production.

Nebeker:

I see. He in person talked with you.

Mumma:

Oh yes. He was over here. He said, “You’ve got to do this, you’ve got to do this.” He controlled the design more than anyone else did.

Nebeker:

How long was he in Dayton, do you remember?

Mumma:

Oh, months I guess, I don’t know.

Nebeker:

Is that right? What was your recollection of him?

Mumma:

Well, he was a nice fellow. I don’t know. I enjoyed talking to him.

Nebeker:

He seemed approachable?

Mumma:

Oh yes. He knew what he wanted and we weren’t in a position to argue with him. We just said we can do this, we can’t do it, you know. See, he’d done something similar with IBM equipment over there in England. But when the Germans went to the fourth wheel in the Enigma, he couldn’t get it done quickly enough.

Nebeker:

Did he return a second time?

Mumma:

I think he came back. I forget. I think he came back to visit.

Nebeker:

Did you have other visitors from England, do you recall?

Mumma:

No I don’t. We had quite a few Navy people come down, Captain Wenger and some others.

Nebeker:

I see. What’s your recollection of Wenger?

Mumma:

Well, all right. He talked to Joe mostly, but I had met him and he looked over the machines, you know. He watched them operate. And also, during the design period, why he’d see what we were doing and how we were supposed to do it.

Enigma project as departure from pre-war work

Nebeker:

So as far as your work at NCR and Joe’s work, this Enigma matter was a digression. You just got pulled into that. It drew a little bit on your electronic work in an important way but it didn’t contribute very much to it.

Mumma:

No, we didn’t add anything new that we didn’t know before; you can say it that way. We didn’t design a computer to do it like we should have.

Nebeker:

Right. If you’d been on another project where you were doing some computer work for the Navy or someone...

Mumma:

Yes, because I at that time didn’t have a concept of how these modern computers work, actually.

Nebeker:

Well, almost no one did (laughter) at that time. Have you thought about--that maybe it was a shame that when you’d gotten so far with electronic calculating before the war that you didn’t get on a calculating project?

Mumma:

Well, I should have but I guess the influence was to try to get them to automate their present products. We did attach our electronic multiplier to some of their accounting machines.

Nebeker:

On accounting machines, Yes. What about--

Mumma:

See we did that on another job too. We picked up the old balance on another type machine we made for banks. I think that came along when the check sorter thing came along. We obtained a magnetic character reader from this school out in California.

Nebeker:

Stanford?

Mumma:

Stanford, yes, they had designed a reader that could read magnetic characters and I was on a committee then to develop the shape of those characters.

Magnetic character reader; standardization

Nebeker:

Oh I’m glad you mentioned--I wanted to ask about that--this standardization of those characters.

Mumma:

Yes, we had the federal banks and the federal people in on it too, you know. And we worked out this printing you’ve seen along the bottoms of checks. You know what they look like.

Nebeker:

Yes, and you were on the committee that arrived at that?

Mumma:

And IBM had a reader and they had a multiple head job to scan the characters, so they, I don’t think, they were as critical of these shapes as we were. We had a single head reading these characters and had to look for a different pattern for each character. We matched patterns.

Nebeker:

So they had to be quite distinct patterns that you were picking up.

Mumma:

Yes, instead IBM had individual pulses when their heads passed over these characters. But we did ours with a single head.

Nebeker:

That was something NCR already had?

Mumma:

We got that from Stanford.

Nebeker:

I see. That reading head.

Mumma:

That reading head and that whole bit. We bought that or got that from Stanford--they had the patents on it.

Nebeker:

OK, was there a real confrontation in that standardization effort between IBM and maybe...

Mumma:

Not too much.

Nebeker:

People were reasonable?

Mumma:

Oh Yes, we had a nice relationship.

Nebeker:

I’ve heard that from other people who have worked on standardization. That companies that are fiercely competitive, at least at the sales end, that the engineers can work together.

Mumma:

Oh yes, there was no problem. They held out for what they needed for their equipment. They knew what their machines would do, you know, and we knew what ours would do and we had to come to a compromise. Engineers work with engineers.

Nebeker:

It went fairly smoothly?

Mumma:

Yes, very nice.

Nebeker:

And in retrospect now, was that well chosen?

Mumma:

Yes, it seems to me. Of course, we still have it in the same pattern--never changed it or anything. But they had one little problem. And that problem was that we didn’t have a way to read the amount on the check. When you write a check you do not print the amount in magnetic ink. We need to get that amount on the check. So . . . what to do, what to do? So, I developed a magnetic ribbon for printing the amount on the checks while they go through the proving operation. I invented that ribbon. I got the patent for it--I think that’s the one patent I got on which NCR made any money. And so we developed the ribbon and sold them.

Nebeker:

And you knew of some magnetic ink that work?

