Hacker News Reader

Comment by ChuckMcM | Hacker News Reader

I got a chance to drive a couple of different engines at the Plumas rail museum. And one of the things that does is help you internalize the amazing amount of mass and momentum a train carries around all the time. The engineer who was with us spoke of freight trains that were a hundred cars of iron ore that, even though they were travelling at only 5mph would take nearly 1000' to bring to an "emergency" stop. While I don't know how true that is, just driving around the engine gave me some insight into how far ahead in time the engineer has to be thinking if they are operating a full train.

View on HN

My experience is largely with bicycles (you're very aware of the energy input/output), automobiles, and boats, but thinking of moving vehicles as momentum systems, to which energy can be added or removed, is useful.

On a bike, you maintain a steady speed by matching drag losses with the energy you're putting into the pedals. Braking and other actions cost you energy you've put in.

For automobiles and trucks, added mass has a significant change to momentum, and I feel that. Boats, from a row or sailboat to a 30m+ power vessel, have their momentum, drag, and interaction with water, waves, currents, and wind.

Trains are a whole 'nother ballgame.

> On a bike, you maintain a steady speed by matching drag losses with the energy you're putting into the pedals. Braking and other actions cost you energy you've put in.

As does the frequent mistake of putting weight on the rear pedal. Realising this made me understand how little energy it really takes to keep a bicycle going, given the correct gear.

As does the frequent mistake of putting weight on the rear pedal.

I don't follow. What do you mean here?

I think he means making sure to actively bring your rear leg forward while pedaling, so that your downstroke doesn't have to push against the weight of your rear leg too.

On a bike, you maintain a steady speed by matching drag losses with the energy you're putting into the pedals. Braking and other actions cost you energy you've put in.

Which is exactly why bicycles are so fond of rolling through stop signs. When you're riding one, it is obvious to you that going from 20 mph down to 10, then back up to 20 is reasonably safe and only takes half the effort as going from 20 mph down to 0 and then back up to 20.

It takes 3/4 of the effort - kinetic energy is proportional to the square of speed.

Not to mention that a person pedaling from a standstill has a lot less power output than a person pedaling an already moving bike.

Not really, and highly dependant on gearing.

Yes really. I'm sure you know power is velocity * force. For a long range, force output is pretty constant for humans, so power output must be lower: http://users.frii.com/katana/biketext.html

Nothing is redundant on a bike, if anything it is the other way around, e.g. the brakes are the wheels, the frame is the suspension, the gear mechanisms are also there to keep the chain on, the passenger is the power source. Nothing can be taken away.

Bikes also pioneer high tech lightweight materials, only recently have 'bicycle grade' aluminium and carbon fibre been used on those boxy car things.

It is a whole different world of 'body on frame' engineering. I sometimes wonder what a train would look like if it was designed from the ground up by a bicycle designer with a Lotus engineer - 'add lightness'.

Some modern 'Tesla truck' tech could be thrown in too.

Imagine having half the amount of wheels, all of them actively powered and actively steered so no screeching on corners. The seats could be 'bus stop grade' and the monocoque being some extruded tube with the hand rails and luggage racks being structural elements, like a rally car roll cage.

The idea of Brunel's wide gauge was to have the wheels at the side of the train rather than have the train sit above the tracks. Imagine if you had independent suspension and stepped a metre down into the train for some super low centre of gravity. Or if that extra space below the train was a massive parcel area so goods were shifted on/off trains with some standard pallets that just slotted in, aircraft style. Think how much traffic could be shifted off the roads particularly if you ran the trains in mixed freight mode through the night.

Rollercoaster style tracks, particularly in tube lines could be used to slow down trains at stations and speed them up afterwards.

As for the doors, these could be integrated into the compartment ends, so rather than there being the normal interconnect between compartments the doors would slide back/fore to keep the train enclosed on corners without the usual rubbery bendy bit needed.

Add airbags and seatbelts, windows that open, a bit of wifi and you could have comfort and actual passenger safety.

>> 5mph would take nearly 1000' to bring to an "emergency" stop.

That seems far fetched. 1000' is a long way at 5mph. The number of engines doesn't matter. Each car has its own brakes in an emergency. Cut pressure in the line and every car will brake itself. Perhaps they meant situations involving hills, not uncommon for ore trains heading from mines.

A train's emergency braking system is somewhat sacrificial (using it can damage engines, cars, etc.) so you're probably correct. I'd say that in either case, most people who are not "train people" will dramatically underestimate the required stopping distance. The mass of a fully loaded train is surprisingly large.

Car drivers tend to underestimate the amount of space it takes for a semi to stop. Heck, every time I load up my car or attach a trailer, it takes time to become accustomed to the decreased braking performance.

Drivers likely could not give an accurate estimation for how many feet it takes to do anything on the road, such as come to a stop in an SUV going 50. Given the extreme mass of a cargo train, I wouldn't be surprised if the distance figures sounded high.

Car drivers underestimate the amount of space it takes a car to stop.

> how many feet it takes to do anything on the road, such as come to a stop in an SUV going 50

Isn't knowledge of stopping distances for vehicles at various speeds and different road surface conditions a requirement for obtaining a driving license? (At least, in the UK I believe it is, and I think the Highway Code tables probably over-estimate these distances for safety purposes)

The Highway Code tables are based on an average family car in the 1960s (IIRC) when the tables were first introduced. They're ridiculously long distances for modern cars, though AIUI the "thinking time" part is nowadays thought to be too low.

