i’m not a vehicle dynamics guy so I go with what the experts telll me. One of our best and brightest, Mark Demboski, was stolen by the Brits in the post Hatfield meltdown (yes much to the chagrin of the arrogant track engineers in the UK it took a North American team to figure it out). Mark spent the remainder of his career trying to help Railtrack/Network rail get control over what the TOCs were doing to the less than permanent way with the new modern bogies on their modern generation of rolling stock (one of the strongest arguments for vertical integration I’ve seen).
Euclid
In any case, when I consider this quest to dampen out truck hunting, this unusual truck design popped into my mind. It certainly seems like it would be very stiff to pivot; maybe too stiff. The early tenders were short, so their truck pivot was relatively less than freight cars of that era.
OK, I see what you’re getting at now.
The interesting thing about these trucks, if I am reading the design right in your description, is that they have none of the cross-articulation flexibility of the three-piece (or in a cruder sense, the Fox, and a more refined sense, the Taylor) trucks. The rigid frame may be to keep the bearing brasses strictly aligned and the axles parallel, with any cross-level accommodation sort of approximated by different deflection of the long leaf springs. That picture is so unusual that I have to think I’m missing something. (Note the emphasis on cross-level accommodation by the long-travel CCSBs in the O’Donnell article, and the mention of the issue that was brought up in the earlier “TSB derailment” thread about carbody stiffness influencing derailment propensity on spirals or going into and out of curves…and how to compensate effectively for it…)
Am I correct in presuming that the springs are pinned to the axleboxes and their top ‘saddles’ slide on the tender bolster? And the telescoping center pivot has no ‘play’ in it to let the H-frame do anything but rise and fall and pivot to follow curves?
I do note that the polar moment of inertia of the H-frame in rotation promises to be very small, smaller even than a weak archbar truck, if the mass of the spri
That in itself would provide the equalization, at least the longitudinal equalization, much the same way that the long flexible spring in the 1830s truck did. One major thing this does is put all the weight loading from the tender body nearly directly over the axleboxes, so the ‘connecting’ H-frame can be quite light, and indeed I don’t see any particular reason why there could not be axial pins to permit the two sideframes to pivot after the general manner of a three-piece truck other than it puts the bearings out of alignment whenever the sideframes ‘articulate’ if there is no provision made for the brasses to make this (relatively slight) accommodation (or for the journals to tolerate it with proper tribology).
I don’t think that would be enough to guarantee that the truck would pivot ‘properly’ keeping all four spring ends over th
Wizlish,
Yes, I agree that the center bearing would be essential to mechanically center the truck. My point about being able to run without it was only to illustrate how the load is transferred from the side sills to the tops of the journal boxes.
I have been seeking answers and explanations about this type of truck for a few years. Earlier, I remember having seen it in a lot of photographs, and generally assuming that it was a basic arch bar truck and the big springs were some type of side bearings. But one day, I took a good look at it and pondered what I was seeing. Then I realized that the big leaf springs were primary suspension. If so, I could not see how the load could be carried on the typical center bearing. So I concluded that it must have a telescoping center bearing.
I don’t think there are many of these trucks in existence, but I found an old C&NW 4-4-0 at the National Museum of Transport in St. Louis. I called as asked that someone take a look at the center bearings. They got back to me with a hand sketch of the telescoping center bearing.
Overall, the one question that I have not been able to resolve is the matter that you point out. That is, how does the truck pivot? First of all, there would be a lot of friction to
Here is a good close-up of the tender of 999. It may be that the whole spring turns with the truck pivot, and the top of the arch is actually a sliding saddle that travels in the pivot arc by sliding beneath a smooth bearing plate on the tender side sills. That would accommodate the truck pivot. Then the ends in the tubs (or in this case, the triangular housings) are free to move only forward and backwards within the tubs in order to accommodate the spreading of the spring ends as the spring is compressed.
Some of the answers to this are likely to come via NYC 999, which had perhaps the most famous pair of trucks like this.
Here is the famous ‘builder’s photo’ which shows the elevation view of the original truck structure:
And here is a picture of the engine in Syracuse, showing a 3/4 view at low level of the spring and sideframe structure:
Note how it is impossible for the spring ends to move laterally or twist relative to the position over the axleboxes, and how tall the springs are at their perches (and how narrow the perches are at the point of contact with the tender body)
Interestingly enough, these trucks survive on the tender in the Museum of Science and Industry … with a notable change that can be seen in some of the early clips of 999 running in service: there are additional plates on the outside of the springs, extending up from the H frame, which I think are expressly (no pun intended) intended to restrain the lateral motion of the leaf springs as originally designed…
(I am sorry for the ambient noise, bad lighting, and shaky technique of this clip, but it does show the context…) I think Staufer’s drawing of 999 also shows these plates.
