DIY Electric Car Forums banner
1 - 20 of 30 Posts

·
Registered
Joined
·
36 Posts
Discussion Starter · #1 · (Edited)
Sometimes you just have to start a dream project. I've tried in the past but really didn't know what I was doing and useing those times as a learning instance. So I finally bought a bunch of DOM tubing, parts and starting building and designing. after I felt I was ready I starting filming and building.

Basic info as of right now is:
-DOM frame
-Lexus gs450h electric motor
-differential steering
- seating for 4
-solar charging for 12v system and maybe HV

More to come as I hash things out and start building


 

·
Registered
Joined
·
6,211 Posts
In an electrically driven differential-steered vehicle, why would you use a single motor and some mechanical steering arrangement instead of just using separate motors for each track?
 

·
Registered
Joined
·
36 Posts
Discussion Starter · #3 ·
In an electrically driven differential-steered vehicle, why would you use a single motor and some mechanical steering arrangement instead of just using separate motors for each track?
Comes down to funds. Not only would you need another motor, gearing, battery and even more battery. Im simply doing the best I can with what I have. I would love to use two but I simply dont have the funds for it
 

·
Registered
Joined
·
6,211 Posts
Comes down to funds. Not only would you need another motor, gearing, battery and even more battery.
Also it much more simple
It would not take any more battery, since the same total power would be required. Yes, it would need two motors and two gearboxes, but unless you are building an exceptionally high-speed vehicle there is no need for the mechanical complexity of the GS450h transmission (or its dual internal motors and planetary power-splitter gearing) - simple fixed-ratio gear reducers (from almost any EV) would work. The two-motor approach would eliminate the mechanical complexity of whatever is being planned for differential steering. The biggest expense would be for two controllers and whatever would be used to coordinate them.

The simplest possible differentially steered electric powertrain is two motors, each with a fixed-ratio gearbox. It's how every child's toy and robotics hobbyist basic chassis that can turn on the spot works.

Im simply doing the best I can with what I have. I would love to use two but I simply dont have the funds for it
I'm not sure I understand the logic of buying an inappropriate drive unit, then saying that choice prevents building a better design of vehicle, but I do understand making do with the stuff you have (even when it was selected for a very different vehicle). Enjoy the project! :)
 

·
Registered
1996 Toyota Land Cruiser
Joined
·
143 Posts
Steering will be done by two disc brakes and a lever for each hand. He can probably build it for less than $100 from parts out of a junkyard. I approve of this build! KISS!
 

·
Super Moderator
Joined
·
1,343 Posts
I think the real answer here is "I'm doing what I want to because I want to", in which case, general advice is not necessary or useful, only specific troubleshooting.
 

·
Registered
Joined
·
36 Posts
Discussion Starter · #8 ·
I think the real answer here is "I'm doing what I want to because I want to", in which case, general advice is not necessary or useful, only specific troubleshooting.
yes and what I can do to achieve what I want to do - I love advice and would love to build this with someone but I have a tight budget and can only do what my skill capacity is. Thats why I love DIY forums like this because most these builds people are building what's in there head and it's just so cool! Plus I have a film degree and I'm finally putting it to use editing these video :LOL:
 

·
Registered
Joined
·
36 Posts
Discussion Starter · #9 ·
Steering will be done by two disc brakes and a lever for each hand. He can probably build it for less than $100 from parts out of a junkyard. I approve of this build! KISS!

It will actually have three disc brakes, one before the 4.1 LSD and one on each side. The sides will provide the Differitial steering provided by a steering wheel and thus turning as pressure is applied to the opposite side. I could go with clutches and torque vectoring but again thats not this project. The brake before the lsd will be the actual brakes and this is a simple design based off of rock crawlers and monster trucks , needing the least about of force to apply the great results. KISS if def what this is, and glad to have you along!
 

·
Registered
Joined
·
6,211 Posts
It will actually have three disc brakes, one before the 4.1 LSD and one on each side. The sides will provide the Differitial steering provided by a steering wheel and thus turning as pressure is applied to the opposite side.
...
The brake before the lsd will be the actual brakes and this is a simple design based off of rock crawlers and monster trucks...
That works, and for the steering portion is the system used by basic 6-wheel and 8-wheel ATV's (such as the Argo), but it's very crude.