Mumma:

Well, we just made our own. We just mixed the magnetic oxide in with the waxes and all that. In the patent it tells you all about the different waxes that were used and how they were deposited on a paper ribbon, by the way, not on cloth ribbon. This was a one-time ribbon. The ribbon kept advancing, you never hit the same spot twice, you know. So that you always got a clean letter every time.

Nebeker:

Sure.

Mumma:

So we made a proof machine that used these ribbons. So we put that ribbon on a proof machine--it was part of the process to get the “amount” printed in magnetic characters on a check.

Automation and NCR production techniques

Nebeker:

What about production techniques at NCR? Were they moving toward automation in producing these mechanical cash registers?

Mumma:

Well, somewhat. They had a lot of hand assembly though.

Nebeker:

Were you involved at NCR the way you were at Frigidaire, trying to automate?

Mumma:

No, I was not in production. I developed a product. I was in what we called Electrical Research at first, and then I got into Engineering afterwards.

Nebeker:

But they didn’t call on that group to try to improve production techniques?

Mumma:

No, they didn’t--in fact that didn’t happen until Anderson came and he threw everything out and started over again.

Nebeker:

Do you think that you could have made contributions to production?

Mumma:

Oh we could have, I guess. I was so busy doing what I had to do that I didn’t think about it.

Nebeker:

Sure, but I was just wondering, in looking back now were there ways of automating some of that?

Mumma:

Sure. Yes, we should have started earlier and gotten into a different type of production, but it would have been awful hard to move that company over to something like that.

Nebeker:

Yes, they had such a momentum in their business.

Mumma:

We did finally--we made high speed printers when we started making the main frame computer. We had to get peripherals for it of course. We got involved in that pretty heavily. I was brought in on that pretty deeply. But up until we took over this company, there was no movement.

Nebeker:

Did you resent that at the time? Did you think NCR--

Mumma:

No...

Nebeker:

Or you accepted--

Mumma:

Yes...

Nebeker:

You were busy? (laughter)

Mumma:

See, I have to admit to another thing. I wasn’t strong on transistors either at that time.

Nebeker:

I’m sorry?

Mumma:

I was not strong on transistors. That’s a whole new way to design circuits. The circuits were designed mathematically. I think they used Boolean algebra.

Nebeker:

Different game from tube design.

Mumma:

Yes, and I knew tubes back and forth but I sure didn’t know the transistor circuitry that well. So we were bound to pick up another group that was well grounded in mathematical design. That’s when I got delegated back to making peripheral equipment.

NCR development philosophy; application of magnetic tape

Nebeker:

Yes--OK. Let me just propose this as a hypothetical view and get your reaction to it: that NCR management in the late ‘30s, ‘40s and maybe later--they were interested in electronic techniques but wanted to protect themselves by getting patents.

Mumma:

Yes, that’s right. You’re pretty much right.

Nebeker:

You think that’s a fair view?

Mumma:

That was the start of it anyway. That was at the insistence of Colonel. Deeds and Kettering--they’ve got to get control of this thing. But--and then when they did get into it, it was attachments to their existing machines, you see. Everything we made was an attachment to their machine. So you see, we never got this other concept until we got into Computer Research. That’s what I’ve felt bad about it ever since but there’s nothing I could do about it.

Nebeker:

Well, that’s understandable because they had machines that worked and--

Mumma:

But, we didn’t start from where we should have and anyway, I’ve felt bad about it but a that’s what happened.

Nebeker:

I’ve also wondered about why magnetic tape was so slow to be adopted. You know you mentioned the cash registers that used punch tape. I mean, the Germans had good magnetic tape in the ‘30s and after the war it was coming to be used in this country.

Mumma:

Well, we used it for music. (pause) Yes, but that was for music. I never got the concept of it until those magnetic tape units came along for storage tapes on the computers, you know. When they put this bar code on it.

Nebeker:

So they hadn’t been used digitally at that point.

Mumma:

No, I hadn’t seen any of them. I never got the concept myself.

Nebeker:

It seems to me, well, of course with the benefit of hindsight one can see that the magnetic tape--

Mumma:

Oh, obviously. I know--

Nebeker:

--would be better than the punch tape.

Mumma:

(laughter) I know, we can think of it now. Of course they used a plastic tape when they went to the magnetic tapes too, instead of the paper but that’s--

Comments on photographs lent by Debbie Anderson (Joe Desch’s daughter)

Nebeker:-

--essentially the same, yes. I just want to check my notes to see if I’ve covered the things I wanted to ask about. I think I have covered the questions that I’ve written down. These were some photographs that Debbie lent me.

Mumma:

Oh I see. She’s a wonderful girl, I’ll tell you.

Nebeker:

She said you might recognize some of these things.

Mumma:

Where’d they get those?

Nebeker:

They’re things that her father had left.

