Monday, May 12, 2014

Wheel Alignment

Recently I had to replace the steering rack in the Z3.  This was a completely new experience for me as I've never messed with any car's steering components.  As it turns out, it wasn't that difficult.  As a friend accurately stated, "It's all a mystery until you've done it once, then it's just work."  The plan was that after I got the new rack on, I'd drive it a couple miles to the nearest Costco to get new tires, and then a mile from there to have the front end aligned.

Of course, I needed to try to get the tires as properly aligned as possible for those two short trips between the house and those two places, otherwise I risked scrubbing the remaining rubber off my tires (at best) or losing control and getting into an accident (at worst).  But how?!  I don't have any fancy alignment systems.  Then I came up with a clever plan.  I still have two 5 foot pieces of 1" angle aluminum laying around.  I clamped one to each of the brake disks on the front end, with them sticking out in front of the car.  I pointed the steering wheel straight forward, and then took a measurement of the distance between the brake disks where the angle aluminum was clamped.  I then adjusted the toe-in, toe-out until the measurement between the end of the angle aluminum was the same as it was at the disks.  So how did I do?  I needed to wait for the shop's initial measurement to find out.

I got the new tires on and dropped the car off at the shop.  I told them when they aligned the car, I wanted it set so that the tires had no toe-in.  In other words, I wanted the wheels aligned for 0° toe-in.  After all, this is what I'd had done before, and is generally accepted by EV enthusiasts as a great way of reducing rolling resistance.  Ultimately every degree you have dialed into the toe-in is more energy that gets scrubbed off your forward momentum.  Of course, we've all come to accept the fact that having the car aligned with 0° toe-in means that the car tends to wander a bit when you head down the road.  You get used to the car sort of following the grooves in the road, requiring you to work the steering wheel a little more to keep the car in a straight line.  Not a big hassle by any means, but certainly not what most people are used to when driving a car.

When I made this request, the owner of the shop stopped and said "Why would you want that?"  I explained that I wanted 0° toe-in for efficiency reasons.  He considered what I said and replied "That won't get you what you want."  "Why?" I asked.  He explained that as any road car moves forward, there is a certain amount of force pushing all the tires back.  OK, that makes sense.  He went on to explain that this force will inevitably cause a certain amount of deflection, or flex in the rubber bushings that make up the steering/suspension, causing the tires to lose some of the toe-in that's dialed in when they are aligned.  He stated that if I started at 0°, once the car was moving forward, and the tires were experiencing that road resistance, that pushes them back ever so slightly, I would end up with a slight bit of toe-out.  He continued saying that in such a case, you'd likely find the car wanting to wander a bit as you went down the road.  That got my attention.

Sensing that this could be very important I replied, "Go on."  He said "We normally dial in a total of 16, or 8 for each wheel.  I would suggest that we halve that to 4, or perhaps 3 on each wheel.  That way, when you're going down the road, you're more likely to be at 0° or just a bit less.  That should get you the efficiency you want, and the car will track truer."   I asked him exactly what measurement 16 was.  What was the increment used?  Is it degrees, tenths of a degree, furlongs, what?  It turns out that it was 1/100ths of a degree.  That means that normal toe-in would be set at .08° for each wheel, and he was proposing .03°.   I told him to do it, .03° it is!

The fact that the increment used for aligning the car was 1/100th of a degree seemed to me to make his claims that much more plausible.  The reason I say that is because if the standard increment used for aligning the car had been higher, say like 1/10th of a degree, he would have been asserting that the deflection in the steering geometry was in excess of .6°, which seems a bit extreme to me.  But with the measurement's increment being 1/100th of a degree, the total deflection in the steering geometry is more like .06°.  A more plausible prospect. I have no way of knowing if what he says is true, but it certainly sounded reasonable.

So the ultimate question is was he right?  To know the answer to that, I'd have to be able to see either no loss of efficiency in the car, or even better, a slight gain in efficiency.  Plus I should see the handling improve as the car should track down the road better.

