Rather than hijack the other 4-8-4 thread I thought it best to start a new one.
From the other thread:
richhotrain This thread brings back bad memories from years ago when I bought the BLI Paragon 2-10-2 and 2-10-4. Lots of starts and stops, front and rear trucks derailing, and the driver wheels derailing on curves. It took a lot of running to overcome the starts and stops, but eventually those problems went away. I had to send both locos back to BLI where they installed stronger springs, and the front and rear truck derailments went away. As far as the driver wheel derailments, the blame was on my 24" radius curves, so I wound up selling both locomotives. Rich
This thread brings back bad memories from years ago when I bought the BLI Paragon 2-10-2 and 2-10-4. Lots of starts and stops, front and rear trucks derailing, and the driver wheels derailing on curves.
It took a lot of running to overcome the starts and stops, but eventually those problems went away. I had to send both locos back to BLI where they installed stronger springs, and the front and rear truck derailments went away. As far as the driver wheel derailments, the blame was on my 24" radius curves, so I wound up selling both locomotives.
Rich
I have one BLI locomotive, a 4-8-4 Northern. It also has driver wheel derailments at two spots on the layout. These are 24" radius curves and the derailment happens exactly in the middle of the half circle loop. Interestingly I have one covered hopper that derails in exactly those same spots. No other cars or locomotives do.
How much work did you do to troubleshoot this? I am thinking I may have a kink in the track or out of gauge issue from laying out the curve but I haven't had time to look beyond a quick glance. Before giving up I was going to relay the track on one of them to see if it solved the problem since I have other 24" curves that don't cause derailments.
Rick
If an engine derails at only one spot, it's fairly likely it's the track, not the engine. If an engine and a car both derail at the same spot, it's almost certainly the track.
Some large engines, like the BLI 2-10-4, require a 24" minimum radius, so by running it on a 24"R curve you're right on the edge of it's tolerance. If there's even a slight kink in the curve, so it's a little sharper, the engine won't work on it.
I'm guessing you're using flextrack, perhaps try substituting a few pieces of 24"R sectional track at that spot and see what happens?
Thanks, that is what I was thinking also. Because only one engine and one car have the problem it hasn't been a high priority to fix.
I didn't know Atlas had 24" sectional track. That is an easy solution.
The true test of good track work is a heavy steamer. Check for horizontal flex by pushing down on the track in the spot where the derailment is occurring, it needs to be solid as the weight of the engine will make it flex down as it goes over it creating a pothole that will make a steamer come off.
Horizontal flex may not cause a problem on a straight section but sure will on a curve.
I made these radius gauge measurers out of appliance box cardboard. They work really well and the price was right. I used a metre stick to draw an arc and carefully cut them out.
Also, I generally let my turnouts float, however, I have several curved turnouts that I needed to stick down as the heavy steamers would torque them at speed and out they would pop. Once they were caulked down, problem solved.
Use your phone to video the problem area up close as the engine goes over and look at it in slow-mo on the big screen TV.
This hobby can be frustrating at times, however, every problem has a solution and it is generally an easier fix than we thought.
Brent
"All of the world's problems are the result of the difference between how we think and how the world works."
hbgatsf Rather than hijack the other 4-8-4 thread I thought it best to start a new one. From the other thread: richhotrain This thread brings back bad memories from years ago when I bought the BLI Paragon 2-10-2 and 2-10-4. Lots of starts and stops, front and rear trucks derailing, and the driver wheels derailing on curves. It took a lot of running to overcome the starts and stops, but eventually those problems went away. I had to send both locos back to BLI where they installed stronger springs, and the front and rear truck derailments went away. As far as the driver wheel derailments, the blame was on my 24" radius curves, so I wound up selling both locomotives. Rich I have one BLI locomotive, a 4-8-4 Northern. It also has driver wheel derailments at two spots on the layout. These are 24" radius curves and the derailment happens exactly in the middle of the half circle loop. Interestingly I have one covered hopper that derails in exactly those same spots. No other cars or locomotives do. How much work did you do to troubleshoot this? I am thinking I may have a kink in the track or out of gauge issue from laying out the curve but I haven't had time to look beyond a quick glance. Before giving up I was going to relay the track on one of them to see if it solved the problem since I have other 24" curves that don't cause derailments.