Mumma:

I’m not sure . . .

Nebeker:

We’ll go to the next one.

Mumma:

That’s his home of course here. Then he got pictures on an oscilloscope.

Nebeker:

Yes, a lot more of these oscilloscope tracings here.

Mumma:

I guess he must have enjoyed it and gotten his camera and taken pictures of the pattern.

Nebeker:

Who is that?

Mumma:

Oh my Lord. Gosh, I don’t know.

Nebeker:

Does this person look familiar here?

Mumma:

Yes, he does. He was one of the first engineers we had. I forget who it is.

Nebeker:

OK.

Mumma:

De Rosa!

Nebeker:

De Rosa?

Mumma:

De Rosa.

Nebeker:

I remember reading about him.

Mumma:

De Rosa, it finally came to me. I have the worst trouble with my memory. By the way here’s some of our first counting rings in there too. I guess you saw those. Here is our high speed vacuum tube counting ring right there we used on the machine that measured muzzle velocity of shells.

Nebeker:

--for the Aberdeen machine?

Mumma:

Yes, for the Aberdeen machine. And here are the denominational orders of the gas tubes.

Nebeker:

So this was the lowest order.

Mumma:

Yes that’s the lowest order here. Then here’s a bank of three denominational orders again. There are all gas tubes here. We put them on racks like that. Each denominational order had ten tubes for counting. The eleventh tube is a self-extinguishing transfer tube that fires when the zero tube fires and sends a pulse to the next higher denominational order in the adder. This permits the use of more than one denominational order in an adder.

Nebeker:

I see. Was this simply a unique machine or did they ask for more of these counting rings.

Mumma:

Well as far as the patent model why that was unique but then we did make a second machine for Aberdeen, which they sent to another manufacturer to make as many machines as they wanted machine, almost in total the way it came, and put it in a box and they assigned a number and they sold these boxes--

Nebeker:

Oh, in numbers? In large numbers? Who produced them?

Mumma:

I do not remember the name of the company. It was, I don’t know, I maybe made 5 or 6 hundred of them. [We are now talking about an electronic multiplier that NCR made for use on the Class 31 machines to make automatic extensions. Vince Gulden made this machine after the war. RM]

Nebeker:

Oh, is that right?

Mumma:

They were tied to what we called our 31 [maybe 390 machine--see Vince. RM] machine. It’s a typewriter-type of machine with totals on it that you could add and subtract with it. And this had the multiplier function so that they could make extensions. With this device they could make automatic extensions just right away.

Nebeker:

Yes. OK.

Mumma:

I forget they had a number assigned to it but they sold quite a few of them. The multiplier was in a big box. It’s not as compact as it is today. I haven’t seen a Louis De Rosa picture in a long time. [This is Vince’s machine. I don’t know how De Rosa enters into conversation except that we had his picture. RM]

Engineers on NCR wartime projects

Nebeker:

Are there any other outstanding engineers you recall from that wartime group?

Mumma:

Oh, from that wartime group? Of course, Vince Gulden worked with us there. And a mechanical engineer, Bob Goebel. Of course he did the design work on the Bombe for the Navy. And his head draftsman was Ed Carey. He's still alive but he’s up in his nineties. He’s older than I am.

Nebeker:

What do you recall of Vince’s work? I’m going to be talking to him this afternoon.

Mumma:

Well he’s an electronics man. He’s probably stronger in electronics than even I am. He’s very good. In fact, he is the one who modified this multiplier I’m talking about and then attached it to the accounting machine. He made all the applications for that, and he can tell you a lot more about it than I can.

Nebeker:

OK, I’ll ask him about that. Anything else you’d like to comment on?

Joe Desch

Mumma:

I’ve been awful lucky the way things came in my life, that’s all I can say.

Nebeker:

Well thank you very much.

Mumma:

Some people didn’t get the breaks that I got. Well, Joe Desch was the break that I had. He was my friend all the way through.

Nebeker:

And you remained close to him?

Mumma:

Very close. I went to his house to see him after we moved down here. One day I knocked on the door and nothing happened. Nobody came to the door. I was there the day he died. He had a stroke previously, but he recovered from that. He was in a rest home for several years. But they fed him a blood thinner so he got back to normal again and went home. He lived by himself there. So, I’d visit him and we kept in touch. I also talked to him on the radio every week. He had a transmitter and receiver and we’d talk.

Nebeker:

Oh, you talked by radio?

Mumma:

Yes, so we kept in very close contact.

Nebeker:

Do you still use your radio?

Mumma:

Oh yes. This retirement village is one place, down here, where we can have a radio and put up an antenna--

Nebeker:

Yes, I certainly saw it! (laughter)

Mumma:

Well, most of these people don’t object to radios. Many other retirement homes don’t allow them at all. There are three amateurs down here.

Nebeker:

Thank you very much.