If levers are actually used, there should be a system to keep both brakes from being applied at the same time when trying to just steer, because that would just waste power if used together. A wheel can be used instead of two levers (and tanks do routinely use a steering wheel); off-roaders routinely use a single lever rocking between two hydraulic cylinders (but they normally make the lever move forward and backward, which would be weird for steering). There would presumably also be a brake pedal for slowing down, which would work with either the separate driveshaft brake or with the same brakes as used for steering. This is a typical single-lever cutting brake control, from Wilwood:
122761

Of course a version home-built from two clutch master cylinders and a random bar would be much cheaper, but the operating principle is the same. The pushrod for each cylinder goes slack when the other cylinder is pushed - it can't be arranged so that the lever pulls on the unused cylinder.

Braking even one track when not trying to slow down is a waste of energy, which is always an issue in an EV. A power-splitter gearset could be used on each output to reduce that, but that would be more complication again. The loss could be reduced by applying the retarding torque with generators and recycling the power to the drive motor, but that would mean creating two custom transmission systems, ironically each of the type which is sitting - unused - in that Toyota hybrid transmission.

But complications and efficiency aside, the worst part is that the brake-to-steer method doesn't even work well. Rock crawlers and monster trucks (and the tractors, etc. that they copied this system from) use differential braking (which they commonly call "cutting brakes") only to help in tight turns - they primarily use normally steered wheels. You might think that holding both brakes off (both levers forward or steering wheel straight) results in going straight ahead, but it doesn't; it just means completely uncontrolled direction. The differential ensures the each track gets approximately the same torque, but those tracks can run at any speed. It's like driving a car with rear wheel drive, an open diff, and no one holding on to the steering wheel: the vehicle wanders at the whim of any factors affecting or side forces, such as slopes, changes in the ground softness or material, or whatever. As a result, the driver has to continually tweak the steering brakes... almost like steering that car with two loose ropes connected to the steering wheel, which can't be used to hold the wheel still in any position because only one can be pulled at a time.

There are tracked vehicles which do use braking to turn and which go straight well, but in those (such as traditional bulldozers) there is no differential and pulling a lever (or pushing a pedal) opens a clutch for that side then brakes - that's more mechanical complication (although a differential isn't needed) and to provide a range of degrees of turning radius (or even the ability to rotate in place) a more complicated system of gearing is sometimes used.

Differential steering transmissions are a fascinating mechanical challenge, but in an EV they're all pointless, since the best solution is no mechanical connection between the sides at all. I do realize that this is about something that can be easily built at minimal cost, rather than something that works well, but I think expectations should be kept at a realistic level... and I hope it will never be expected to travel at significant speed.
 

·
Registered
Joined
·
36 Posts
Discussion Starter · #12 ·
@Rusted B&B stated that he is going to use a LSD, or limited-slip differential. That will eliminate most of the issues with direction control that you are talking about @brian_ and send equal torque to both sets of tracks.
I have no idea why @brian_ is attacking my build so much, or me for that matter. but this has been done many times and I just wanted to share my ideas and you nailed it on the head. LSD eliminates these issue and is very reliable

Like these guys
 

·
Registered
Joined
·
6,211 Posts
@Rusted B&B stated that he is going to use a LSD, or limited-slip differential. That will eliminate most of the issues with direction control that you are talking about @brian_ and send equal torque to both sets of tracks.
Good catch - I had skipped past the limited slip diff, which is exactly the opposite of what will work well. The nature of the problem will depend on the type of LSD:
  • with a worm gear type (Torsen, Quaife) LSD, when the driver attempts to steer by braking one track so it turns more slowly, the LSD will respond by fighting that, applying more torque to the slower-turning track
  • with a clutch-pack LSD that squeezes the clutch packs in response to drive torque, the LSD will make the vehicle harder to turn when more power is applied; yes, in a straight line full power would be most stable (but still not really straight), but it will become unstable as soon as power is reduced
  • a Detroit Locker is more extreme: it will completely prevent any steering above some level of applied torque, and revert to simply open below that level
In fact, an open diff is the only type which will actually send equal torque to both tracks. The entire purpose of any limited-slip diff is to apply unequal torque in an attempt to get more equal speed, reducing the tendency of one wheel to spin (otherwise known as "turning").

If any LSD is used, the worm gear type is likely the most desirable, because it reacts to output speeds speeds (which is what you want to control) rather than drive torque (which is unrelated to the need for directional stability).

