I recently completed a repair on a Pecos River Brass 4-6-4 Santa Fe that I thought was worth sharing. Symptoms first appeared as uneven running and lurching a bit a low speed. A quick look on the track revealed way too much free play in the connecting rods and axles in relation to each other. A closer look on the bench revealed the extent of the problem. Sadly, I forgot to photo-document the 'before' condition, but suffice to say that no amount of lube can save soft brass in a sliding, load bearing application. 
The four crankpins on the front and rear axles were completely shot, worn down almost to the screw going through them. The washers holding the rods on were also worn and oblong. Fortunately, the rods themselves were fine, which was a relief. The crankpins on the center axles also appeared ok. Nevertheless, it was clear a major rework was in order.
Here is the problem child. She's a beauty, but high maintenance, lol.
Disassembly of the engine is pretty straightforward, and the axles can all be removed by removing a plate on the bottom of the chassis. I was able to remove the front and rear axles without removing the wheels, disturbing the quartering, etc. I had to make some fixturing, but I was also able to do the removal and replacement of the crankpins with the axles assembled.
The wheels are brass of course, so I proceeded very carefully with removing the crankpins. As far as I could tell, the wheels themselves, and the crankpin fit in them, were fine, so I really didn't want to add any additional work by damaging the hole in the wheel. I started by grinding the worn ends flat so that I had a reasonably true surface to press on, then pressed the pins out, being careful to back up the inner surface of the wheel so that I didn't put any load into the face of the wheel. Definitely didn't want to crack or deform any spokes!
After removing all of the crankpins and taking detailed measurements, I was able to determine that they were not cylindrical, but actually slightly tapered. At this point, I decided it was worth modeling things up in Creo, so that I could be sure of getting an accurate part with a proper fit. Here's a view of the wheel assembly model:
What I really wanted to see in the model was the section view, showing the relation of the crankpin, washer, screw, and rod. The taper is so slight it is not really visible in this pic.
Here is the drawing showing the dimensions and tolerances I was after. 12L14 steel is commonly used for pins and shafts, and is easy to machine.
Determining the exact taper of something like this is pretty sensitive to the accuracy of your measurements, and depends on the original parts being true and uniformly manufactured. In this case, I had measurements of the four pins suggesting the dimensions shown above, which correspond to a 0.5 degree taper. But I didn't know if that was really the original design intent, or just the result of manufacturing variation, measurement error, and so on. As the taper was critical to getting a good press fit, I made up a test part to evaluate.
After test fitting the part in all of the holes on the wheels, I found the required taper was actually less than what I had determined based on my measurements. I made another test part with a 0.25 degree taper, and checked again. This fit perfectly in all the holes on the wheels, so I updated the dimensions accordingly. However, I noticed while doing this second fit check that the taper was seating in the holes at different heights, meaning that while the taper was correct and fit well in each hole, the diameters of the holes were slightly different. This meant each of the four pins would have to be custom fit to each hole.
At this point, I was ready to begin 'production' manufacturing the four crankpins. Here's a pic of the parts in progress with the taper machined on the blanks (the original beat up pins are in the baggies as well). I made the taper about twice the length I would need, so that I could accommodate the different hole diameters by cutting the pins off the blanks at varying locations along their length.
Here's one of the parts in the lathe, getting center drilled for the 1.6mm hole that becomes the M2x0.4 thread:
After drilling, I placed each part in turn in the chuck and test fit the wheel that it would be installed in. With the wheel pressed on snugly by hand, I measured the first end position (the small end of the taper that would be on the inside of the wheel). I cut the part at that point and tapped the M2x0.4 thread. After tapping, I advanced the cutoff tool to the other end of the pin and cut it from the blank. After a little deburring the parts were complete.
On to the installation... Here is one of the crankpins in the installation tool, along with the tool I made to back up the inside of the wheel while pressing the parts in. The installation tool has a 1mm counterbore in it for the crankpin, so that when pressing them in, the tool gets pressed flush to the face of the wheel. This sets the installed height of the crankpin.
Here are a few pics of the installation process. With the installation tools on the pin and backing up the inside of the wheel, it was just a matter of pressing it in until the clamp was tight.
Here is the finished installation on one of the wheels. The installed height was nicely repeatable thanks to the installation tool.
As the washers were also shot, I made a set of replacements for those as well. Based on the model, I increased the outer diameter of the washers vs the original, for a closer fit in the counterbore on the connecting rod:
And finally, the moment of truth! Here is the axle installed with the rod in place on the crankpin:
With the washer in place:
And fastened:
Several hours of running later, everything looks good!
Next up, I'll do a post about repairing a cracked brush in the Pittman motor. Did I mention she was high maintenance? 
