We haven’t been over there (GunLab.net) in a while, and Chuck is always up to something cool. Recently he had something nice to say about us, in a longer post on reverse-engineering; to be explicit, reverse-engineering the MP44 trunnion. But forget what he says about WeaponsMan.com, how cool is it to be making an MP.44 trunnion for (almost) the first time since a T-34 did a pivot turn on the ruins of the factory?
Here at Gun Lab we do a fair amount of reverse engineering, most of what we like to make have no drawings. However when there are drawings or solid models available we will use them. With this said I have found that most of what is available on the internet or in books is just not correct.
A case in point is the MP-44 trunnion. I have all the drawings that I have been able to find on this part, a number of different sets are out there, and when compared with the actual part have found them to be lacking. Some are just wrong and in some cases I don’t think the person has actually looked at a part.
Now, we have a set of MP.44 drawings here. We’ve actually been meaning to show a few of them to illustrate how MP.44 design features migrated into the AR-10 and thence to all its descendants. They’re terribly reproduced, no longer to scale, but they are dimensioned MP.44 drawings.
Say “Thank you,” class:
Now, you might wonder how it can be possible with apparently original (even if lousy), dimensioned drawings, you can’t just poke the numbers in and try to run the part. There are a number of reasons that you could expect drawings to diverge from shop practice. In the real world, in fact, it’s a constant battle to keep the drawings and the processes both aligned properly on the same part. In the 20th Century this got particularly bad because of engineer/draftsman/master machinist/machine operator job specialization and social stratification. Those could be four different guys whose only workshop interactions were with the adjacent guy in the org chart, and whose contacts were all correct.
There’s no way you produce stuff efficiently without the engineers going out on the shop floor, but some are loath to do that, and some shop staff are loath to have an engineer looking over their shoulders. There’s no way you produce stuff efficiently without a steel-cutter being able to walk back into the engineering spaces with a part and a problem, right to the guy who drew the drawings — but that is forbidden more often than it is allowed! So even in the best, cleanest, and least disrupted shops, lines got crossed, things fell apart, the center did not hold… wait, we got carried away there for a bit. But communications were imperfect, even in a perfect factory.
Then, add into the mix, we’re talking about the Third Reich in 1944-45. If the Germans had perfect factories, the Allies bombed them. Meanwhile, the gaping maw of the Eastern Front demanded endless human sacrifices, and in each successive draft call manufacturers could protect fewer and fewer key workers. The “fix” the government proposed for this was that they would provide labor, but that labor was at best displaced refugees from the ill-fated German settlements in the East, but more commonly slave labor from occupied nations.
Something had to go, and one of the things that went was correcting and updating drawings. Seriously, if you compare surviving German drawings to the M1 drawings, your mental picture of “German efficiency” will never recover. (Well, maybe a little when you realize that two large air forces were gamely trying to reduce German industry to the state of the Germans’ forebears in the Neander valley).
Now back to the MP-44 trunnion. We were contracted a while back with making a limited number of new trunnions for the MP-44. He sent us a very good original one and we had a poor copy of one at the shop. Using these two pieces we started the project of reverse engineering it. The easiest thing to do was look for engineer drawings off the web. These are the ones that I found.
His look like they’re from the same set we’ve got here. He has stripped them of dimensions, perhaps because he’s not working with SI (metric) dimensions, but more likely because the dimensions were not “on” compared to the physical parts he had to measure.
The measurements have been removed from these copies, however you can find them on the internet. I did use the basic drawing as a starting point. The sheets were cleaned and measurements were taken using a cmm, micrometers and pin gauges. Tolerances were set using not only the trunnion but also matching parts. When there was a doubt other parts were located to increase the measurement standards. This allowed us to come up with a reasonable solid model that we felt was accurate enough to start programing.
A CMM is a coordinate measuring machine. Think of it as a sort of 3D scanner that touches off against a part and records that position in 3D space. These can be used to gather a cloud of points, or more efficiently, to capture key dimensions.
The problem with using a CMM against a part you are re-engineering is that you’re working off one part, and you don’t know where in the tolerances that part was. (That’s also our beef with David Findlay’s excellent Firearms Anatomy books — for practical reasons, Findlay worked off a single sample of the firearm).
Given enough parts to measure, you can develop a degree of statistical certainty about where the original measurement was supposed to be. Working with most non-US products, you can also cheat a bit by knowing that engineers like to spec things in fairly round millimetric measures — dimensions that end in X.0 or X.5 millimeters, most of the time.
Anyway, here is the first post on re-engineering the MP.44 trunnion, and here is a follow-up post (in which the model turns out to need some improvement). Meanwhile lots of work improving the shop and working on GunLab’s other projects, such as the VG1-5 limited production run.
Note on an Unpleasant Subject
Technical posts like this and GunLab’s would be banned under a gag order slipped into the Federal Register by the State Department — yes, the very people who negotiated the deal to accelerate the nuclear armament of the hostage-taking terror state of Iran this week. The deadline for comments is 3rd August. As we previously wrote (more background there, at the end of a barrel-heating post):
Comments go here at Regulations.gov or by email to: DDTCPublicComments@state.govwith the subject: “ITAR Amendment—Revisions to Definitions; Data Transmission and Storage”. Ceteris paribus, this link should open in your email application with the correct subject header.
Again, there’s more at that previous post on how to comment, but at this time it’s crucial that you comment. A State Department than can censor the Internet is a State Department that has lost touch with America.