Still, that's not a bad thing, per se. If we underestimate based on a vaguely remembered overestimated value from the highway code based on 1960s cars, then the effects cancel out and we're basically correct for modern vehicles?

I /think/ I recall seeing something that claimed with longer reaction times it actually ended up worse overall.

That said, I think often the bigger problem is people not realising especially in more unusual conditions (like snow in much of Britain) quite how much longer braking distances become.

Yes, but I don't think it requires or correlates to real understanding of what the distances are. I doubt that people understand how it varies with speed, or what that number of feet really looks like on the road. The process I have experienced is I read what a theoretical distance is in the booklet the state gave me for the exam, and then select their indicated answer on a multiple-choice test.

This is true. A buddy of mine works for the engine and brake systems of trains. I was helping him install the new skid plates and bumper he ordered from 4WheelOnline for his Jeep when he mentions the critical points and headaches of maintaining the brake system of trains.

A 30,000 ton ore train in Brazil will take about that long to stop, due to the nature of that weight strung out over 2.5 miles of distance.

That said, I think an emergency stop (air loss) would stop it faster.

> A 30,000 ton ore train in Brazil will take about that long to stop, due to the nature of that weight strung out over 2.5 miles of distance.

I once learned in a physics class that weight has nothing to do with the stopping distance, but rather the steel on steel friction coefficient.

Weight and ground pressure cancel each other out in the friction equation. The issue is where all that heat goes. Heated brakes will fade, glaze, fail, break, burn or a combination of the above, so there are limits on how much braking force you can apply to wheels

There are lots of assumptions in that entry-level physics exercise, for instance that all the energy can be dissipated without destroying anything (that the brakes are able to utilize the full friction available) and also that the entire weight of the vehicle rests on top of the wheels performing the braking.

Excellent point - it’s the rolling resistance indeed. I pointed out the size for reference (easier than saying 480,000 lbs rolling resistance :)

You don't want to lock up the wheels in an emergency. Locked wheels will skid on the rails, creating flat spots on the wheel. Later, if not repaired, the rotating wheel can hammer the rail at the flat spot, causing the rail to break.

It's an emergency. Who cares about flatspotted wheels? Parts can be replaced.

If you slip the wheels it will take longer to come to a stop. Sliding friction is always lower than static friction.

I agree, but the optimal achievable situation might be to be on the edge of locking wheels like ABS. Anyway, my point was that damaging the wheels is not the issue at hand.

It takes a while until the brakes in the last car start braking because the pressure doesn't drop everywhere simultaneously.

Why not also use an electric cable? Sounds like a problem, since the first car will have to "resist" the rest of the train.

Electricity wouldn't scale. Solenoids on each brake system would take current. Adding new cars would increase the load on whatever is providing that current. To ensure that cars could be swapped in and out, the cables on all cars would have to be large enough to provide current for even the longest train.

Air brakes, once pressurized, don't take much to maintain. Relatively small pipes work on even very long trains. You can add cars all day without increasing the steady load on the compressor. A longer train will take longer to pressurize, but once there it would not need the compressor to run constantly. Air is also not as dangerous as running large electrical cables the length of a train.

I meant electricity just for the triggering, not for the action. A small electric signal could for example open a valve in the air tube. Thus all cars would break at the same time.

Some trains already to this, where they put in electronic valves in the air lines for in car. Basically, the electronic valve controls if the existing brake is vented or connected to the central air line. This still allows for the fail-safe behavior of being able to brake the train from the central air line by venting it.

This lets them start easier/quicker from a stop, as they don't have to re-air the entire train and less wear on the brakes of the trains in the front of the train as all the cars start braking at the same time. This is particularly useful for trains that have to start on a hill as taking minutes for your brakes to fully turn off can complicate things.

Passenger trains have used electropneumatic brakes for decades. A few freight trains use them. Trains with distributed power (locomotives spaced along the train) have sectionally controlled braking. Sometimes, there are "repeater air cars" with air compressors and a radio link to the locomotive. Those are used mostly on flat terrain, where more braking power is needed but not more engine power.

Because the pneumatic system is fail-secure, since a break in the line causes the pressure to drop and brakes to engage. I suppose you could have a system which required constant power to hold the brakes open, and loss of power would cause them to engage, which would have the same result?

"Shortly after 08:00 hrs on 8 November 2010, a passenger train running from London Charing Cross to Hastings failed to stop at Stonegate station in East Sussex. The train ran for a further 2.45 miles (3.94 km) with the emergency brake applied, passing the level crossing at Crowhurst Bridge before coming to a stop 3.22 miles (5.18 km) after first applying the brakes."

There's a whole sequence of events behind this, from the work to remove slippery leaves from rails, to the sand applied during braking if wheels are slipping, to the IT systems and management processes identifying why the train didn't have any sand. It's a fascinating 151 page report.

https://www.gov.uk/raib-reports/station-overrun-incident-at-... (PDF report linked at the bottom).

It was going far faster than 5mph. 5mph is little more than walking pace.

True, but the last part of the graph shows what I think is fairly normal braking from 24mph to 0.

Counting pixels, the distance covered from 10mph to zero is about 350 feet. With this distance for a lightly loaded passenger train, 1000 feet doesn't seem impossible or even unreasonable for a heavy freight train at 5mph.

Reply on news.ycombinator.com