I suspect that someone in the Chicago area could wangle about 5 minutes of time to document the construction details of one of these trucks – or someone familiar with Internet sources could find some better close-up pictures of the relevant structure and provide good-quality links. There is no question that these were considered the ne plus ultra of really high-speed design in the early 1890s; there is also no question that (as with the torque rods on the DR 19 1001 motor locomotive) some pretty
which was associated with this picture, one of the clearest views of the ‘revised’ truck structure.
I would bet a dollar that those are pins at the tops of the plates over the axleboxes that have short links to horizontal eyes at the ends of the springs. The connection at the bottom of the ‘center’ plate is heavily gusseted at the bottom, which is telling me that the top of this spring slides (acting essentially like a ‘normal’ metallic-spring CCSB upside down) with the further joy that if the top of the spring perch is curved a bit, you have line contact which lets the whole business rock fore and aft as needed, so the support is two parallel line contacts across the four springs no matter how the trucks rotate.
Without the gusseted center plate there would be substantial stresses on the spring eyes, which would at best give significant pin wear and at worst cause distortion of the spring eyes or other components. Evidently the considerable extra unsprung mass did not outweigh the improvement in stiffness and resistance to distortion…
The photo of the tender of 999 is what I intended. But I do recall that thread posted on RPN. That started with someone asking about another odd tender truck as shown in the first post. I introduced this truck that we are discussing here into that thread. In looking into my files, I see that I was starting to write about it and ask questions in 2003. I can see what you are saying about the 999 trucks, and I think you could be right. That would fully explain how the truck could pivot.
A relatively late version of this type of truck was shown and described in an article by Joseph Follmar in the fall 1995 (Vol. 22, No. 4) issue of North Western Lines. The article was about C. St. P. M. & O. class I-1, 4-6-0s built around the turn of the century. This execution incorporates rather sophisticated castings. It appears as though it is equipped with raised buttresses that back the spring against tipping over. Overall, it is quite similar to the 999 trucks.
What I find to be very interesting is that there is a large set of seemingly unrelated track/train factors that, even when they are all within proper specification; can derail a train if they all happen to occur at the same time. One of these factors is relatively high train speed, so the derailment consequence will be a high speed derailment.
In other words, even if these random factors are all within the normal operating requirements of the train, they are almost certain to cause a high speed derailment if they all happen to occur at the same time. So it is like a game of chance, and given the deadly seriousness of the consequences, it reminds me of Russian roulette. The consequence I am referring to is the truck-hunting derailment.
Normally, one thinks of derailments being caused by something that goes wrong, such as a broken rail, broken wheel, broken flange, dropped drawbar, burned off axle, broken truck parts, shifting load, etc. But the truck-hunting derailment can occur from the right combination of cross level error, gage error, dry rail, continuous welded rail, empty car, relatively more flexible car body, and speed.
It’s that random possibility that causes the problem. And it might take a derailment, or hopefully a lesser incident, to make it known. Nobody designs such a defect into a car, or any other item.
It’s kind of like the rocking car issue on jointed rail. I seriously doubt that the railroads, or the railcar manufacturers, took that into account when designing/ordering new cars, especially a new length of cars. And those cars probably have run millions of miles without incident. Unless they are moving over jointed rail at a certain speed, in which case they’ll roll themselves right off the tracks.
There’s an article in the latest Classic Trains about just that problem.
What I find to be very interesting is that there is a large set of seemingly unrelated track/train factors that, even when they are all within proper specification; can derail a train if they all happen to occur at the same time. One of these factors is relatively high train speed, so the derailment consequence will be a high speed derailment.
In other words, even if these random factors are all within the normal operating requirements of the train, they are almost certain to cause a high speed derailment if they all happen to occur at the same time. So it is like a game of chance, and given the deadly seriousness of the consequences, it reminds me of Russian roulette. The consequence I am referring to is the truck-hunting derailment.
Normally, one thinks of derailments being caused by something that goes wrong, such as a broken rail, broken wheel, broken flange, dropped drawbar, burned off axle, broken truck parts, shifting load, etc. But the truck-hunting derailment can occur from the right combination of cross level error, gage error, dry rail, continuous welded rail, empty car, relatively more flexible car body,
As I have learned from racing Engineering is the ‘art’ of fixing what broke so it won’t break again. When that area of the vehicle has been engineered to not break, the next area of failure will then be discovered, to have a engineering solution applied to it’s fix and so on an so forth until every breakable area of the vehicle has been reengineered and the path to breakage begins all over again because the ‘fixed’ vehicle now permits increased speed and increased physical loads. This cycle applies to every thing humans use.