As for the 4-8-4, that loco should be a lot more tolerant of 24" radius curves than my 2-10-2 or 2-10-4 were. The more drivers on a steam engine, the more prone the loco is to tight curves.
At one time, my rule of thumb was 3 inches of radius for every driver set on a steam engine. Using that rule, a 4-8-4 should navigate 24" radius curves, but a 2-10-2 or a 2-10-4 would need a minimum of 30" radius curves. That said, I had problems with the 2-10-2 and 2-10-4 on 32" radius curves.
It may be that what you think are 24" radius curves or what I think are 32" radius curves are actually tighter toward the middle of the curve. Easements can be formed on either end of the curve, but the middle of the curve will exhibit no tolerance for error.
You mention the possibilty of kinks or out of gauge trackwork on the curves. The trackwork could possibly be out of gauge somewhere along the curve, but a kink is more likely. Do you solder the rail joiners in your curves? That should eliminate kinks if done properly. Another possibility is uneven track where one rail is higher or lower than the other rail. Check out that possibility. Superelevation of the curve can also help to eliminate derailments.
Alton Junction
hbgatsf I didn't know Atlas had 24" sectional track.
I didn't know Atlas had 24" sectional track.
Rick, what does this derailment look like immediately as hit happens?
Does the offending driver flange move outboard or drift inboard on the curve?
If the flange moves laterally out of the gauge toward the outside of the curve, you have one of two* possibilities, and the radius is not one of them for the curve you are using:
The outside rail is low or the gauge is pinched.
If the flange moves over the rail inward, toward the center of the circle whose arc the curve represents, it is also a low rail or a pinched gauge.
The only other possibility in my experience is a change of radius that is not obvious to you, but it impacts the locomotive. Naturally, we're not talking about a widening, but a reduction in radius.
*I'm assuming the locomotive's range of motion, as intended, is full for the drivers in the way of lateral motion, and that the forward trip pin isn't snagging on something. It won't be a trailing car forcing the locomotive to derail, so you can rule out that; the loco is so much heavier that it should not be horsed out of the gauge by a trailing component.
There's really not much you can do with the engine to make it not derail consistently. Best idea is to replace the track.
I'll get hate for this, but my BLI ATSF 2-10-2 runs fine on my 22" curves. It does derail on my single 19" curve track. (I use sectional track over flex track on curves for the very reason that its almost a guaranteed smooth curve.) I have the MRR 2012 Virginian track plan.
It looks awful running on my curves so that's why its unfortunantly more of a shelf queen...reserved for a future, larger layout.
---------------------------------------------------------------------------------------------------------------
Modeling the PRR & NYC in HO
Youtube Channel: www.youtube.com/@trainman440
Instagram (where I share projects!): https://www.instagram.com/trainman440
Especially on the second run of ATSF 2-10-2's, the "modernized" version with drifting valves, BLI designed plenty of lateral motion into the valve gear. They also have a much better mechanism in them than the first run 2-10-2's, and perform much better. (First run 2-10-2's have a plastic gear that will fail. It's a question of when it fails and not if. They all need replaced. Second run has metal gears--much better. I've had both versions apart).
The 4-8-4's, including brass hybrids, are actually more "rigid" and with their large drivers are more likely to give me problems on my layout than the Santa Fe 2-10-2 ever did. The second run ATSF 2-10-2 has some drivers actually sprung (I had to replace a missing spring and fortunately had the right item on hand or in the parts bag). The 4-8-4's, depending upon which one you have, have no drivers sprung.
You have to remove the kinks to run these at the lower limit radius. The 4-8-4's really don't like a vertical kink.
This is part of the reason I used Kato HO Unitrack on my layout--it doesn't allow me to put a kink in. Also, I've improved any radius possible to the largest Kato radii--above 31" radius, and 34" nominal radius if you use the short curved pieces that come out of the #6 turnouts, which you can buy separately--you just need plenty of them (9 for a 90 degree turn).
