How do tanks turn efficiently?
August 4, 2009 5:36 AM Subscribe
Tanks have long, high-friction tracks - good for going forwards, not so good for turning as the tracks would have to skid. What do tanks do to overcome this problem?
It's reasonable to speculate that they have high-power engines, that they have a selection of low gears, and that they try to avoid making too many sharp turns, but are there any other design features that overcome this problem?
It's reasonable to speculate that they have high-power engines, that they have a selection of low gears, and that they try to avoid making too many sharp turns, but are there any other design features that overcome this problem?
One goes faster than the other. It tears up the ground horribly. Fortunately, that's the enemy's problem.
posted by dmd at 5:44 AM on August 4, 2009 [1 favorite]
posted by dmd at 5:44 AM on August 4, 2009 [1 favorite]
if the treads are steel, they're only going to be high-friction on soft surfaces such as soil, where they bite deeply; they should slip pretty easily on hard surfaces like concrete.
posted by jon1270 at 5:48 AM on August 4, 2009
posted by jon1270 at 5:48 AM on August 4, 2009
It may sound ridiculous, but everything I learned about tank maneuverability, I learned from playing Battlezone. If you can find one, go play!
posted by Cat Pie Hurts at 5:49 AM on August 4, 2009
posted by Cat Pie Hurts at 5:49 AM on August 4, 2009
Each tread is independently controlled. One can go faster than the other for a gentle turn or they can go in opposite directions to turn in place.
posted by exogenous at 5:58 AM on August 4, 2009
posted by exogenous at 5:58 AM on August 4, 2009
This is a huge problem for all tracked vehicles. This is why you see rubber tire loaders and such forth at construction sites when vehicles need to work on pavement or other damageable ground. Turning tracked vehicles tears the ground up big time. But yeah tracked vehicles (fully-tracked, not half-tracks) can generally pivot in place by running their tracks in opposite directions. I don't really know about tanks, specifically, but I've operated a lot of heavy construction equipment.
posted by jeb at 6:02 AM on August 4, 2009
posted by jeb at 6:02 AM on August 4, 2009
Best answer: Designing the tank so that it's as 'square' as possible helps prevent skidding on turning as longer thinning vehicle would tend to skid more. And keeping the weight in the middle.
posted by fearfulsymmetry at 6:05 AM on August 4, 2009
posted by fearfulsymmetry at 6:05 AM on August 4, 2009
Best answer: I wonder whether it's possible to raise some of the wheels that keep the track on the ground. If most of the weight were supported on just a couple of wheels towards the middle of each track, there wouldn't be much of a problem.
posted by jon1270 at 6:09 AM on August 4, 2009
posted by jon1270 at 6:09 AM on August 4, 2009
Sort of related -- here is a video of a Bradley doing a powerslide on concrete:
posted by Jinkeez at 6:14 AM on August 4, 2009 [1 favorite]
Response by poster: Thanks to those of you who have responded so far!
Perhaps I should clarify - I am already aware that tanks can turn on the spot by running one track forwards and the other backwards. What I'm interested in is overcoming friction when doing that, as to do it the tracks have to skid sideways (see the "Power Requirements" section at the bottom of this page for a diagram of why the tracks have to skid - to turn a tank on the spot, to the right, the fronts of the tracks have to skid right and the rears of the tracks have to skid left).
I'm hoping to hear about how tanks are designed to minimise this problem; Posts like those from fearfulsymmetry and jon1270 are exactly the sorts of things I'm looking for.
posted by Mike1024 at 6:44 AM on August 4, 2009
Perhaps I should clarify - I am already aware that tanks can turn on the spot by running one track forwards and the other backwards. What I'm interested in is overcoming friction when doing that, as to do it the tracks have to skid sideways (see the "Power Requirements" section at the bottom of this page for a diagram of why the tracks have to skid - to turn a tank on the spot, to the right, the fronts of the tracks have to skid right and the rears of the tracks have to skid left).
I'm hoping to hear about how tanks are designed to minimise this problem; Posts like those from fearfulsymmetry and jon1270 are exactly the sorts of things I'm looking for.
posted by Mike1024 at 6:44 AM on August 4, 2009
From the quoted page:
In order to steer a tracked vehicle, it is necessary to drive one track faster than the other, causing the vehicle to turn toward the slower track. This is called "skid steering" or "differential steering". While the theory is simple, it’s execution is not.
Incorrect. As noted above, one track drives in reverse. Usually the tank will stop first, then execute the turn on a dime.
However, you really don't see tanks driving on roads all of the time anyway. They are loaded on to flat cars on a railroad and off-loaded in the battle sector. They then deploy primarily off-road, as the road junctions are the best defended areas in any defensive sector. The very purpose of the tank is to go around these well-defended areas and attack where the attacker wants to.