Unfortunately I'm unable to say with any certainty whether the car is more or less efficient in rolling now.  The only measurement I have at my disposal to determine this is the average watt hours/mile.  As most of you know I've been collecting that data with every trip the car makes since it hit the road.  Doing so was what helped me discover that the Helwig Split H60 brushes were more efficient than the standard H-49 brushes that came on the Netgain motors.  But the value, or worth of that number is dependent on all other things in the car remaining the same.  If you change two things at once, there is no way to tell what's responsible for the change in readings.  Well having just put a brand new set of tires on the car, I've completely negated the relevancy of any efficiency numbers I've gathered since those changes.  That said, the car is using a bit more energy per mile.  It had been averaging 320 Watt hours/mile, but it looks like we're up to about 350.  Is it the tires?  Is it the alignment?  Impossible to say.

What I can say is that the car drives great.  It tracks down the road beautifully and is better to drive than ever.  In that regard, the service shop owner seems to have been correct.  I'm calling that a win.

Truth be told, I'm much less interested in the minor increase in efficiency that might be on the table, than in gaining what most people would refer to as "normal" driving car.  Not unlike when I abandoned the nifty little proximity cut off switch I'd designed for the power steering pump.  It worked just fine, and saved a bit of energy, but you had to turn the steering wheel a couple inches before the power steering kicked in, which meant for those first few inches there was no power steering.  It wasn't all that difficult, but it was strange, and I had to explain it to anyone that drove the car.  I don't want a car that drives oddly, and neither does anyone else for that matter!  But for me it's probably for a slightly different reason.  I don't want anyone who allow to drive the car saying "It's great, except for this one strange thing..."  I don't want anyone's first EV experience tarnished by having to work the steering wheel like a saw as the car meanders around the road.

The point is, if you're having your car aligned with 0° toe-in, thinking that it's remaining that way when you're moving down the road, and you're enjoying a lower rolling resistance, you might be wrong.  Your car might be rolling down the road with a slight toe-out.  What would be great is if someone else with a 0° toe-in set on their car could gather a lot of efficiency data, and then have their car re-aligned to .03 or .04° toe-in per wheel and then report back.  Any takers!?

Oh, and incidentally, the alignment I did in the garage was .63°.  Not bad for a guy with a couple of sticks and a measuring tape.


Alex said...

Hi Tim, it might take me a year or so to get to the point where I can test your theory, but I'll try to remember when I get there!
Did you ever install that vacuum pump for the power brakes? I have to shop for actual components now!

Tim Catellier said...

Hi Alex,

Well, that's kind of a funny story.

The original MES/DEA pump has an internal pressure switch that shuts the pump off when it's drawn a sufficient vacuum. I was told by the vendor that if I simply reversed the polarity of the wires to the pump, it would bypass that switch. Doing that would allow me to determine if the pump was still functional or not. So I tried it and no joy, the pump was dead.

Finding the pump was truly dead, I ordered a replacement from EVTV. After it arrived, I measured it realized that it was going to take quite a bit of fabricating to get it in the car. It was still hot out, somewhere around 105°F in the garage, so I thought I'd put it off until cooler weather and just drive with no power brakes (carefully).

A couple months ago when I started pulling the old pump out, I found that the 12V+ line running to the pump had an abrasion on it, and signs of arcing. Apparently this had happened on the line in an area that I can't see or normally get to. I thought "Wait a minute, if this shorted out, it would have likely blown the fuse. Did I ever check the fuse?" The answer is no, I had not. I opened the fuse box and sure enough, the fuse to the pump had blown. I replaced it and the pump fired right up.

Thus ends the tale of the worst troubleshooting story I've ever been a part of. An electrical component in the car failed and not only was checking the fuse not the FIRST think I did, I didn't do it AT ALL!! That's just embarrassing.

So now I have a spare vacuum pump in my garage, sitting in its box. But I figured, that's OK. Eventually I will have to replace that MES/DEA pump. It's a diaphragm type pump, and those diaphragms just don't last forever.