Good luck with the track work!
If you fix the track work for 4-8-4's it will also make the performance of Tangent 86' hi cube boxcars much better as well.
John Mock
wjstixI'm guessing you're using flextrack, perhaps try substituting a few pieces of 24"R sectional track at that spot and see what happens?
I have areas of my layout where the radius will dip to 22 or 24 inch radius.
Sectional track in these areas is the way to go. There is no tolerance for variation with installation when you are attempting to get right down to the minimum operating radius for equipment.
Sectional track prevents a short section of under-minimum that will only cause problems.
-Kevin
Living the dream.
That's why a lot of layouts have a minimum MAINLINE radius and a minimum radius - There's more breathing room with the larger mainline minimum, and in industrial areas where those big road locomotives do not go, you're still OK.
But with a freely flexible type of flex track liek Atlas or even Peco, the curves form naturally. You're not going to get a pinch point less than minimum in the middle of the curve unless you are very sloppy. Plus there are tools controlling the radius - like the Ribbonrail gauges or Fast Tracks Sweep Sticks. So even if you use the stiffer flex like ME, using those curve templates will ensure the curve is the radius you want without resorting to track joints every couple of inches with sectional track.
And if the linear length of the curve is more than one piece fo flex track - solder two sections together BEFORE you make the curve. Then you won't get a kink in the middle. And if at all possible, even with a soldered joint, don't locate the rail joint right in the middle of the curve.
Simple tips for more reliable trackwork.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
rrinker That's why a lot of layouts have a minimum MAINLINE radius and a minimum radius - There's more breathing room with the larger mainline minimum, and in industrial areas where those big road locomotives do not go, you're still OK.
Which is very prototypical. There are lots of 4 axle only industries out there these days.
An "expensive model collector"
richhotrainAs for the 4-8-4, that loco should be a lot more tolerant of 24" radius curves than my 2-10-2 or 2-10-4 were. The more drivers on a steam engine, the more prone the loco is to tight curves.
There's a lot of factors that can do into that, not just number of drivers. My Bachmann 2-10-0 and 2-10-2 can take sharper curves than my BLI 2-10-4. Depends on the manufacturer, wheelbase, driver size, flexibility of the drivers, etc.
Get a Ribbon Rail track gauge for any minimum-radius curves you put down, to make sure the track really ends up being what you think it is. But be prepared to take up the track and do it again.
It takes an iron man to play with a toy iron horse.
wjstix richhotrain As for the 4-8-4, that loco should be a lot more tolerant of 24" radius curves than my 2-10-2 or 2-10-4 were. The more drivers on a steam engine, the more prone the loco is to tight curves. There's a lot of factors that can do into that, not just number of drivers. My Bachmann 2-10-0 and 2-10-2 can take sharper curves than my BLI 2-10-4. Depends on the manufacturer, wheelbase, driver size, flexibility of the drivers, etc.
richhotrain As for the 4-8-4, that loco should be a lot more tolerant of 24" radius curves than my 2-10-2 or 2-10-4 were. The more drivers on a steam engine, the more prone the loco is to tight curves.
That was my point.
Thanks for the tips. I am doing scenery work and can't run the Northern through the problem areas, nor do I remember which way the wheels derailed.
I did solder the joints before flexing to create the curve. The derailments happen at that joint so my suspicion is that there is a problem there due to the flex. Sounds like the sectional track is the solution.
hbgatsf I didn't know Atlas had 24" sectional track. That is an easy solution.
Id recommend you fix it before its too late. Youre fortunante enough to realize the problem before your layout is more or less "finished", its better to rip it up now than leave it as is just to realize later down the line you shouldve fixed it sooner.
My $0.02
Good luck!
Charles
Trainman440 Id recommend you fix it before its too late. Youre fortunante enough to realize the problem before your layout is more or less "finished", its better to rip it up now than leave it as is just to realize later down the line you shouldve fixed it sooner. My $0.02 Good luck! Charles
Well, the problem has been there for over 20 years. Until I got the Northern it wasn't an issue that I was inclined to fix.