This is because the treads themselves are the weakest link (no pun intended) in the entire tank. The tank is immobilized when one of the treads breaks.
posted by Ironmouth at 7:01 AM on August 4, 2009
In order to steer a tracked vehicle, it is necessary to drive one track faster than the other, causing the vehicle to turn toward the slower track. This is called "skid steering" or "differential steering". While the theory is simple, it’s execution is not.
Incorrect. As noted above, one track drives in reverse. Usually the tank will stop first, then execute the turn on a dime.
However, you really don't see tanks driving on roads all of the time anyway. They are loaded on to flat cars on a railroad and off-loaded in the battle sector. They then deploy primarily off-road, as the road junctions are the best defended areas in any defensive sector. The very purpose of the tank is to go around these well-defended areas and attack where the attacker wants to.
This is because the treads themselves are the weakest link (no pun intended) in the entire tank. The tank is immobilized when one of the treads breaks.
posted by Ironmouth at 7:01 AM on August 4, 2009
Tanks dont need to turn efficiently. They get something like 1mpg and they exist to kill the enemy. The designs are such that theyre better off performing poorly on a variety of surfaces instead of ideally on one or two.
posted by damn dirty ape at 7:32 AM on August 4, 2009
posted by damn dirty ape at 7:32 AM on August 4, 2009
In order to steer a tracked vehicle, it is necessary to drive one track faster than the other, causing the vehicle to turn toward the slower track. This is called "skid steering" or "differential steering". While the theory is simple, it’s execution is not.
Incorrect. As noted above, one track drives in reverse. Usually the tank will stop first, then execute the turn on a dime.
How is that incorrect? That's how tanks traveling forward turn: one track goes faster and the vehicle turns toward the slower track. But both tracks are still rotating the same direction. Your scenario is for a stopped tank turning in place.
OP: I'm a little confused on what you're asking: I'm hoping to hear about how tanks are designed to minimise this problem
The problem is minimized by having an engine powerful enough to overcome the friction of turning in place (the worst case scenario for friction). And, as has been noted, tracked vehicles have terrible fuel economy. My brother and I were talking to a guy that sold old military vehicles and he had one tank there, a Russian tank I think from WWII or the 50s, maybe a T34. He said it would use 1500 gallons of diesel a day to run. It did not look like it had been moved in a long time.
posted by 6550 at 7:42 AM on August 4, 2009
Incorrect. As noted above, one track drives in reverse. Usually the tank will stop first, then execute the turn on a dime.
How is that incorrect? That's how tanks traveling forward turn: one track goes faster and the vehicle turns toward the slower track. But both tracks are still rotating the same direction. Your scenario is for a stopped tank turning in place.
OP: I'm a little confused on what you're asking: I'm hoping to hear about how tanks are designed to minimise this problem
The problem is minimized by having an engine powerful enough to overcome the friction of turning in place (the worst case scenario for friction). And, as has been noted, tracked vehicles have terrible fuel economy. My brother and I were talking to a guy that sold old military vehicles and he had one tank there, a Russian tank I think from WWII or the 50s, maybe a T34. He said it would use 1500 gallons of diesel a day to run. It did not look like it had been moved in a long time.
posted by 6550 at 7:42 AM on August 4, 2009
Best answer: Modern tanks have very elaborate transmission system for applying power at varying rates to the treads to make turns. In the Abrams, it is a fly by wire type system where an abstract input to the wheel is translated to the appropriate turning forces on the treads. It does create significant sideways force on the tracks. This is the reason that the tracks are covered with rubber pads and the reason why they have be laboriously replaced fairly often. A friend of mine was the officer assigned to follow the tank companies around on exercises in Germany and pay people for the damage the tanks did ripping up everything they drove over.
Tanks try to put as much weight on the center of the tread as possible where the sideways force is the lowest (the front and back of the tread moves the most, as you can imagine if you think about the forces involved).
This guy does a pretty good job of explaining the history of the transmissions and the breakdown of the skid forces at the end.
posted by Lame_username at 10:02 AM on August 4, 2009
Tanks try to put as much weight on the center of the tread as possible where the sideways force is the lowest (the front and back of the tread moves the most, as you can imagine if you think about the forces involved).
This guy does a pretty good job of explaining the history of the transmissions and the breakdown of the skid forces at the end.
posted by Lame_username at 10:02 AM on August 4, 2009
Response by poster: OP: I'm a little confused on what you're asking: I'm hoping to hear about how tanks are designed to minimise this problem
Imagine I'm designing a tank. My last tank didn't turn very easily due to the tracks' friction, so I'm considering a larger engine for more power. But before doing that, I'm checking to see if other people have devised clever solutions to my problem that don't call for a larger engine.
