Milwaukee Road Passenger Car Project

[u18b]

Shifting back in time, coach 4400 was the very first lightweight coach Milwaukee built back in 1934.  I finished the body for it today.  Milwaukee used this car as a test car and later showcased it at the World's Fair along with an EP-2 electric.  It lasted in revenue service until 1961 and did not receive any post-war modifications. 



It is very close to a Fox Velley car, except the windows are larger.  On the prototype cars that Fox Valley modeled, Milwaukee narrowed the windows so that they could add more seats to the cars, at the expense of less leg room.

I created a new core template for this and future 1934-era cars.  The roof has a slightly different logarithmic curve to it that ends in a slight point at its peak.

The hardest part was breaking the seal on what was probably the last brand new, shrink-wrapped copy of Hiawatha: First of the Speedliners in existence.     That was the most expensive train book I've ever bought.

Wow Aaron!  I want one!  I love it.
This was a one of a kind car.   And the 1933 World's Fair EP-2 may be next on my list.
Milwaukee Road made a big deal about this car.  I believe it was their very first transition away from heavyweights.
As Aaron said, they were "unveiling it" at the World's Fair.

Here is a brochure of the exhibit.

[Sokramiketes]

Ah yes, the first of the new construction!  It was an iteration just after they converted a heavyweight to new car sides.  That one is interesting too with six wheel trucks and a clerestory roof if I recall correctly, but with the trademark sides and arched windows of the future speedliner cars. 

[chessie system fan]

Interestingly, the 4400 coach is exactly the same as that converted heavyweight, minus the roof.  That car is also on my to-do list.

Also, I recently acquired a Fox Valley car, and I was surprised to discover that Matt cheated!  The car sides should be tapered near the bottom, but instead Fox Valley did a slab, straight side.  The curve at the edge hides it.  The curve is too pronounced on the 1936+ cars, but I might explore doing the same thing on the 1934 cars, depending on how well the taper prints.

[peteski]

Also, I recently acquired a Fox Valley car, and I was surprised to discover that Matt cheated!  The car sides should be tapered near the bottom, but instead Fox Valley did a slab, straight side.  The curve at the edge hides it. 

Is that really cheating, or only a compromise made to work around the limitations of injection molding process? Since the model's walls are magnitude thicker than the 1:1 sheet metal, if the taper of a thick wall can represent the 1:1 taper, then is that wrong?

[chessie system fan]

I can't speak to the difficulty of tooling injection molding, so I don't know if tooling a car side of this profile is too difficult.



(The perspective is end-on.)

[peteski]

You're right Aaron, I see what you mean.  I was going from memory (it has been a while since I looked at my set).


As you mentioned, the sides are flat all the way to the curved bottom edge.  No taper.  I guess we really don't know it this is due to an oversight or mold making limitations.

[chessie system fan]

Considering I have the same source material Fox Valley used, I suspect purposeful. Ease of pad printing, perhaps? And considering I might be the first person to ever notice it--and only because I know what to look for--he might not have made the wrong choice. We're talking tenths of millimeters here.

[NSEGeorge]

If I understand the car, it tapers in at the bottom from the roof line to the curved section. Even on a four slide mold, the core where the part sits has to be ejected from the press by pins within the mold for the part. You can push the item out with pins, but if you inject plastic into a reverse angled core, the plastic will stick in the mold. To inject this in plastic the 'core' needs to be able to slide out of the body so the inside would have to be slight less than straight (there is a .5% draft angle) so the part would be thicker at the top than it would be at the bottom. While this would work, it would make the windows thicker at the top than the bottom and over all decidedly thicker than on a straight side car. My guess is that it was decided that was a no go.

The only other way to produce that taper in injection molding would be to have what is called a 'collapsable core', which is just what you would think. Its a complicated kind of mechanics that has to move out of the way to eject the part from the press. Expensive as heck, and a very hard thing to justify for a single road, limited market car.

IF one were to design the car with the floor integrated with the sides maintaining the taper it would be possible to eject, but then you would have the problem of the under floor curved section having to be flat on the back. Compromises are a reality of injection engineering.

There is no issue with pad printing on a tapered shape of this dimension, so I doubt very much if that was a consideration in the design for Matt.

Overall, this car, and others like it, prove there is a need for the next level in additive manufacturing. We've been using Mimaki full color 3D printers for props for a several years, but they are so expensive as to be not practical for the mass market yet. I believe these printers will come down in price over the next few years, and then we will have something for modelers that doesn't require painting, and something for those who paint with extreme detail fidelity with no supports to cut off.

Times they are a-changin'!

NSE George

[chessie system fan]

Thanks for the technical information, George!  It's appreciated.  That explanation makes sense.

Meanwhile, this observation end turned out to be more challenging than I thought.   



Since it's easier to see, this is a good spot to show what the inside of my cars look like. The bodies are 1.5mm thick, but the window areas are 0.5mm thick and have 0.5mm wide interior boundaries around each window.  There is also a ledge along each side to support the floor.  The floor is 2mm thick.



After the body is completed, I'll add a breakaway X pattern that will both help with printing, keep things square, and support the floor.



The solid area in the middle doesn't extend the whole length.  I'll add car road number or class type info there so that I will be able to easily keep everything organized.

[Jim Starbuck]

Thanks for the technical information, George!  It's appreciated.  That explanation makes sense.

Meanwhile, this observation end turned out to be more challenging than I thought.   



Since it's easier to see, this is a good spot to show what the inside of my cars look like. The bodies are 1.5mm thick, but the window areas are 0.5mm thick and have 0.5mm wide interior boundaries around each window.  There is also a ledge along each side to support the floor.  The floor is 2mm thick.



After the body is completed, I'll add a breakaway X pattern that will both help with printing, keep things square, and support the floor.



The solid area in the middle doesn't extend the whole length.  I'll add car road number or class type info there so that I will be able to easily keep everything organized.

This is just amazing work Aaron!

[u18b]

Aaron,
have you printed anything with that X pattern yet?

I assume that is to prevent warpage during printing and curing.

I'm interested to see how this works for you.

[chessie system fan]

I've printed X patterns of varying designs on every model the past few years.  This current iteration has been more successful than the others. The 45 degree lattice structure does a good job of keeping everything firm during the print, yet doesn't sag like straight 90 degree supports do.

Obviously, this doesn't work if I'm printing the car flat because the million supports needed would defeat the purpose.

[chessie system fan]

I completed the bodyshell for the observations Omeme and Opechee.



This leaves one car left to do for the 1937 train.

[u18b]

Here you go  Aaron.







The screw head diameter: 0.172" (4.38mm)
Kingpin diameter: 0.120" (3.06mm)
Kingpin height (just the part that sits inside the truck bolster): 0.072" (1.83mm)
Top of the truck's bolster over railhead (not including the 2 bumps): 0.195" (4.95mm)
Height of those bumps: 0.020" (0.52mm)
The last 2 dimensions might not be exact.

Aaron.  You asked about that screw a while back.   Is it possible that an Atlas frame screw would work.  I have some you can try when you come over.

[u18b]

I've printed X patterns of varying designs on every model the past few years.  This current iteration has been more successful than the others. The 45 degree lattice structure does a good job of keeping everything firm during the print, yet doesn't sag like straight 90 degree supports do.

Obviously, this doesn't work if I'm printing the car flat because the million supports needed would defeat the purpose.

This one seems beefier than the other ones you did.  Like on that generator car.