Arcane question of the week
|
Dave Nelson
How far above the railhead would you say is the center of gravity for your average empty STMFC?
Really.
Our software team has added curve resistance to the rolling resistance values and while doing so tossed in something else for tipping over at some relationship of curve radius, super-elevation, and speed. I think it’s far too sensitive but I need a decent center of gravity estimate to make the case.
Dave Nelson |
||
|
|
||
|
John Barry
A lot lower on a flat car than a house car, with a gondola somewhere in between. Seriously, the mass distribution propoerties vary not only by car type but by construction. The trucks were generally a significant portion of the weight, but the underframe and superstructure were also significant portions. My gut feel is that the center of mass moved downward somewhat as construction methods improved. Welded Z frame center sills were much lighter than the bolted C channels for instance. Using the same trucks with less weight above them would lower the CG. Good question even though most of us don't have the data to answer it. John Barry ATSF North Bay Lines Golden Gates & Fast Freights 707-490-9696 PO Box 44736 Washington, DC 20026-4736 From: "'Dave Nelson' Lake_Muskoka@... [STMFC]" To: STMFC Sent: Saturday, March 19, 2016 12:02 AM Subject: [STMFC] Arcane question of the week How far above the railhead would you say is the center of gravity for your average empty STMFC? Really. Our software team has added curve resistance to the rolling resistance values and while doing so tossed in something else for tipping over at some relationship of curve radius, super-elevation, and speed. I think it’s far too sensitive but I need a decent center of gravity estimate to make the case. Dave Nelson |
||
|
|
||
|
Alex Schneider
Presumably consideration was also given to the fact that the carbody.and trucks were only held together by center pins, although there were pads on both sides of the bolsters which could support weight. Thus the body could tip off the trucks as well as the trucks tipping off the rails. Alex Schneider Sent from my Verizon Wireless 4G LTE smartphone -------- Original message -------- From: "John Barry northbaylines@... [STMFC]" <STMFC@...> Date: 03/18/2016 23:34 (GMT-06:00) To: STMFC@... Subject: Re: [STMFC] Arcane question of the week A lot lower on a flat car than a house car, with a gondola somewhere in between. Seriously, the mass distribution propoerties vary not only by car type but by construction. The trucks were generally a significant portion of the weight, but the underframe and superstructure were also significant portions. My gut feel is that the center of mass moved downward somewhat as construction methods improved. Welded Z frame center sills were much lighter than the bolted C channels for instance. Using the same trucks with less weight above them would lower the CG. Good question even though most of us don't have the data to answer it. John Barry ATSF North Bay Lines Golden Gates & Fast Freights 707-490-9696 PO Box 44736 Washington, DC 20026-4736 From: "'Dave Nelson' Lake_Muskoka@... [STMFC]" <STMFC@...> To: STMFC <STMFC@...> Sent: Saturday, March 19, 2016 12:02 AM Subject: [STMFC] Arcane question of the week How far above the railhead would you say is the center of gravity for your average empty STMFC? Really. Our software team has added curve resistance to the rolling resistance values and while doing so tossed in something else for tipping over at some relationship of curve radius, super-elevation, and speed. I think it’s far too sensitive but I need a decent center of gravity estimate to make the case. Dave Nelson |
||
|
|
||
|
Nelson Moyer <ku0a@...>
The car standards for the La Mesa Model Railroad Club are by far the most
stringent standards that I'm aware of, and I pasted in the center of gravity standard in answer to your question. I. Center of Gravity 1. Cars should be constructed to obtain the lowest center of gravity. The minimum CG for any car is CG30. 2. ANY weight added to a car to make its' required weight should be kept as low as possible (NMRA RP20.1). 3. The center of gravity of each piece of rolling stock (locomotives excepted) shall be determined on a Protractor or Tilt Table and noted as its Center of Gravity Index, (or CG)expressed in degrees, i.e., CG35. This shall be the farthest deflection from the normal upright (0 degree) or vertical position to which the car can be tilted, from a flat, level, horizontal surface (90 degrees) on which it stands without tipping over. This will be determined by using a vertical protractor arm moving against the flat vertical side of the car, as the car is tipped sideways. The protractor pivot center must be coincident with said horizontal flat surface. Adapter blocks shall be used to establish a flat plane for the measurement of equipment not having flat sides. A high CG index number indicates a low center or gravity, i.e: CG90. 4. Rolling stock will be tilt (protractor) tested on both sides of car to establish the CG Index. The poorest of the two readings (the lowest reading) will become the CG Index for that car. Side to side differences exceeding 5 degrees should be corrected with counterweights or by relocating/removing added weights. 5. Equipment may be required to be modified as necessary to achieve the optimum weight/center of gravity. 6. The MINIMUM center of gravity index for any NEW registration car shall be 30 degrees (CG30), brass models MAY be exempted from modification that would ruin their value- provided they cause no operating problems. 7. Existing equipment registered before the establishment of the CG index WILL NOT be exempted, except on a "proof only' basis, and will need to be upgraded to the current standard when repaired or inspected. Nelson Moyer From: STMFC@... [mailto:STMFC@...] Sent: Friday, March 18, 2016 11:02 PM To: STMFC <STMFC@...> Subject: [STMFC] Arcane question of the week How far above the railhead would you say is the center of gravity for your average empty STMFC? Really. Our software team has added curve resistance to the rolling resistance values and while doing so tossed in something else for tipping over at some relationship of curve radius, super-elevation, and speed. I think it's far too sensitive but I need a decent center of gravity estimate to make the case. Dave Nelson |
||
|
|
||
|
destorzek@...