Good examples include fearfulsymmetry, who pointed out the importance of having the tank as 'square' as possible, and jon1270, who pointed out that by reducing pressure at the ends of the tank's tracks, friction there will be reduced.
I'm not designing a tank, by the way - but I am advising some people designing a rescue robot, and instead of an engine they use some electric motors.
posted by Mike1024 at 10:46 AM on August 4, 2009
Imagine I'm designing a tank. My last tank didn't turn very easily due to the tracks' friction, so I'm considering a larger engine for more power. But before doing that, I'm checking to see if other people have devised clever solutions to my problem that don't call for a larger engine.
Good examples include fearfulsymmetry, who pointed out the importance of having the tank as 'square' as possible, and jon1270, who pointed out that by reducing pressure at the ends of the tank's tracks, friction there will be reduced.
I'm not designing a tank, by the way - but I am advising some people designing a rescue robot, and instead of an engine they use some electric motors.
posted by Mike1024 at 10:46 AM on August 4, 2009
Best answer: (I was an M1-Abrams tank crewmember in the late 1980's)
Lame_username is on the right "track" (sorry, couldn't resist!). When an Abrams is turning at speed there is tremendous force applied to the treads, and if just one tread segment separates the whole track becomes broken. We called this "throwing track" and it's very bad - it means the tank is immobile and therefore quite vulnerable to attack. So drivers were trained to try and slow down when turning. Most of our training took place on earthy surfaces which provided some buffering should the tracks slide during a turn, but throwing track was still common.
Maintaining the treads was a daily task while on field exercises, tightening each tread segment on both sets of tracks.
posted by shino-boy at 11:04 AM on August 4, 2009
Lame_username is on the right "track" (sorry, couldn't resist!). When an Abrams is turning at speed there is tremendous force applied to the treads, and if just one tread segment separates the whole track becomes broken. We called this "throwing track" and it's very bad - it means the tank is immobile and therefore quite vulnerable to attack. So drivers were trained to try and slow down when turning. Most of our training took place on earthy surfaces which provided some buffering should the tracks slide during a turn, but throwing track was still common.
Maintaining the treads was a daily task while on field exercises, tightening each tread segment on both sets of tracks.
posted by shino-boy at 11:04 AM on August 4, 2009
Best answer: It's not having the tank as "square" as possible, but as rectangular as possible with the distance between the treads being the long side of the rectangle, and as long as as possible, and the length of the tracks as short as possible, thus the tracks deviate less from describing the arc of the turn.
For differential-drive robotics, there is also this approach. It's done here with wheels instead of tracks, but the principle is similar - when you have multiple tracks on each side, permanently angling the tracks slightly results in a big payoff for turning efficiency, at a smaller (but not insignificant) loss for straight-line efficiency.
Some robots motorize this angle, so that all wheels/tracks are straight when traveling straight, and then rotate them to describe the appropriate arc during differential-drive turning.
For a lo-tech improvement, I'd suggest giving the ribs of the tracks a \__/ shape (in cross-section) instead of a ___ shape, so when the tracks are mounted on the wheels, the cross section becomes a \HH/ shape, instead of a _HH_ shape. This means that when the tracks are dragged sideways by the turning, they rise up over the ground and slide sidewides, instead of gouging into it and dragging an accumulating mound of turf with them.
posted by -harlequin- at 11:47 AM on August 4, 2009
For differential-drive robotics, there is also this approach. It's done here with wheels instead of tracks, but the principle is similar - when you have multiple tracks on each side, permanently angling the tracks slightly results in a big payoff for turning efficiency, at a smaller (but not insignificant) loss for straight-line efficiency.
Some robots motorize this angle, so that all wheels/tracks are straight when traveling straight, and then rotate them to describe the appropriate arc during differential-drive turning.
For a lo-tech improvement, I'd suggest giving the ribs of the tracks a \__/ shape (in cross-section) instead of a ___ shape, so when the tracks are mounted on the wheels, the cross section becomes a \HH/ shape, instead of a _HH_ shape. This means that when the tracks are dragged sideways by the turning, they rise up over the ground and slide sidewides, instead of gouging into it and dragging an accumulating mound of turf with them.
posted by -harlequin- at 11:47 AM on August 4, 2009
A robot, eh? Depending on the size / weight of the thing, it brings to mind some cheap kids toys I have seen. Rather than having the wheels or tracks able to turn, a little post is extended, from the centre of gravity, which then rotates the whole vehicle with the tracks above ground.
posted by Meatbomb at 6:34 PM on August 4, 2009
posted by Meatbomb at 6:34 PM on August 4, 2009
This thread is closed to new comments.
posted by Fleebnork at 5:43 AM on August 4, 2009