The Russian project in the YouTube video uses an open diff, according to the captions in the linked episode; they also say "That's not quite correct, but we will try to use on drive tests", and in that episode nothing is operating. There are a lot of videos of it running, but I don't know if any explain what was wrong with the original plan. In the "first run" video it does go down the nice smooth road at up to 50 km/h, wandering all over it while the driver saws away at the steering wheel... with an unknown type of differential. This vehicle apparently does get driven a lot, but judging from these two videos there is likely no useful technical information ever shared.
 

·
Registered
1996 Toyota Land Cruiser
Joined
·
143 Posts
Good catch - I had skipped past the limited slip diff, which is exactly the opposite of what will work well. The nature of the problem will depend on the type of LSD:
  • with a gear-type (Torsen, Quaife) LSD, when the driver attempts to steer by braking one track so it turns more slowly, the LSD will respond by fighting that, applying more torque to the slower-turning track
This is the type of LSD he is using, my guess, but your interpretation is wrong. Differentials are designed to differentiate, first. Torsen is short for Torque-Sensing and the geometry of the differential gears allows to diff to differentiate when there is a great difference in speed between two sides of the diff, like for instance a racecar coming out of a tight hairpin bend or a tank skid-steering. But when there is a smaller difference in speed between both sides, then it starts to lock up, like for instance a hard acceleration run where both rear tires are struggling for traction. If it worked like you say then out of the hairpin corner the racecar would have the inside tire spin up in a smokeshow, which is actually what happens in an open differential.
 

·
Registered
Joined
·
6,211 Posts
This is the type of LSD he is using, my guess, but your interpretation is wrong. Differentials are designed to differentiate, first. Torsen is short for Torque-Sensing and the geometry of the differential gears allows to diff to differentiate when there is a great difference in speed between two sides of the diff, like for instance a racecar coming out of a tight hairpin bend or a tank skid-steering. But when there is a smaller difference in speed between both sides, then it starts to lock up, like for instance a hard acceleration run where both rear tires are struggling for traction.
The marketing name is cute, but a Torsen doesn't sense torque at all. Quaife probably sells more of these things in the aftermarket than JTEKT (the current owner of the Torsen name and design), and Quaife calls it their Automatic Torque Biasing (ATB) limited slip differential - torque bias is the output, torque isn't the input. By connecting the output sides through a nearly one-way gear drive (that's why the worm gears in some versions), it only allows the slower-turning output to receive more torque; it is effectively speed-sensing, rather than torque-sensing. Torsen documentation refers to the Torque Bias Ratio, and that's the result of the gear action, not the input to it. The side which turns faster transfers torque to the slower-turning side, as in any reasonable LSD.

If it worked like you say then out of the hairpin corner the racecar would have the inside tire spin up in a smokeshow, which is actually what happens in an open differential.
The scenario of the inside tire spinning wildly doesn't occur with a Torsen because as soon as it is the faster-turning tire it gets less torque. An open differential puts the same torque to both outputs, so when one has less load and thus less traction - typically the inside tire in a turn but also the tire on a patch with less traction or the tire with less load on it or even lifting due to uneven ground - it continues to get the same torque as the other side and slips.

But as I said, a LSD will make the vehicle harder to turn, which is indeed directional stability... just stability that the driver has to fight.
 

·
Registered
1996 Toyota Land Cruiser
Joined
·
143 Posts
Well, yes, torque will increase on the braked side but speed will decrease and as it comes to a stop the diff will fully open like an open diff. Luckily with a disc brake there will be more than enough mechanical advantage to overcome the torque. Maybe you've just started thinking about a vehicle like this today, but a recreational tank like this not a new concept and is a common build for gearheads all over the world using this tried and tested formula.
 

·
Registered
Joined
·
36 Posts
Discussion Starter · #19 ·
Well, yes, torque will increase on the braked side but speed will decrease and as it comes to a stop the diff will fully open like an open diff. Luckily with a disc brake there will be more than enough mechanical advantage to overcome the torque. Maybe you've just started thinking about a vehicle like this today, but a recreational tank like this not a new concept and is a common build for gearheads all over the world using this tried and tested formula.
been done for decades and it's going to be a fun concept build - solar charging, carbonfiber, EV, moving turret - and all around foolery
 
1 - 20 of 30 Posts
Top