---In STMFC@..., <Lake_Muskoka@...> wrote : How far above the railhead would you say is the center of gravity for your average empty STMFC?
Really.
Our software team has added curve resistance to the rolling resistance values and while doing so tossed in something else for tipping over at some relationship of curve radius, super-elevation, and speed. I think it’s far too sensitive but I need a decent center of gravity estimate to make the case.
Dave Nelson ============= I've never seen a discussion of this in the traditional trade literature, likely because it only affected passenger trains, and then, the locomotive, being an iron tank filled with water mounted above a rather tall running gear, would have been affected before the train, so your best bet would be to search period literature for the effects of CoG on locomotives. Dennis Storzek |
||
|
|
||
|
Not trade literature, but a 1947 Scientific American article http://www.scientificamerican.com/article/the-physics-of-disaster/ Tim O'Connor I've never seen a discussion of this in the traditional trade literature, likely because it only affected passenger trains, and then, the locomotive, being an iron tank filled with water mounted above a rather tall running gear, would have been affected before the train, so your best bet would be to search period literature for the effects of CoG on locomotives. Dennis Storzek |
||
|
|
||
|
spsalso
I read this in the (1947) article cited: "Amtrak’s 150-mph (241-km/h) Acela creates its own bank angle by tilting up to 4.2 degrees. If the Acela is operating on a curve whose out-side rail is raised 2 inches (5 cm), the Acela can speed as if it is on a curve that is raised an additional 7 inches (17.8 cm) higher—for a total super-elevation of 9 inches (22.9 cm)." And it would appear to do that without significantly changing the height of its center of gravity. In the context of this discussion of tipping forces, that's quite an accomplishment. Ed Edward Sutorik |
||
|
|
||
|
spsalso
On a bit more thought, perhaps the Acela moves its center of gravity sideways to accomplish this feat. Since the cars tip, perhaps the center of gravity really can be moved enough to make the kind of difference noted in the article. Which is quite interesting. Thanks, Tim Ed Edward Sutorik |
||
|
|
||
|
Jack Mullen
Dave,
That could be the arcane question of the month, if not the year. As I'm sure your are aware, this isn't readily found. Some values can be picked up in reading track-train dynamics studies and derailment analyses, most of them well past steam era. Currently (for those of you whose calendars don't turn past 1960, that means WAY past steam era) AAR interchange rules limit combined car+load CG height to 98". It's been remarked that it's not possible for a loaded Plate C boxcar to exceed this value, so if you assume a load mass equal to the LD LMT, located 1/2 IH above the floor, you could derive an upper bound for the car's CG - for a modern car of course. And UMLER now has a field for empty CG height,though it's only a required entry for certain cars. Looking back to the steam era, or closer to it, at least for steam-era equipment., I can offer a few random tidbits. My references are packed for moving, so this is dimly recalled gleanings from the distant past. For a flatcar, something like 3' or a few inches less should be a reasonable value. For a boxcar, IIRC the empty CG is generally within a foot or less above the floor. Having gone through the rock-and-roll derailment era of the '70s and '80s, I think large covered hoppers tended to have an empty CG around 5', and loaded around 8', and both values were regarded as higher than average, so again that may help with an upper limit. For steam locos themselves, as a rule-of-thumb, the height of the CG should be around the bottom of the boiler. It's not clear to me whether you're talking about tipping to the inside of a curve in a stringline derailment, or overturning outside in an excess-speed derailment. Since speed is mentioned, I think it may be the latter. Empty-car CG should be low enough that overturning shouldn't occur at any plausible combination of train speed and curve, so if the model is predicting otherwise, something is wrong. In practical terms, it's likely that an L/V derailment, either wheel-climb or rail rollover would occur sooner. Jack Mullen Dave Nelson asked: How far above the railhead would you say is the center of gravity for your average empty STMFC? Really. Our software team has added curve resistance to the rolling resistance values and while doing so tossed in something else for tipping over at some relationship of curve radius, super-elevation, and speed. I think it's far too sensitive but I need a decent center of gravity estimate to make the case. Dave Nelson [Non-text portions of this message have been removed] |
||
|
|
||
|
Dave Nelson
Who better to ask than the members of this list? You guys are all great.
