How should I build the driveline?

[mccheynebeeman]

Hello all,

I have a project I am building that has 8 wheels & 8 wheel steering. I have used 8 sprint control arm assemblies (strut towers and brakes and rotors.. etc) along with their CV axles to join up to four seperate differentials. I was origionally thinking that I would have an engine & transmission inline with the vehicle, droping the drive using CV axles to two 90 gear boxes, which will drive two sprockets with 80 size chain to each corresponding diff. The problem I don't like about this idea is that if I need to tension the chain (which I would) I'd need to either move both diffs side to side for each chain, or move the 90 gear box up and down. Since I wanted everything sealed in a custom box, this would make it really hard to seal.

So, I was thinking about using 3 carrier diff gears (same ones as on the current diffs) inline instead of chain to connect everything. The middle gear would connect to the 90. This way, everything can be secured down without needing future adjustments.

Anyway thoughts?

(to see some pictures of this machine for a reference of what I'm talking about got to

http://www.4wheeler.ca/index.php?showtopic=20755)

Thanks in advance,

Frazer RM Ross

 

[mloew]

Could always do this:

&

[ignore]
[img]http://www.army-technology.com/projects/mrav/images/mrav3.jpg[/img]
& 
[img]http://www.mil-plus.com/public/photo/00000087.jpg[/img]
[/ignore]

Best regards,

Matthew Ian Loew


Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[mccheynebeeman]

I made a post in the "Gear and Pulley" mechanical section on using crossed helical gears and it was stated that they are not suitable for high torque applications. (Interestingly enough, I've found gear suppliers stating that they are, and some suppliers stating that they are not). I do like the elegant solution that they offer, although my limited garage tools might hinder my ability to make 4 properly working helical gear boxes.

ca2sbl - when I had initially thought of building this project, that's exactly how I envisioned the drive train to work, but only knew of semi-trucks that had the dual driveline diffs. I'll look into them more (but I imagine they will be quite pricey). Another issue is mating the current CV cups (from the GMC sprints) into this gear box.

I’m not really happy by going with chains, but it looks like it is the only solution without spending a lot of money on the project (these ideas are great though for future projects). If I did go with chains, could someone quickly scan through the calculations below to make sure I did them correctly.


As for calculations, the engine produces 58 lb/ft @ 3600 rpm. First gear reduction is 2.4:1 and the differential inside the automatic transmission is 3.16:1. That would give me a maximum torque of 58 lb/ft * (2.4 * 3.16) = 440 lb/ft at the output shaft. This shaft would be inline with the vehicle and change 90 degrees in direction via the bevel gear box, which needs to be rated at this maximum torque (440lb/ft or 5280 lb/in)..

I've found a gear box (

http://catalogs.hubcityinc.com/bevel/model800.html)

that is rated at 443 lb/ft @ 400 RPM.

Now given that I want a 1.45:1 reduction on the main drive axle, this would be putting 444 lb/ft * 1.45 = 643.8 lb/ft on some of the chains. The Main drive sprockets would be 22 teeth and the sprockets on the diffs would be 32. Using this ANSI Chain spec (

http://www.4wheeler.ca/index.php?act=Attach&type=post&id=24693)

I figured 80H Ansi size chain will work for the 22 tooth sprocket on the main shaft (can handle 446lb/ft torque) and 80H on the diffs with the 32 tooth sprockets (can handle 650 lb/ft).

Thanks for all the help! The pictures are great!

 

[mccheynebeeman]

Warpspeed, sorry I missed your line "I still feel that crossed helical gears may be a suitable solution for a very light weight low powered vehicle." The vehicle is going to weigh around 1200-1500lbs and have the engine mentioned in the calculations above (54HP) (That might put things into perspective on it being light or heavy). I do not know what crossed helical gears can handle since I still have not been able to find any specs.

 

[ca2sbl]

mccheynebeeman:
The M151 series of 1/4 ton trucks (jeep) were used by the U.S. military from 1960 until 1985. The 71 Hp gas engine put out 128 lbft of torque at 1800RPM. In first gear (5.712:1) the torque would be 731.136 lbft. The differential gear ratio is 4.86:1

The front suspension is "A" arms, coil springs, tube shocks, open steel cross u-joints (Cardan type) with drum brakes. The whole front suspension was mounted on a subframe that was attached to the unit body with four bolts. The outside width is 64 inches.

