it won't cost fuel economy besides whatever weight it adds... imho the benefit far outweighs the weight...

 

I have a couple of contributions to make – firstly about the technicalities of Physics, measurement and units, and then about the day-to-day practicalities of owning and driving a modern car like the Jazz/Fit. Here's the first:

Mass and Weight: – There are very precise terms when used in science and engineering. Precision is needed to avoid the sort of misunderstanding that causes quite normal people to call each other nasty names - or at worst results in embarrassingly crashing an expensive spacecraft into Mars instead of going into orbit around it.

Everything has Mass. (Ok, except for certain subatomic particles like photons or neutrinos, but we’re talking about things you can see, touch, and drive along a road.) That mass doesn’t change and it’s measured in kilograms except in that old part of the British Empire, the United States, where they quaintly still use the old British pound. A mass of a kilo is a kilo wherever it is – sitting on the ground, driving in a car, floating around inside the Space Station, or on Mars. And 1 kg is near enough 2.2 lb.

Gravity is a distortion of space-time caused by mass. (If anybody tells you they know how and why it works, they are lying. However the math is more straightforward.) This distortion causes different masses to try to move towards each other. An unrestrained object close to the Earth’s surface will accelerate straight down at 9.81 metres/second/second, or if it’s in America, 32.2 feet/second/second.

But most objects are not free to fall. If something is sitting on the ground, it will push on the ground; and because the ground isn’t going anywhere, it pushes back up with the same force. The forces are equal and opposite, so nothing moves. This force is the Weight of the object. If everything is at rest, it’s a constant. Force is measured by Scientists in a unit called Newtons.

You are likely to weigh around 1,000 Newtons, or hopefully a little less.

The Newton is one of those metric units that doesn’t mean much to most people. They do know what a kilo or a pound is, so they tend to refer to the Weight of something in these terms. What they really mean is its Mass, but if nothing is moving it doesn’t matter. Once things start moving, with dynamic forces and accelerations involved, the difference becomes important.

The Weight of a kilo or pound, i.e. the force exerted between it and Earth by gravity, is 1 kilogram-force (kg-f), or 1 pound-force (lb-f), respectively. The pound-force is a recognised engineering unit, the kg-f less so, and they are all convertible one to another. To get from kg-f to Newtons you multiply by the acceleration of gravity. 1 kg-f is 9.81 N (and it’s also 2.2 lb-f.) So if you “weigh” 220 lb, what that really means is that you have a mass of 220lb, or 100 kg, and at rest you push against the ground with 220 lb-f, 100 kg-f, or 981 N.

But if you happen to be on one of those free-fall rides, your mass hasn’t changed but you will temporarily weigh nothing. And nor will what you just had for lunch…

So there we have it. Mass is a property of an object, one that doesn’t change, and is measured in kg, or lb.

But Weight is a force, and can’t be measured in kg, or lb. The correct units for a force are kg-f, lb-f, or Newtons. It’s just that we tend to take short-cuts in the language…

 

People on this forum are talking about changing springs, changing ride height, and adding anti-roll bars, and it’s pretty clear that most of them are guessing at what they are doing. Many of the settings they are trying are effectively resulting in no suspension at all, and there are very few situations where this might be a good idea.

There are many things you are trying to achieve with your suspension. The chief one is to keep you tyres in contact with the ground as much as possible. Roads have bumps. Some have big bumps. Some corners have a hidden bump that can throw a badly set-up car right off the road. And, if you want to enjoy driving your car, the last thing you need is to be frightened about what it might do to you.

I’ve just been for a few quick laps around our local racetrack. It’s called Pukekohe and it has a variety of corners – with good tyres and a good setup, some are 200 km/h, (120mph) while the slowest hairpin is about 80 km/h (50 mph). But any of you who has seen the Australian V8 touring cars racing here will know it has a couple of places with nasty bumps. Some teams seem to get their cars to ignore the bumps. To others it’s a problem. The EK9 Civic I was in last week has a very nicely tuned K20A and some very effective rally suspension with plenty of travel. It didn’t feel the bumps at all – and we were able to take some very cheeky shortcuts over the kerbs without upsetting its composure. So even on a track, having proper suspension is a good idea.

