truck tuning: consistent delrailment on a tight radius

[Point353]

If the derailment is due to the trucks being rotated to their maximum rotation angle, then the middle axle is not the cause of those derailments.  The loco's wheelbase is just too long for that curve. The truck's wheelbase is ok.

I wasn't referring to the wheelbase of either the loco or the trucks.

[EmdFan]

I have not read the whole thread, but I do know that there have been some issues with the SD35 trucks not having enough “rocking” motion. I work for Atlas but do not have anything to do with repairs. I would suggest considering sending it in for our techs to take a look at. I do not believe it’s a widespread issue but I have heard about this on a number of occasions.

[peteski]

I wasn't referring to the wheelbase of either the loco or the trucks.

Ok, then I still do not understand the reason for tightening the gauge of the middle wheelset of each truck.

[Point353]

Ok, then I still do not understand the reason for tightening the gauge of the middle wheelset of each truck.

Not tightening the gauge of middle wheelset in both trucks, but tightening the gauge of middle wheelset in the (long hood) truck that derails to match the gauge of middle wheelset in the (short hood) truck that does not derail.

[peteski]

Not tightening the gauge of middle wheelset in both trucks, but tightening the gauge of middle wheelset in the (long hood) truck that derails to match the gauge of middle wheelset in the (short hood) truck that does not derail.

Ah, ok.

[NtheBasement]

Maybe I missed it but did the OP re-measure the track gauge after installing the new flex?  No sense munging a loco to deal with bad gauge.

Looking at the photo way back, I don't see any solder at the rail joints.  Maybe the joints are just a lot neater than what I've been able to do.  Also nice job offsetting the joints.

I had a similar conundrum on my helix because you can't solder-then-bend and you can't arrange for the joints to be on any tangents.  I wedged the outer edges of the ties inward a bit to create a bit of a straight before soldering to get straight joint, then released the wedge and let the cement dry.  Since the helix track isn't visible I didn't bother sliding ties under the joints.

[DKS]

I had a similar conundrum on my helix because you can't solder-then-bend...

Why not? I've done it, and I'm sure plenty of others have.

[Boilerman]

I sould check the rail spacing with an NMRA gage to verify that the track is in gage and if it is then check the travel of the trucks on the unit that derails.
Also I don't have ant gaps on the curve sections on my layout as the wheels will pick those gaps and can derail.
I used Atlas Cod 55 flex and I solder the joints prior to forming the radius and I remove the spikes as needed when forming the radius and this has worked for me,
That has eliminated derailments on curves on my layout.

[freedj]

New Conclusion:  The locomotive derails when going counter clockwise with the long hood forward.

I have picked up some 10R" Code 55 sectional track and run the loco in each permutation:
* Long hood forward - Clockwise - OK
* Long hood forward - Counter Clockwise - Derails
* Short hood forward - Clockwise - OK
* Short hood forward - Counter Clockwise - OK

This is consistent with what I see on the layout.

This was on a level test bed. 

The derailment appears to occur when the trailing truck hits full rotation.  While there is only one truck turning the problem truck is fine.  Once the 'OK' truck fully enters the curve the problem truck derails.  I don't know how to measure what full lock is on the trucks.  Maybe i can come up with something with 123 blocks and a piece of straight track.

I have also held up my 1/3 of a circle of 10"r to what is on the layout and it is fairly close.  The layout is mostly a bit more than 10", but there are one or two areas that seem like they might be less.  It's really hard to gauge.

[wm3798]

There may be one more adjustment you have to make.  First, carefully remove the shell from the drive, and put it aside.  Then, using a small jeweler's screwdriver, separate the side frame from the truck.
Now place the truck, the sideframe, and the locomotive drive on a paper towel.

Now take that paper towel out to the driveway, and put it under the tire of your car.  Drive over the locomotive, backing up and repeating the maneuver as often as is necessary, then go to the hobby store and buy another engine.

Place the shell on the new drive, and see if that doesn't fix the problem.

Lee

[freedj]

I've rigged up a test fixture to check truck articulation.  These numbers are only useful in relation to each other.

* Long hood forward - Counter Clockwise - 65mm
* Long hood forward - Clockwise - 70mm

These measurements were taken by taping a piece of straight to the straight edge and clamping that down with the 123 block.
The loco is railed on the clamped piece of track, and a floating piece of sectional straight track.
The loco is pushed up against the 123 block.  The truck makes contact first.
The floating piece of track is pulled away from the straight edge until it wants to spring back.
The small square is added to the floating track while holding the floating track still, and butted against the truck.
At this point the loco is pinched between the square and 123 block by the trucks.  The frame does not contact anything. 
The large square is brought in square to the straight edge and the scale is read.

These measurements were repeatable +/- .5mm.

I am not exactly sure how these translate to degrees, but there is a 5mm difference between directions. 

Questions: 
Should an atlas sd-35 be able to negotiate a 10r" curve (i probably should have asked this a long time ago)?

What does it look like to swap the trucks on this loco?  Those little wires that connect to the frame kind of scare me.  Do they just pull out?  It looks like there are three clips that hold the frame halves together.  Can i pop the lower two and get enough slack to drop the trucks?

Since this measurement is mirrored for the short hood forward, why does it not derail when running short hood forward in either direction?

The only thing i can think of is the gauging of the leading truck.  I know the long hood truck is slightly wider than the short hood truck, but still in spec.
How do you actually adjust the gauge on the wheels?  I've read about screwdrivers and prying, and also truck disassembly, but I havn't been able to find a good description or a video.  Only a general outline with the caveat of "gently so you don't bend an axle".

[ Guests cannot view attachments ]

[freedj]

Now take that paper towel out to the driveway, and put it under the tire of your car.  Drive over the locomotive, backing up and repeating the maneuver as often as is necessary, then go to the hobby store and buy another engine.

Believe me, i've been considering it! 

[freedj]

I have not read the whole thread, but I do know that there have been some issues with the SD35 trucks not having enough “rocking” motion. I work for Atlas but do not have anything to do with repairs. I would suggest considering sending it in for our techs to take a look at. I do not believe it’s a widespread issue but I have heard about this on a number of occasions.

Is the above consistent with what your describing as "rocking"?  It is still under warranty so sending it in may be my best option.  I feel like i have eliminated my trackwork as the root cause.

[strummer]

There may be one more adjustment you have to make.  First, carefully remove the shell from the drive, and put it aside.  Then, using a small jeweler's screwdriver, separate the side frame from the truck.
Now place the truck, the sideframe, and the locomotive drive on a paper towel.

Now take that paper towel out to the driveway, and put it under the tire of your car.  Drive over the locomotive, backing up and repeating the maneuver as often as is necessary, then go to the hobby store and buy another engine.

Place the shell on the new drive, and see if that doesn't fix the problem.

Lee

  ...another reason why I love this site!

Mark in Oregon

[wm3798]

I confess I've used that before, but usually as a maintenance tip for Bachmann engines...  Back over it in the driveway on your way to the store to buy an Atlas or Kato engine...

Lee