All of these parts are available new or used on the military surplus market. I have seen good, used differentials for $95U.S. and new units for $300U.S. There are civilian aftermarket automatic and manual locking conversions available.

These subframes can be bolted to box section beams, with additional cross beams to make up the chassis of your machine. To connect the output from your transmission to the M151 differential, you would need M151 driveshafts cut and welded to the Sprint drive axels.

There are several web sites for military vehicle collectors that have links to military parts dealers that
can supply you with all of these parts, including complete assemblies.

Keep having fun,
Russ

 

[Warpspeed]

I really don't know either. But tooth loading is I suppose, relative to gear diameter. As the driven gear has to enclose the diff hemisphere, and the ratio is 2:1 both drive and driven gears are going to unavoidably be of a fairly large diameter.

The tooth loadings will be quite low, and even with point contact, I cannot see actual gear failure ever being an issue. For it's physical size it will be very lightly loaded.

The serious hard core power transmission engineers might look at this as being only capable of a quite low power density, and not being a particularly elegant or efficient design. But does that really matter ?

 

[mccheynebeeman]

ca2sbl:

It would make complete sense to not re-invent the wheel and go with these diff boxes. But weight is a bit of an issue (granted a gear box capable of 731lb/ft matched to a similar angle gear reducer box is around 80-90lbs). I'll try to find some more info on these.

warpspeed:

"The serious hard core power transmission engineers might look at this as being only capable of a quite low power density, and not being a particularly elegant or efficient design. But does that really matter ?" I have always fought against what is supposed to be because it is the accepted way, compared to what I want it to be. In that sense I would want something quite inexpensive, fairly easy for me to install (no major machining) and light (so, no.. truly it does not matter). I really do like the crossed gear set because of a possible single axle connecting all three diffs on one side and then another axle to drive the remaining diff (so the transmission can sit low in the vehicle). One problem (one I will face with a chain driven setup) is that I will need to create a custom housing for each gear set.

One thing I noticed you mentioned above about the steering is leaving the rear pair of tires fixed while the other six tires could turn and not use rear steering (ie. monster trucks). I planned on steering all 8, with the rear four steering opposite to that of the front 4. Granted of course, that this could be dangerous at high speeds, I was thinking of using a set of handlebars instead of a steering wheel. With proper gearing this would make it awkward to over steer as the user at the handle bars would comfortably only be able to articulate the system to a limited degree (for safe driving at high speeds). The benefit to all wheels steering would be that when needed, those handlebar could be completely rotated and thereby steer the vehicle at a much sharper turning radius than with only front wheel steering.

Here is a picture of my proposed setup for the steering:

http://199.126.216.230:81/pictures/random/8wheeled_steering.jpg

 

[Warpspeed]

I tend to agree with your line of thinking here. Four commercial military diffs, while mechanically attractive and very robust, will quite likely end up relatively heavy. And the 4.86 ratio is a bit of a worry too. One way around that might be to consider six larger diameter wheels instead eight smaller wheels? That would simplify a lot of other things, and might also reduce vehicle weight and complexity.

Road cars that use four wheel steering, steer the rear wheels in the opposite direction to the fronts for maneuverability at parking speed only. It would be highly unstable like that at highway speeds. These systems steer the rears (very slightly) in the same direction as the fronts at high speed, which is probably not at all what you want.

That eight wheeled military vehicle above, and twin steer trucks have the right idea. All the steering should be done with the front sets of wheels, and the rear most axle should not steer at all for best stability. The only problem I can see with that, is getting enough wheel clearance to gain sufficient steering lock at the front most axle.

I suppose the whole steering system is going to be a compromise between good low speed turning circle, and high speed stability. But if it also has the option of skid steering as well, it could turn in it's own length to negotiate obstacles at very low speed, and use the normal steering at higher speeds.