But I’m seeing posts from people trying to drive suspensionless cars on ordinary roads. That’s just crazy. More of that later, I hope…

There seems to be a lot of doubt about how or why a rear anti-roll bar works. But before discussing that, let’s start with tyres, friction and that elusive property μ, or coefficient of friction.

Many people have heard of it. It’s a number which relates the maximum horizontal force a tyre can exert on the road to the weight (remember, weight is a vertical force). For a road tyre on a dry road you might be looking at close to 1. That means that the tyre will allow you to stop or turn with an accelerating force close to the weight of the car. You can stop or turn (and if you have 4wd or all the weight of the car is on the driving wheels, you could even accelerate at close to the acceleration of gravity, close to 1 g.) If the road is wet or otherwise slippery, this number will be quite a lot less than 1. Good racing tyres at full working temperature go much more than 1.

But it’s not quite that simple. In a straight line – pure acceleration or braking – it is true that the force generated by a tyre is quite closely proportional to the weight it carries over quite a weight range. For instance, if it was carrying a weight of 900 lb-f (about 4.0 kN) it might have a maximum braking force of 810 lb-f (3.6 kN). Put half the weight on it and you get similar proportions – 450 lb-f (2.0 kN) gives 405 lb-f (1.8 kN) maximum braking. Double the load and you’ll pretty much double its braking ability. You have to overload a tyre by a long way before this linear relationship starts to break down. At several times its design load, the tyre will start to distort badly enough that the proportion, the coefficient of friction, starts to drop off.

But it’s not quite like that with sideways loading. At 900 lb-f, you might get 800 lb-f of cornering force. μ is .89. At 450, that might give you 410 lb-f cornering. μ is .91. But at 1800 lb-f of weight, you get some quite serious sideways tyre distortion while cornering and although the extra force pushing the tyre into the road does give you more available sideways force, it’s not in proportion, and you only get a maximum of 1450 lb-f of horizontal force. μ has dropped to .81.

And this is why we can use the suspension to tune the cornering balance of the car.

(Next time I can get a few spare minutes I’ll continue…)

 

A typical FWD car might have around 2/3 of its weight on the front wheels, 1/3 on the rear. (Again, remembering that Weight is the force pushing the tyres on to the road.)

Combine this with the fact that tyres get progressively less efficient at generating sideways forces as the weight on them increases and you can see why the front tyres are the first to lose grip as you try to increase the cornering speed.

Let’s put some numbers to this, and let’s start by imagining the car turning a long left hand bend. All four tyres are the same, which is what we are stuck with in most cars, for reasons of practicality. For a car like the Fit, the numbers might come out like this:

Weight on the wheels:

LF 550 lb-f, RF 1050 lb-f,
LR 250 lb-f, RR 550 lb-f.

With that loading on that right front tyre it is going to be limited to a maximum of about .84 g turning acceleration.

But the much more lightly loaded right rear won’t be anywhere near its limit, which could be over .9 g. The front of the car will lose grip first. It will understeer, wash out, push, whatever you like to call it.

So now let’s stiffen the rear suspension, a little with springs but mostly with a proper, powerful rear anti-roll bar.

Now the weight on the wheels while turning hard to the left is more like this:

LF 630 lb-f, RF 970 lb-f,
LR 170 lb-f , RR 630 lb-f.

See how we’ve taken quite a bit of weight off the right front, and added it to the right rear. It’s absolutely true that by loading up that right rear tyre we’ve taken some of its potential grip away. But the right front tyre is still much more heavily loaded, so the rear still can’t get to it’s limit, which would still be about .89 g. The big thing is that we have increased the cornering power of the front wheels. We are now looking at a maximum turning acceleration of about .86 g at the front, a useful 2.4% increase in turning force.

Another effect of the change is that there is less steering input needed to initiate and maintain the turn. The slip angles at front and rear have become more similar. So corrections are easier – even if the unexpected happens and the rear of the car steps out, you haven’t got half a turn of steering lock to unwind before you can start to correct the slide. Also you will note that the left front tyre also has more weight on it – so more power can be put through it, allowing much better and earlier acceleration out of the turn.

