Roundhouse (MDC) Harriman Passenger Cars derail

Several ideas:

1) Check out - The MDC Harriman cars are about 60' long, putting them into NMRA's category "L", for which 20" is the recommended minimum radius. You can run them on 18", but the trackwork has to be very smooth and kink-free, and ought to have transition curves (even a half section of 22" radius at each end of the 18" curve would help). 18" curves may be too sharp to successfully use body mounted coupler pockets on 60' cars.

2) Use an NMRA or other good gauge to check that all the wheelsets are in correct gauge. The MDC wheels are not too tightly pressed on the axles, so that the wheels can be moved or twisted out of gauge. Make sure the wheels are centered on their axles, and that they are not off-center so one wheel rubs on the back of the truck sideframe. Make sure the wheelsets spin freely, and that there is no casting flash on the sideframes that might be rubbing on the wheels. While checking the wheels, make sure none of the wheels are cracked, and that none of the flanges are broken.

3) Make sure the trucks pivot freely under the cars and do not get caught on the stirrup or end steps. Make sure the truck mounting screws are not too tight or too loose, so the trucks pivot freely. It is best to adjust these screws so one just allows the truck to rotate, but not to rock from side-to-side, and so that the other screw is a little looser so the second truck can rock slightly. (This allows the car to follow slight twists in the track without causing wheels to lift off the rail and derail, but also keeps the car from rocking from side-to-side when pulled in a train.)

4) MDC changed the design of the coupler mounting over the years. The trucks with integral arms supporting the coupler pockets are not too bad. But for a time, MDC offered separate coupler pocket/support arm pieces that "floated" around a hub on the top of the trucks. I have found that these do not work very well (especially when backing as the arms tend to buckle and give a sideways thrust causing derailments) and should be rigidly attached to the trucks or better, be replaced with better trucks/couplers (or body mounted couplers on gentler curves.)

Hope some of the above help. Gary Q

Thank You I will check all this

Greetings from Germany Juergen

18" (45cm) radius is a sharp curve. You should have 24" (60cm) or better if at all possible. If you must use such sharp curves, makes sure you have easement curevs (spirals) at each end of the curve.

Geezer has very good suggestions. I would add that you should also note where on your layout the derailments occur: if they happen repeatedly in the same places, then your track work is substandard there. Common problems areas are turnouts (Weichen), and rail joints on curves.

If you are using sectional track, make certain that all joints are tight and smooth. Run your finger over them (carefully - you don't want to cut yourself on a bad railjoint.) If railjoints don't line up properly, do _not_ force them into line - that is sure to cause kinks. Also, use larger radius sections at the beginning and end of each curve -- these will form a approximate spiral.

Good luck!

OK I use NMRA and correct the wheelspace I built on 2 cars body mount kadee #5 I spend a little more weight on the cars all better, only derails on one point of the layout (a s-curve) i can not correct the rail. all locos and cars no problem I trade the MDC-wheels to metal-wheels from my scrap-box i have only wheels for 2 cars no derail more which manufacturer have these metal-wheels?

juergen

Kadee. 36" (scale) diameter. Roco - 10.5mm diameter RP25 profile.

You might check the weight of the cars. NMRA recommended weight is

1 oz + 0.5 oz per inch of car length. A 60 foot prototype car will be about 8 inches long, so the cars ought to weight about 5 ounces. I weigh my cars with an el-cheapo spring balance which I liberated from the wife's kitchen. Office supply stores sell postal scales for weighing letters which work well for weighing rolling stock. I have some Roundhouse 48 foot open vestibule wood coaches. After weighting them to NMRA recommendations derailments went away.

David Starr

Ouch! S-curves are _very_ bad news - so bad that I will do just about anything to avoid them when designing layouts. I will even go to slightly under minimum design radius to get rid of an S-curve.

If you must have an S-curve, you need a straight section between the curves. This straight section should be at least as a long as the longest piece of rolling stock. You may be able to get a section of straight in there by shifting both curves by a centimetre or two.

If you can't get rid of the S curve, you may have to restrict the Harriman cars to some other route. Such restrictions were and are made by railroads, so it would be correct to make such a restriction.

all locos and cars no problem

Atlas, Kadee, Lifelike/Proto 2000, Jaybee, Northwest Shortlines, Walthers, etc. Choose 36" wheels. (In N. America, we specify wheels by their prototype diameters in inches, not by their actual diameters in mm.)

Good luck!

And two #4 turnouts too close together can make a nasty S curve too. I've had practical experience with this situation. Bruce

Thank you all for your help i will show at the weekend i think i can the track a little re-organize the s-curve built with grades in the middle juergen