But it is still an interesting question… gets you thinking of how the weight difference between a flat car and boxcar of reasonably similar construction translates into a different center of gravity… I would not think that much given the sides and roof panels for a boxcar are so thin. In turn, this leads on to the question of why do empty covered hoppers have such a high center of gravity? Those sides must be much more substantial than I thought.
Last… I quite enjoyed reading about the test procedure in San Diego. It isn’t readily apparent to most model railroaders but long trains are really different.
Thanks all for your comments,
Dave Nelson
From: STMFC@... [mailto:STMFC@...]
Sent: Saturday, March 19, 2016 1:38 PM To: STMFC@... Subject: RE: [STMFC] Arcane question of the week
That could be the arcane question of the month, if not the year….
It's not clear to me whether you're talking about tipping to the inside of a curve in a stringline derailment, or overturning outside in an excess-speed derailment. Since speed is mentioned, I think it may be the latter. Empty-car CG should be low enough that overturning shouldn't occur at any plausible combination of train speed and curve, so if the model is predicting otherwise, something is wrong. In practical terms, it's likely that an L/V derailment, either wheel-climb or rail rollover would occur sooner.
Jack Mullen |
||
|
|
||
|
Schuyler Larrabee
“Any super elevation in quite a bit of extra safety.” [sic]
So wrote Dave Nelson. And for preventing a train or car from rolling off to the outside of the curve at speed, sure, that makes sense.
But what about a train that has to stop on a curve with significant superelevation? The ERIE ran through a town called Great Bend, south of Binghamton in Penna., so named for the great bend in the Susquehanna River at that point. The curve on the ERIE was quite sharp, and the superelevation was really significant, I believe 8” difference in the tops of the rails. At least 7”, but I think 8”. Freights ran through Great Bend at 50 mph, or better, so the superelevation was a good thing, and probably even more so for passengers. But I saw a couple of freight trains stopped on this curve, and it was quite intimidating to see cars leaning that far. ISTR that the employee timetable for the Susquehanna Division included special instructions about this curve, including advisories about how to start a stopped train to avoid overturning a car. Unfortunately, my library has never fully recovered from the move, and I can’t put my hands on them now.
Schuyler
From: STMFC@... [mailto:STMFC@...]
Sent: Saturday, March 19, 2016 6:07 PM To: STMFC@... Subject: RE: [STMFC] Arcane question of the week
Who better to ask than the members of this list? You guys are all great.
But it is still an interesting question… gets you thinking of how the weight difference between a flat car and boxcar of reasonably similar construction translates into a different center of gravity… I would not think that much given the sides and roof panels for a boxcar are so thin. In turn, this leads on to the question of why do empty covered hoppers have such a high center of gravity? Those sides must be much more substantial than I thought.
Last… I quite enjoyed reading about the test procedure in San Diego. It isn’t readily apparent to most model railroaders but long trains are really different.
Thanks all for your comments,
Dave Nelson
From: STMFC@... [mailto:STMFC@...]
That could be the arcane question of the month, if not the year….
