Portal Axle

[DannytheRadomski]

Some heavy duty off-roaders such as the Hummer H1/Humvee have portal axles to raise their clearance. Basically the drive axles are connected to gears that transfer the torque to the wheels. This allows the axles to be at the top of the hub instead of the middle for better clearance.

My idea is inverting that. What if you had portal gearing hooking up to the bottom of the hub? Then everything could be mounted a few inches lower and lower the CoG. The biggest problem I see is the increase in unsprung weight, but could the pros outweigh the cons?

[Jay Lawrence]

Danny,

I think there are a number of concerns with this, apart from the increased mass.

* Increased transmission losses
* Increased plunge requirements
* Increased manufacturing cost due to accuracy requirements (welding a-arms etc. is much more critical so as to not misalign the gears)
* Increased rotational inertia
* Having to lower the diff substantially (I imagine the rear sprocket would end up hitting the ground)
* Packaging nightmare (have to fit brakes, bearings, wheel speed sensors, etc. plus drive gear)

I don't think you could actually mount anything lower, except the diff and drive shafts, and on cars that I have seen with decent rear ends these parts are already lower than the wheel centre (upswept driveshafts in rear view).

[Markus]

Jay, your concerns have some assumptions behind them so not all of them are necessarely relevant.

In a certain concept portal axles might actually be a worthy option. What comes to unsprung weight, some of the top e-teams already use upright gear reduction and even corner mounted engines, thus the negative effect can't be that big. That means also packaging is possible.

A-arms are usually attached to uprights with balls so I can't see how welding a-arms would effect gear tolerances.

I could see the implemention in something that doesn't use a gearbox or doesn't have the first reduction, so the transmission losses are very similar.

Imagine something like crankshaft -> combined clutch + 2-speed gearbox (planetary gearset like overdrive, with 1-N-2 control) ->
rear diff -> portal reductions. Low CoG, small packacing and quite efficient. Possible? Yes. Complex? A bit. Sensible? Likely not.

Just to lower CoG I don't think portals are worth it. But if the implemented concept is missing first or second reduction or both then it might be a concept to look into.

[Racer-X]

Interesting...

What about a portal axle with a spool/solid rear and a swing arm type suspension while using an engine that isn't from a bike?

Reason I say that is because its the only way I could see any real gains. That way you're lowing pretty much the whole rear suspension. Also I don't think you could get the diff much lower or need the extra reduction on most cars if you stuck with a chain drive bike engine.

[Owen Thomas]

It's a cool concept, and it could probably work if done right; but is it worth it?

Ignoring the technical benefits or drawbacks, from a vehicle and project perspective this is one of those "bright idea" projects that use a lot of resources for little (if any) gain.

Not ignoring the technical details, I would say the only reason to lower the CG of the differential/drive axles is if you're using a REALLY heavy diff. Heavy diffs are usually torque biasing, and a higher localized CG over the rear axle would theoretically increase vehicle performance due to weight transfer. I can't say that outright, since tire load sensitivity and roll distribution and all that other fancy stuff plays a huge factor, but on a vehicle level it doesn't seem to make much sense. </sidebar>

I'm not trying to rag on you, Danny; I admire the way you find potential applications to FSAE just about everywhere. However, there are issues with looking for applications without first needing them.

Too many of us fall into the trap of finding a problem to fit a novel solution. This is backwards, and we all know it, but it happens anyway. If it's not solving a specific problem and just a project someone really wants to do, you have to look at what goal it is helping to accomplish. In this case, if the #1 vehicle goal was to lower overall CG, this is a good call. If it's to improve robustness/reliability (i.e. reduce complexity, more analysis, more testing, etc), then the dropped axle may not be such a great idea.

You can see another drawback from the "cascade effect" where Racer-X begins to justify the specific design of the drive system by changing critical or core systems, like the engine and suspension. The details like drive shafts should never be used to direct decisions of larger systems, it should be the other way around.

Sorry to put a wall of crusading text into a lighthearted technical discussion, but this stuff can't be said enough (especially to new engineers and designers).

[Will M]

Originally posted by Owen Thomas:
In this case, if the #1 vehicle goal was to lower overall CG, this is a good call. If it's to improve robustness/reliability (i.e. reduce complexity, more analysis, more testing, etc), then the dropped axle may not be such a great idea.

Or if you #1 goal is aero then putting the axle at the top of the hub may free more space for the under tray to work.
But then you need to have a fairly wild suspension and there could be compliance issues.

The chance of it working with great success is slim.
But it is a great design exercise, and you may find a new 'killer app' with enough work.

Keep at it!

-William

[Pete Marsh]

Another thread mentions a problem with the utilisation of snowmobile drive trains is the relatively high speed of the drum and the need for an extra reduction.
If you had your heart set on such a thing, then this geared axle idea would let it be turned around and mounted behind the axle, and the low shafts also suit and allow the motor to mounted very low.
Figure out how to "manualise" the CVT, and your good to go with a super compact and out of the way power train. Lots of room left for a turbo or supercharger on your big twin if you want to get serious with a minimum wheel base 60:40 100HP monster!

Pete

[Jay Lawrence]

Markus,

Hadn't really thought of the electrical car setup and it's hub drive (I'm hoping to ignore e-cars until they go away).

With the welding A-arms I was referring to A-arm related accuracy (including pickup points etc.) which I would think would have to be super accurate to allow a remote gear drive to work (I'm just picturing the job a tripod joint does and replacing it with a gear that then drives another in-upright gear). How would you accommodate for plunge and angular change of the driveshaft? I guess a beam would overcome this, but it hadn't crossed my mind earlier.

[DannytheRadomski]

Owen, I'm not in FSAE yet, but right now I see it as a design competition more than anything. As FSAE is now, usually nothing really sets designs really apart other than amount of time and money. There is nothing novel or fundamentally different. WWU and University of Auckland were able to build custom engines and some teams build their own gearboxes, but that is extreme as I can think of designs going.

I am mostly introducing these to entertain myself more than anything, but novel thinking can't hurt FSAE.

[Markus]

Jay,

Portal gears don't exclude using tripod joints. Or any other type of joint (and I don't mean the 'dam here)...