Air bag suspension with adjustable ride height would be another interesting toy to play with.

 

[GregLocock]

You need to steer the rear axle if you want the tightest possible turning circle, including overhangs.

There was a WW2 UK armoured car that had 4 wheel steer, they ended up disabling the rear wheel steer as they had too many rollovers. I do not think that limiting the steer angle is sufficient, at high speed, I think you need to actively fix the rear half of the vehicle in the straight ahead configuration. Managing the transition will be complex.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[mloew]

The chain drive on the steering may have an unacceptable amount of lash in it. I recommend using a direct mechanical linkage or gears rather than chains.

Best regards,

Matthew Ian Loew


Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Warpspeed]

If I was going to steer the front and rear pairs of wheels only, one way would be to mount a conventional steering rack down the length of the vehicle and use some bell cranks at each end.

Nothing really wrong with handlebars either. Power steering might be worth a thought on a ~1200 Lb vehicle with handlebars and a fairly direct steering ratio. A power steering rack would be a simple solution to that.

But I still believe rear wheel steering is going to require some fresh dry underwear after the first really high speed bend.

 

[mccheynebeeman]

ca2sbl: I found the m151 differentials used for 95USD. The weight appears to be 70 pounds from what I could gather on my investigation. Granted if I had known about the m151 I would have started to build my project around these, but now heavy modifications would need to be implemented if I were to use these now.

mloew: Thank you for the advice. I was considering this as a possible issue, but now that you have stated the same I will try gears in an effort to keep the steering tight.

GregLocock: If the scenario of a driver in a vehicle turned the steering wheel 5 degrees to the left and the car was turning at X radius, was matched to a front and rear steering vehicle that when the driver turned the steering 5 degrees to the left the vehicle turned at the same X radius, would this make the vehicle anymore unstable at high speeds? Are there other factors to consider then just a sharper turning radius that might make it flip over easier givin the above scenario? hmm

Warpspeed: Power steering is a definite, as well as extra dry underwear. The skid steering (via braking one side or the other) was only going to be used where traction was an issue, since there are not going to be locking diffs I could force one side or the other to always have drive. Air bags will be thought about in the next project that will also use the m151 diffs.

 

[Warpspeed]

It is reasonably easy to estimate turn radius and turn centre by drawing a line through the rotational axis of every wheel. The lines will meet (or cluster) around the turn centre. Steering both ends would approximately halve the turn radius for any given steered wheel angle. It is a bit rough and ready, but this may help visualise what is going on.

I have been thinking a bit more about the skid steering and the brakes. How about having three brake pedals ? One for the brakes on each side, and a central master pedal that operates both sides equally via a balance bar. That should feel fairly "natural" to drive.

 

[mccheynebeeman]

Warpspeed: steering both ends would reduce the radius by half, but what if the gearing from steering wheel were to be 2:1 (ontop of whatever ratio there is now). Then the steering would be no different than that of a front steered vehicle without the 2:1 reduction (as seen by the driver), except for the fact that you can turn the steering wheel (on the all wheel steering) twice as many revolutions to get the tightest turn. If it were setup this way, I can not see it being anymore unsafe than our regular highway vehicles.

As for the skid steer effect, I'm glad that we see eye to eye in that I was going to use that descrbied setup. (If two pedals were only used, high speed braking could potentially get really dangerous)

Note: Estimated top speed is around 80kms/h, average speeds would be around 30-50kms/h

 

[Warpspeed]

The danger of rear wheel steering is that it will definitely cause vehicle oversteer as speed increases. Oversteer is dangerous because as the rear wheels begin to lose traction the turning radius reduces even further.

Basically the rear end slides outwards and the vehicle rapidly goes into a spin. All production cars have some built in rear understeer, even rear beam axle cars steer the rear wheels. The rear wheels are automatically steered by body roll.

Roll understeer is stable and safe. By deliberately steering the rear wheels in the oversteering direction, high speed stability will be marginal to non existent. The vehicle is liable to very sudden and violent loss of control at higher speed. It will either go into a sudden uncontrollable spin, or tip over.

