Design changes oftentimes need to be made on an assembly that is close to achieving production status. Changes such as these need to be very carefully thought out.
This statement is both a call for help as well as stating an impossible condition. By its nature, fixing a design is going to require changing something. When I hear this, I immediately think” I am going to need to change something, it just depends on who I am going to upset at how much I am going to change. Another way to look at this is to make subtle changes that will remain confined to a small area of the assembly and not force other changes to be made throughout the assembly-keeping a tight lid on Pandora’s Box.
With the advent of microprocessors, oftentimes making a change is as simple as a minor programming change. Recently, I provided development assistance for a robotic vehicle development effort. When I made note that the structural loads were too great for the condition presented for analysis, the solution turned to be a relatively minor programming change. In truth, we did make a change, but it was one that involved no mechanical hardware.
In another life, I was faced with the same dictate, but I knew going in to the task that things would have to be changed. The task boiled down to determining at what level the changes would be acceptable. Since this was a military program, and specifically in the area of large caliber artillery projectiles, fundamental items like weight, overall length, adding mechanisms, electronics, etc. were out of the question. The problem can be described as follows:
The projectile, indeed a shell, contained submunitions, which at the proper time were ejected from inside the shell out the back end. Such devices have to be constructed in a very robust fashion, as they must tolerate anything from 10,000 to 30,000 times the acceleration of gravity during their trip down the canon barrel. During the ejection event, a pyrotechnic charge detonates within the shell, releasing what is known as a base plug. The base plug takes all the pressure loads during firing and is typically made from solid steel. The submunitions within the projectile are pushed backwards out into the atmosphere, traveling at speeds that could exceed 2000 ft./s and spinning at rates of 400 RPS or higher. Once the submunitions are exposed to the slipstream, they undergo a deceleration and despinning event. This will cause them to hit the base plug, likely damaging them to the point where they will not function. The problem was how to move the base plug out of the way without making any significant changes to the entire design. The submissions cannot be redesigned in any manner, so that leaves the base plug. Loads on the base plug were mostly compression loads. What needed to be done to correct the base plug-submunition collision problem is to kick the base plug in a radial direction away from the centerline of the projectile. The trick is where to find the energy to do this. After digging around in an engineering dynamics text, I came across something that looked useful – spin axis transfer. If a spinning body suddenly comes apart, the center of gravity of the once-solid piece remains the same, but the two parts fly off in separate directions. This actually had been tried in an earlier attempt to solve the problem by cutting a half disk -shaped section out of the base plug portion internal to the projectile. Once the base plug was blown off the back of the projectile, the spinning motion would kick this half-disk to one side followed by the base plug being pushed in the opposite direction. This did work to a certain extent, but was inadequate. So now the problem became how to amplify this effect? Running some manual calculations combined with some dynamics computer programs I put together demonstrated a fairly simple method to do this: force the half disk to pivot by pinning it to a place close to the base plug outer radius, and removing material to allow this piece to swing outwards relative to the base plug centerline once released from the confines of the projectile. The half plate will now absorb some of the base plug spin energy causing it to swing rapidly outwards. Once it reaches approximately 180° relative to the base plug, other design features will kick the plate off the pivot pin, releasing the half plate. This effectively increases the effectiveness several times. A test was put together that demonstrated the effect of this design change.
Yes, changes were made, but none that would affect any other part of the design.
Norman T. Neher, P.E.
Analytical Engineering Services, Inc.
Elko New Market, MN
www.aesmn.org