And that is the power of a rear sway bar.

You could get similar theoretical results by using much stiffer rear springs, but ride quality suffers. You will reduce the usable suspension travel in the light rear of the car. It will start to bounce, and as I said before, you will simply never enjoy driving a car when you are frightened of a mid-corner bump…

 

Nearly a year ago we bought a 2005 GD3 JDM Fit. It’s in brilliant condition for a ten-year-old car. It still looks almost brand new. Mechanically there is nothing worn at all. But that’s not surprising for a Honda at 80,000 km (50,000 miles.) And the price was very favourable. Its primary use is daily transport in city traffic, but it will also be used for long trips where its big internal space and quite reasonable fuel economy will be very handy. But after our first long-distance drive in it there was a very real danger there wouldn’t be another. There was no way I could put up with the jarring, bouncy ride. Fortunately I’ve managed to fix it, and get rid of the ghastly understeer reasonably easily...

I’m aware that there are people around the world who race these things. (But why not use something a little more suitable? Like a proper VTEC Honda?) In particular there seem to be a number in the USA who use them for Autocross events. There’s a good chance I might do the same here, but there is a very big difference. Our Autocrosses are almost always on grass. Whereas the American ones are more serious affairs on smooth paved surfaces. They will be using sticky tyres and generating g-forces well in excess of normality, so the little suspension changes I’ve come up with won’t work for them. But if you are just using yours for personal transport, my findings may be interesting for you.

The principles are the same of course no matter what the car is used for. The wheels still need to go up and down and the dampers (shock absorbers) need to be correctly configured to dissipate the bump energy absorbed by the springs, and control the progressions in and out of pitch and roll. Importantly the roll stiffness at the rear needs to be high compared to the front, so that weight can be taken off the outside front tyre in a turn to reduce understeer while the inside front tyre is loaded more to increase traction.

One of the more interesting rally cars I was ever involved with was a GD Jazz with a K20A JDM DC5 Integra power unit carefully squeezed into the front. Intelligent suspension settings and an ultra-close set of ratios in the six-speed enabled it to be the fastest 2WD car in any rally it contested (or at the very least to be in close contention) and it won many awards and trophies. The overwhelming lesson from this car – and all my years of motorsport – is that lowering a car to the point where any corner of the suspension tends to run out of upward wheel travel is always a big mistake. You can try to eliminate this problem with stiff springs to reduce suspension movement, but there quickly comes a point where the increased loading placed on the tyre by the stiff suspension on hitting a bump (and there are always bumps) not only decreases the available grip but also makes the car more unpredictable and harder to drive. For wheel movement in the other direction, too little available wheel droop will cause tyres to leave the ground. You need wheel travel both ways, upwards and downwards.

One question which remains in my mind is whether or not the cars made in Japan and USA have even the slightest similarity in terms of suspension settings or even layout.

All I can do is quote the various measurements I have made and let others tell me what they find.
(And I’m still wondering why you would want to use a Fit or Jazz with its original engine in any serious sort of motorsport. Why not pick a proper performance-oriented Honda?)


In New Zealand we were for quite a long time in the very privileged position of having a genuine suspension guru looking after the suspension specifications of most of the Toyotas sold here. Chris Amon, whose best points tally in a Formula 1 season was 4th, without ever quite managing to ever win a race, was once described by Ferrari F1 team director Mauro Forghieri as the best test driver he ever had. After publicly (and rightly) criticising Toyota he turned their ill-handling Japanese designs into very rewarding cars to drive, with impeccable road manners – and he only changed spring, anti-roll bar, and damper rates. I’ve carefully measured the various models he played with, and their horrible JDM counterparts, so I know what he did to make a FWD road car handle well.

Typically, you need around four inches of upwards wheel movement (bump), which will be reduced a little once you include the bump stops; three inches of downwards wheel movement (droop); similar spring rates front and rear (measured at the wheel, not the spring itself); and a significantly stiffer rear bar than the front. If there is less travel than this there will need to be compromises.