The only road vehicles that I know of that use rear wheel steering are some really large fire engines, where a second driver at the back steers the rear wheels to get around very tight intersections. The other application I have recently seen, was a very large mobile crane that had ten huge wheels, five down each side. The rear pair could be steered independently (by hydraulics), and did not operate directly from the steering wheel. Presumably this extra steering is used for very low speed maneuvering of the crane into position in tight situations.

You would have loved this crane, ten wheel drive ! The front four wheels were steered, then four fixed wheels, and the rear pair were locked straight ahead for road operation.

 

[GregLocock]

No, at high speed there are many subtle effects to do with rear wheel steering. What you are proposing is an oversteering machine, it will have many awful steering characteristics, including a speed at which the controls will reverse, ie above that speed you will have to steer left to turn right.

Probably the best discussion of this is Dixon "Tires, Suspension and Handling" from

http://www.sae.org,

but basically you have to sit down and work it out yourself, there is not a whole lot written down that is not proprietary.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[mccheynebeeman]

The danger of rear wheel steering is that it will definitely cause vehicle oversteer as speed increases. Oversteer is dangerous because as the rear wheels begin to lose traction the turning radius reduces even further.

What you are proposing is an oversteering machine, it will have many awful steering characteristics, including a speed at which the controls will reverse, ie above that speed you will have to steer left to turn right

hmm.. I guess I am still having a hard time fully visulizing this issue, but a solution could be that I build it with the oversteer configuration and if the vehcile is truely too unstable at proposed speeds then I can always remove the rear steering and weld up the linkages to remain fixed.

By the way, what figures of high speeds are we talking about?

 

[Warpspeed]

High speed is any speed where centrifugal force starts to become significant.

To try to visualise this, imagine you had a four wheeled vehicle where the front wheels were steered, and the rear wheels were castors. You know, those little wheels fitted to furniture that swivel around and steer all by themselves.

The vehicle would have absolutely no directional stability. As soon as you try to steer the front, the back just goes wherever it wants to go. That is what Greg means by steering left to go right, and my smart comment about a change of underwear.

For directional stability, the rear wheels ABSOLUTELY MUST resist any sideways movement. What you are proposing is to steer the rear wheels in the same direction that castors would naturally try to steer, giving a similar unstable response.

Try riding an office chair down a ramp to get the feeling of having absolutely no directional stability, or control.

 

[GregLocock]

The calculations for the critical speed at which control reversal will occur are too complex to summarize here, but are available in every vehicle dynamics textbook that is written for student use. The problem is that you won't know many of the numbers you'll need in those equations I think.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Warpspeed]

The concept of oversteer instability is difficult to express in plain easy to understand text.

But the idea is that the vehicle will have a strong tendency to try to slide sideways, or yaw uncontrollably at speed. It will attempt to spin out, or suddenly and very quickly swap ends.

The actual point of "letting go" may be calculable, but in practice a bump, pot hole, wet patch, or sudden change in road surface, will cause a very sudden and dramatic loss of control.

As the whole machine starts to yaw, the front wheels, being attached relative to the chassis, will change direction with respect to the intended direction of travel.

This unfortunately makes things worse, which is what makes it such an unstable condition. To correct, the front wheels need to be very quickly turned in the opposite direction to the direction of the corner you are negotiating.

But because it is unstable, insufficient steering correction will not save you, too much, and it will start to spin in the opposite direction. The danger is that a sideways sliding vehicle can always hit a solid obstacle, and/or flip over.

Rear wheel steering almost guarantees these sorts of instabilities and handling problems.

 

[mccheynebeeman]

Well, I can somewhat now see what you two are trying to explain. I have been reluctant not to change the opposed steering concept since I did not think there would be too much of a difference, but at this critical speed I can now understand the instability of the setup. I really wanted rear steering to not only have another "eye popping" visual but also greater maneuverability for the machine. But, since the terrain is not suitable for an unstable vehicle (climbing a hill side or riding through an open field with gofer holes) and I'm not about to risk a limb for looks, I'll lock down the rear 4 tires. Like you said warpspeed, I can always use the skid steering for tight maneuvers.

Thank you all for the help!