It's not clear to me whether you're talking about tipping to the inside of a curve in a stringline derailment, or overturning outside in an excess-speed derailment. Since speed is mentioned, I think it may be the latter. Empty-car CG should be low enough that overturning shouldn't occur at any plausible combination of train speed and curve, so if the model is predicting otherwise, something is wrong. In practical terms, it's likely that an L/V derailment, either wheel-climb or rail rollover would occur sooner.
Jack Mullen |
||
|
|
||
|
Jack Mullen
Dave Nelson said:
Consider that a box car underframe has a lot less steel than a flat car. It's not so much a matter of adding the weight of the boxcar superstructure as redistributing mass higher above the rail. I'm still going from memory here, so feel free to fact-check, but I think the USRA box cars were no more than a ton or so heavier than the composite gon, and likewise to the proposed USRA flat cars. For another example, C&NW had some 50' merchandise box cars built prewar that were under 50000# light weight, well within the range of 50 ft, 50T flat cars. Admittedly, that's a somewhat light 50' box, but by no means unique. I was thinking specifically of 4600 CF and larger grain cars, where much of the car structure has been reduced to what's basically a big rectangular steel tube. Again, not really a matter of how heavy the sides are, as how much of the total body is fairly high up. Jack Mullen |
||
|
|
||
|
mark_landgraf
Taken from an NKP Center of Gravity Diagram book - date unknown - presume 1930 to 1935 Tank Car - 10,000 gallons - mty wgt 46,064 lbs - CoG = 48.25 inches loaded with 10248 gallons of gasoline - actual wght 105,862 - CoG 72.187 inches loaded with 10,248 gallons of mineral oil - actual wght 125.656 - CoG 75 inches Coal Hopper - 50 ton - mty wght 39,600 - CoG = 39.0625 inches loaded with a 10 inch heap - actual wght 140883 - CoG = 70.625 inches loaded with a 30° heap - actual wght 151,632 - CoG = 63.125 inches Gondola - 50 ton - mty wght 44,100 lbs - CoG = 33.125 inchesg loaded with coal - 10 inch heap - act wght 143341 - CoG = 63.75 inches Loaded with coal - 30° heap - actual wght 157445 - CoG = 68.5 inches Boxcar - 40 ton - 3736 cuft - mty wght 46450 lbs - CoG = 56.0625 inches loaded with oats - actual wght 136000 - CofG = 84.75 inches Loaded with corn - acyual wght 136000 - CoG = 69.0625 inches Reefer ARTCo blt 1926-1927 - 40 tons - mty wght 52362 - CoG = 61.5 inches mty car with 9615 lbs of ice in bunkers - actual wght 61977 - CoG = 69.03 inches Super elevation back prior to the FRA, and when cars were not as tall was 7 inches max where passenger trains were operated, and the speed & radius required it. For frght only lines the max was 4 to 5 inches max. Generally 1.5 inches of underbalance was - and still is - a safe maximum for frgt cars. Passenger trains normally operated at 3 inches of underbalance. Underbalance is how many inches a pendulum deflects while going around a curve. Envision an equilateral triangle with 56.5 inch side set up inside a car. A pendulum attached to its upper junction point. On straight track the pendulum hangs straight down at the zero point on the scale. Then we head into curve with no superelevation, the pendulum deflects outward 1½ inches. If that same curve had 1½ inches of superelevation, and the train was moving at the same speed as above, the pendulum would stay at zero. Hint - the triangle uses legs that are the same length as track gauge. Even today's taller hicube cars when stopped on a curve with 4 inches of SE, still need a good crosswind - say 60+ MPH - before there is a likelihood of an mty car blowing over. Loaded cars would take tornado like winds to blow them over. Today you would need about 20 inches of superelevation to reach where a car would tip over. The actual angle would be determined by a triangle formed with a base of 56.5 inches and the other two equal legs of the length from the gauge point to the CoG elevation on the centerline of the car. When the CoG point tips far enough so that it is outside to the gauge line, the car tips over. Mark Landgraf Albany NY railroad office engineer From: "jack.f.mullen@... [STMFC]" To: STMFC@... Sent: Saturday, March 19, 2016 10:14 PM Subject: RE: [STMFC] Arcane question of the week Dave Nelson said:
Consider that a box car underframe has a lot less steel than a flat car. It's not so much a matter of adding the weight of the boxcar superstructure as redistributing mass higher above the rail. I'm still going from memory here, so feel free to fact-check, but I think the USRA box cars were no more than a ton or so heavier than the composite gon, and likewise to the proposed USRA flat cars. For another example, C&NW had some 50' merchandise box cars built prewar that were under 50000# light weight, well within the range of 50 ft, 50T flat cars. Admittedly, that's a somewhat light 50' box, but by no means unique. I was thinking specifically of 4600 CF and larger grain cars, where much of the car structure has been reduced to what's basically a big rectangular steel tube. Again, not really a matter of how heavy the sides are, as how much of the total body is fairly high up. Jack Mullen
|
||
|
|
||
|
Dave Nelson
Whoa, now that’s what I like… real data.