In thinking further, I definitely see the analogy between hunting-truck derailment problem and the harmonic rocking problem in that they both involve a pattern of repeating movement that builds in force in a self-perpetuating manner. Both are caused by factors that would not normally be thought capable of causing a problem. In the case of harmonic rocking/rollover, one factor is a speed range of approximately 17-22 mph. In the case of runaway hunting, two factors are continuous welded rail and dry weather. Who would have imagined that any of those three factors would cause derailments?
In the case of truck-hunting derailments, the “fix” was thought to be the Constant Contact Side Bearing. However, that fix seems to have proven to be inadequate. So the fix for that is to layer on lots of backup fixes such as just the right blend of inspection, maintenance, repair, rules for replacement, and hunting detectors.
But that fix for the fix seems to be too complex to be reliable as we see in the case of the Canadian derailment cited in the OP here. One problem is that the preload cannot be tested. Another problem is that it needs to be tested rather frequently because the life of the
My understanding was the the original CCSBs used metal springs and the ‘elastomer’ was considered an improvement (no determinate spring period).
If you remember John White’s comments in the American Passenger Car about rubber springs in general, it will come as little surprise that many of the reported problems with the ‘old’ elastomer CCSBs appear to be reported in Canadian winter time!
My own feeling is that the ‘ideal’ fix for this will involve some form of wave spring. I like your idea of a standard tool that can hook into ‘fixed’ points in the CCSB structure and (1) read the preload when levered down to a ‘datum’ position, and (2) give a force indication that can be read at the time a ‘feeler’ gauge or sheet of paper can be ‘just passed’ between the CCSB nose and its contacted surface as the tool is gradually depressed.
I got the impression that most of the ‘issues’ with inadequate CCSB performance were not caused by ‘sticking’ or cocking of the CCSB nose piece relative to the CCSB frame, but rather by some problem with the material or dimensions of the elastomer. Where the fun is going to start is what happens when the supposed significant number of these deflicted bearings starts to be detected in practice … can one of these be changed out ‘in the field’ just by jacking the car up a little witout needing to unhook the foundation or lift beyond the point where the center pivot is wholly disengaged? Or is it one of those situations where rust makes it difficult to separate things without cutting or hammering?
Here is another high speed derailment caused by truck hunting while the constant contact side bearings were failing to perform properly. In this one, it appears that the train initiating dynamic braking contributed to the rail climbing event that occurred during the truck hunting episode:
I observe many trains on BNSF & UP with these long centerbeam cars going through Rochelle on the webcam and locally in Downers Grove. Generally, the trains they are in are not going over 50 mph. Of course, I belive that Rochelle has a speed restriction over the plant. Do any of the USA engineers on the site know of speed restrictions for trains with these cars?
When you read this and the previous report I posted, plus the one Wizlish posted in the OP, you can see what a remarkable problem this really is. You can also see that this has been a learning experience, and not everything has been learned yet.
A truck-hunting derailment is quite a nasty event. Everything is running along fine until you get to speeds above 50 mph, and then a pattern of car ride instability begins to develop. It comes and goes in spasms that can get quite violent at times. Sometimes the whole car body will violently oscillate side to side several times per second. In this condition, a car could run 100 miles without derailing, but suffer thousands of violent hunting spasms with each taking the car right to the verge of derailing.
But CN took issue with the report. "CN challenged the RSA regarding truck hunting and indicated that the methodology was unsuitable and not consistent with normally used industry testing parameters." How many derailments do they need to convince them of the problem.
Once someone besides them offers to pay for the fix they will be more than happy to apply it. Also admitting you are wrong, even after this is proven beyond a doubt is a cardinal sin in the world of railroading.
I would bet a dollar that those are pins at the tops of the plates over the axleboxes that have short links to horizontal eyes at the ends of the springs. The connection at the bottom of the ‘center’ plate is heavily gusseted at the bottom, which is telling me that the top of this spring slides (acting essentially like a ‘normal’ metallic-spring CCSB upside down) with the further joy that if the top of the spring perch is curved a bit, you have line contact which lets the whole business rock fore and aft as needed, so the support is two parallel line contacts across the four springs no matter how the trucks rotate.