Atlas GP9 - 2006 vs. 2016 Universal Worm Shaft Compatibility

[C855B]

I have a trio of Atlas GP9s, one from the 2006 run and two from 2016. The earlier version is very noisy, and it wasn't a leap (removed the trucks) to trace the problem to the worm shafts. I "Beardenized" it with some reduction, but it's still a coffee grinder. The new ones are blissfully quiet. So...

Does anybody know off the top of their head if the 2006 version will take the 2016 unversals and worm shafts as drop-in replacements? I guess I could find this out for myself by disassembling one of each version and trying the swap, but they're not the easiest to get back together (motor saddle on the 2016s is a tight fit). I'd like to avoid that.

[Steveruger45]

Hi Mike not sure I can tell from the parts diagrams here. https://shop.atlasrr.com/t-partsdiagrams.aspx
Anyhow when I was de-coffee-grindering  a couple of B40-8 locos, after much investigating I found the main cause was motor alignment.  A new motor saddle fixed them both yet the old ones looked perfectly ok. 

[C855B]

Good thought. I will check the saddle for asymmetry or just generally try it in different orientations and see if that helps. The old saddle isn't quite as clear about proper orientation as the new one.

[tehachapifan]

Something else that I think is a possibility with at least some Atlas locos is the orientation of the bearing blocks. While they appear perfectly square, I believe the inner dimensions...the part that comingles with the frame fingers...are perhaps a tad different and can possibly change the alignment of the worms ever so slightly. This is just a theory, however....I have not actually measured the inner dimensions.

[Steveruger45]

Russ, those bearing blocks also often have molding sprue on the flat faces that bear up against the side frames.  That can screw up alignment.   The serrated faces bear up against the block support fingers.  Atlas make a revised bearing block, its part number is the same as the original but with the letter A at the end.  These newer design bearing blocks don’t have the lips on the frame sides and only have the lips  on the finger sides of the blocks.   I’ve found them to be somewhat better to get properly in place and seated.  The thrust faces on the newer design are also different in that one side of the block is more raised in way of the shaft hole.  I don’t know the correct orientation but I use them with the more raised side facing the worm.  I’ve also bent up the fingers in the frame that support the bearing blocks a tad to raise the worm away from the truck tower pinion gear a tad, as a last resort as this could cause a lot of alignment issues in itself.  For the motor alignment I check the clearance at each flywheel to the frame on top and bottom and on both frame sides.  There’s a lot more but the above has helped me get a good few growler locos to run quietly.

[C855B]

... after much investigating I found the main cause was motor alignment. ...

Bingo. Not completely silenced, but much reduced. Solution was a 0.015" shim underneath the motor saddle, in addition to the Beardenizing from previous work. I have another 2006 GP9 on the bench in the process of a GP9B conversion, so we'll divert that to the motor shop    and see what can be done with it.

[Steveruger45]

Hi Mike,
Yep shimming often works, especially on the locos that don’t use the one piece saddle but have the plastic rings at each end of the motor.
I use various thicknesses of card stock for shims, such as old business cards and old Christmas cards and such to find the sweet spot and if needed a thinner card with masking tape applied in case the sweet spot is between the various thicknesses I have on hand.  Sometimes shimming or I should say spacing between the two halves of the frame at the fuel tank area helps too in case it gets noisy again after fitting the tank back on.