Thanks so much for posting it. Dave Nelson From: STMFC@... [mailto:STMFC@...] Sent: Sunday, March 20, 2016 6:01 PM To: STMFC@... Subject: Re: [STMFC] Arcane question of the week Taken from an NKP Center of Gravity Diagram book - date unknown - presume 1930 to 1935 Tank Car - 10,000 gallons - mty wgt 46,064 lbs - CoG = 48.25 inches loaded with 10248 gallons of gasoline - actual wght 105,862 - CoG 72.187 inches loaded with 10,248 gallons of mineral oil - actual wght 125.656 - CoG 75 inches Coal Hopper - 50 ton - mty wght 39,600 - CoG = 39.0625 inches loaded with a 10 inch heap - actual wght 140883 - CoG = 70.625 inches loaded with a 30° heap - actual wght 151,632 - CoG = 63.125 inches Gondola - 50 ton - mty wght 44,100 lbs - CoG = 33.125 inchesg loaded with coal - 10 inch heap - act wght 143341 - CoG = 63.75 inches Loaded with coal - 30° heap - actual wght 157445 - CoG = 68.5 inches Boxcar - 40 ton - 3736 cuft - mty wght 46450 lbs - CoG = 56.0625 inches loaded with oats - actual wght 136000 - CofG = 84.75 inches Loaded with corn - acyual wght 136000 - CoG = 69.0625 inches Reefer ARTCo blt 1926-1927 - 40 tons - mty wght 52362 - CoG = 61.5 inches mty car with 9615 lbs of ice in bunkers - actual wght 61977 - CoG = 69.03 inches Super elevation back prior to the FRA, and when cars were not as tall was 7 inches max where passenger trains were operated, and the speed & radius required it. For frght only lines the max was 4 to 5 inches max. Generally 1.5 inches of underbalance was - and still is - a safe maximum for frgt cars. Passenger trains normally operated at 3 inches of underbalance. Underbalance is how many inches a pendulum deflects while going around a curve. Envision an equilateral triangle with 56.5 inch side set up inside a car. A pendulum attached to its upper junction point. On straight track the pendulum hangs straight down at the zero point on the scale. Then we head into curve with no superelevation, the pendulum deflects outward 1½ inches. If that same curve had 1½ inches of superelevation, and the train was moving at the same speed as above, the pendulum would stay at zero. Hint - the triangle uses legs that are the same length as track gauge. Even today's taller hicube cars when stopped on a curve with 4 inches of SE, still need a good crosswind - say 60+ MPH - before there is a likelihood of an mty car blowing over. Loaded cars would take tornado like winds to blow them over. Today you would need about 20 inches of superelevation to reach where a car would tip over. The actual angle would be determined by a triangle formed with a base of 56.5 inches and the other two equal legs of the length from the gauge point to the CoG elevation on the centerline of the car. When the CoG point tips far enough so that it is outside to the gauge line, the car tips over. Mark Landgraf Albany NY railroad office engineer _____ From: "jack.f.mullen@... [STMFC]" <STMFC@...> To: STMFC@... Sent: Saturday, March 19, 2016 10:14 PM Subject: RE: [STMFC] Arcane question of the week Dave Nelson said: … gets you thinking of how the weight difference between a flat car and boxcar of reasonably similar construction translates into a different center of gravity… I would not think that much given the sides and roof panels for a boxcar are so thin. Consider that a box car underframe has a lot less steel than a flat car. It's not so much a matter of adding the weight of the boxcar superstructure as redistributing mass higher above the rail. I'm still going from memory here, so feel free to fact-check, but I think the USRA box cars were no more than a ton or so heavier than the composite gon, and likewise to the proposed USRA flat cars. For another example, C&NW had some 50' merchandise box cars built prewar that were under 50000# light weight, well within the range of 50 ft, 50T flat cars. Admittedly, that's a somewhat light 50' box, but by no means unique. In turn, this leads on to the question of why do empty covered hoppers have such a high center of gravity? Those sides must be much more substantial than I thought. I was thinking specifically of 4600 CF and larger grain cars, where much of the car structure has been reduced to what's basically a big rectangular steel tube. Again, not really a matter of how heavy the sides are, as how much of the total body is fairly high up. Jack Mullen [Non-text portions of this message have been removed] |
||
|
|
||
|
Nelson Moyer <ku0a@...>
While real data is interesting, it doesn’t scale to 1:87, so it has little to no modeling value. After all, how many of us load out tank cars, etc. with real commodities?
Nelson Moyer From: STMFC@... [mailto:STMFC@...] Sent: Monday, March 21, 2016 11:03 PM To: STMFC@... Subject: RE: [STMFC] Arcane question of the week Whoa, now that’s what I like… real data. Thanks so much for posting it. Dave Nelson From: STMFC@... [mailto:STMFC@...] Sent: Sunday, March 20, 2016 6:01 PM To: STMFC@... Subject: Re: [STMFC] Arcane question of the week Taken from an NKP Center of Gravity Diagram book - date unknown - presume 1930 to 1935 Tank Car - 10,000 gallons - mty wgt 46,064 lbs - CoG = 48.25 inches loaded with 10248 gallons of gasoline - actual wght 105,862 - CoG 72.187 inches loaded with 10,248 gallons of mineral oil - actual wght 125.656 - CoG 75 inches Coal Hopper - 50 ton - mty wght 39,600 - CoG = 39.0625 inches loaded with a 10 inch heap - actual wght 140883 - CoG = 70.625 inches loaded with a 30° heap - actual wght 151,632 - CoG = 63.125 inches Gondola - 50 ton - mty wght 44,100 lbs - CoG = 33.125 inchesg loaded with coal - 10 inch heap - act wght 143341 - CoG = 63.75 inches Loaded with coal - 30° heap - actual wght 157445 - CoG = 68.5 inches Boxcar - 40 ton - 3736 cuft - mty wght 46450 lbs - CoG = 56.0625 inches loaded with oats - actual wght 136000 - CofG = 84.75 inches Loaded with corn - acyual wght 136000 - CoG = 69.0625 inches Reefer ARTCo blt 1926-1927 - 40 tons - mty wght 52362 - CoG = 61.5 inches mty car with 9615 lbs of ice in bunkers - actual wght 61977 - CoG = 69.03 inches Super elevation back prior to the FRA, and when cars were not as tall was 7 inches max where passenger trains were operated, and the speed & radius required it. For frght only lines the max was 4 to 5 inches max. Generally 1.5 inches of underbalance was - and still is - a safe maximum for frgt cars. Passenger trains normally operated at 3 inches of underbalance. Underbalance is how many inches a pendulum deflects while going around a curve. Envision an equilateral triangle with 56.5 inch side set up inside a car. A pendulum attached to its upper junction point. On straight track the pendulum hangs straight down at the zero point on the scale. Then we head into curve with no superelevation, the pendulum deflects outward 1½ inches. If that same curve had 1½ inches of superelevation, and the train was moving at the same speed as above, the pendulum would stay at zero. Hint - the triangle uses legs that are the same length as track gauge. Even today's taller hicube cars when stopped on a curve with 4 inches of SE, still need a good crosswind - say 60+ MPH - before there is a likelihood of an mty car blowing over. Loaded cars would take tornado like winds to blow them over. Today you would need about 20 inches of superelevation to reach where a car would tip over. The actual angle would be determined by a triangle formed with a base of 56.5 inches and the other two equal legs of the length from the gauge point to the CoG elevation on the centerline of the car. When the CoG point tips far enough so that it is outside to the gauge line, the car tips over. Mark Landgraf Albany NY railroad office engineer _____ From: "jack.f.mullen@... [STMFC]" <STMFC@...> To: STMFC@... Sent: Saturday, March 19, 2016 10:14 PM Subject: RE: [STMFC] Arcane question of the week Dave Nelson said: … gets you thinking of how the weight difference between a flat car and boxcar of reasonably similar construction translates into a different center of gravity… I would not think that much given the sides and roof panels for a boxcar are so thin. Consider that a box car underframe has a lot less steel than a flat car. It's not so much a matter of adding the weight of the boxcar superstructure as redistributing mass higher above the rail. I'm still going from memory here, so feel free to fact-check, but I think the USRA box cars were no more than a ton or so heavier than the composite gon, and likewise to the proposed USRA flat cars. For another example, C&NW had some 50' merchandise box cars built prewar that were under 50000# light weight, well within the range of 50 ft, 50T flat cars. Admittedly, that's a somewhat light 50' box, but by no means unique. In turn, this leads on to the question of why do empty covered hoppers have such a high center of gravity? Those sides must be much more substantial than I thought. I was thinking specifically of 4600 CF and larger grain cars, where much of the car structure has been reduced to what's basically a big rectangular steel tube. Again, not really a matter of how heavy the sides are, as how much of the total body is fairly high up. Jack Mullen [Non-text portions of this message have been removed] [Non-text portions of this message have been removed] |
||
|
|
||
|
destorzek@...
I thought the original request was for data to be used in programing a train simulator???
Dennis Storzek |
||
|
|
||
|
Dave Nelson
Dennis, is correct… the question was asked on behalf of the Open Rails simulator (free, Open Source license). The team added curve resistance and while doing so implemented a first pass at dealing with the unbalancing effect that occurs when a car moves thru a curve (what super elevation compensates). Unfortunately the lack of good data on typical center of gravity and safe levels of unbalancing made the early formulas way too sensitive. The answers provided here have been most helpful… we’ll create a number of defaults to use in the absence of data coming from specific cars and I, for one, will be adding more accurate data to most.
There is still a bit of fudging going on WRT air resistance but other than that the simulator now calculates rolling resistance for all cars and locomotives many times per second, factoring in car weight, solid or roller bearings, grade, and curves in one version of the Davis Formula. I’m ever hopeful I can persuade one of the programmers to go for buff and draft effects on the full train but they’re leery of a potential frame performance loss. Any other questions? Dave Nelson From: STMFC@... [mailto:STMFC@...] Sent: Tuesday, March 22, 2016 10:46 AM To: STMFC@... Subject: RE: [STMFC] Arcane question of the week I thought the original request was for data to be used in programing a train simulator??? Dennis Storzek [Non-text portions of this message have been removed] |
||
|
|
||
|
mark_landgraf
Dave Most train performance calculators do use the Davis equation. One of the most important factors is to keep the flanges as close to centered on the gauge as possible. This will occur when operating on track that is superelevated so that trains operating over are operating at 0 inches UB. Operating either faster or slower moves the flanges either out or in, but either causes more rolling resistance. Aerodynamics gets bad when going faster than 70 mph. Don't get me wrong, a train of solid highcubes moves cleaner than a mixed train of boxes, gons and flats. The fuel cost and horsepower curve crosses the aerodynamic speed curve at about 70 mph. The additional HP to get above 70, and the cost of fuel, just don't justify the minimal time savings. About 1980, the USDOT had a free train performance calculator that was written by a university in Cambridge MA. That was the era of TPCs. Today you use TPC runs to calibrate transportation network simulators. Have you tried celebrating the model with real data? I realize that you would need to collect data by observation. If you have a track chart for a real territory, preferably one you can drive on the road next to the tracks for an extended period, you can build the track library. You should be able to build a library of loco characteristics from engine builders. Gathering the consist can can be done by video. Pacing the train will provide speed and timing. Mark Landgraf Sent from my BlackBerry 10 smartphone on the Verizon Wireless 4G LTE network.
Dennis, is correct… the question was asked on behalf of the Open Rails simulator (free, Open Source license). The team added curve resistance and while doing so implemented a first pass at dealing with the unbalancing effect that occurs when a car moves thru a curve (what super elevation compensates). Unfortunately the lack of good data on typical center of gravity and safe levels of unbalancing made the early formulas way too sensitive. The answers provided here have been most helpful… we’ll create a number of defaults to use in the absence of data coming from specific cars and I, for one, will be adding more accurate data to most. From: STMFC@... [mailto:STMFC@...]
Sent: Tuesday, March 22, 2016 10:46 AM To: STMFC@... Subject: RE: [STMFC] Arcane question of the week I thought the original request was for data to be used in programing a train simulator??? Dennis Storzek [Non-text portions of this message have been removed] |
||
|
|
||