Modern cars tend not to have bad camber and toe changes with wheel travel, but it’s never a bad idea to check these just in case. Spring rates need to be just stiff enough to limit pitch (nose lift on steady acceleration and nose drop on steady braking) to something reasonable, and the anti-roll bars likewise limit body roll, again in a steady-state cornering situation.

Rear roll stiffness can be increased to the point where the inside rear wheel becomes completely unloaded, or even lifts slightly, in a turn. That’s not a problem. Because of the nose-heavy front-to-rear weight split, the outside rear tyre will always be more lightly loaded than the outside front, so front tyre grip will always be the limiting factor. Oversteer will only ever happen if the transition into the corner isn’t properly governed by the dampers, or the suspension actually bottoms out. Both conditions are unfortunately common when suspension is modified without careful thought and measurement, especially when it is set too low. (And, yes, other situations like a half flat rear tyre, or cold rear tyres on the first corner of a race…)

The low-speed rebound valving in the dampers needs to suit the spring rates, allowing the car’s body to return promptly to its rest height without overshoot, while the low-speed bump valving assists the spring and bars to take care of transitions (pitch and roll). High speed damping needs to be limited just enough so the springs, not the shocks, absorb most of the vertical motion of bumps. And that’s about it. It’s not really difficult engineering but it is almost impossible to find it built into a car these days…

 

The Brochure from Honda New Zealand describing the GD Jazz praises its “Class-leading ride and handling.”

Well, for me the ride and handling is as bad as it gets. Easily the worst car I have ever owned. I took the front anti-roll bar off. (Not a 5-minute job.) It helped, but not very much, and the cringeworthy understeer was lessened, becoming almost tolerable.

But after finally getting completely frustrated with all the bumps and bounces, and ripping the suspension out of the car to measure it, it turned out that the front wheels could move upward only 15 mm before the dampers come into contact with the hidden, long, stiff bump stops on their shafts. Cutting over 50 mm off the bump stops (don’t worry, there’s still 40 mm of them left!) suddenly released two whole more inches of suspension travel! What a difference!

It was even worse at the rear – there was only 5 mm (That’s right, 5 mm!!) of bump travel before meeting the bump stops. These are hard to get to but they can be cut 60 mm and still leave 55! Because of the geometry this liberates just over 70 mm more unhindered rear wheel travel – nearly three inches!

The difference in ride was astonishing. It actually took bumps in its stride. The front damping is pretty much perfect, while the rear was a little over-damped in rebound – perfectly suited to the stiffer rear springs I wanted to try later.

Perhaps at some point German engineers did refine the suspension as claimed in the brochure. But then somebody else completely wrecked it with these stupid, long, stiff bump stops and soft rear springs. Where the race-car-stiff front anti-roll bar came from I have no idea, but removing it was a very good move. I now finally had a Honda which handled very nicely thank you, especially on rougher roads, and it laughed at speed humps. A very long way from where we started. Being picky, I could say that it still understeered a bit, and that a little less body roll wouldn't hurt. Stiffer rear springs should sort both those out (a 25mm rear anti-roll bar would be even better, but very much harder to arrange) – but I wasn't in any real hurry to change it now. It became an actual pleasure to drive.

Then, while idly sorting through the boxes of cast-off race and rally springs, as it happens I found some suitable Tein springs. 250 mm long (very nearly 10”), 65 mm internal (2 ½”), and rated at 6 kg/mm, a little over 325 lb/in, they are useful 30% increase on the 250 lb/in original rears. They fit perfectly into the rubber top seats of the Fit – and I even fitted the rubber wrap from the bottom coil of the Honda springs. Ride height is measured at just 3 mm (1/8”) lower than the stock springs.

On driving it I am sure that this is as stiff as you would ever want on the road, and that the stock JDM shock absorbers (52610-SAA-J111-M1 written on them) are just coping with them. They don’t quite make up for the roll stiffness lost by removing the front anti-roll bar, but they seem to have eliminated the understeer – just like the Amon Toyotas. And speed humps are still fine, although you can certainly feel that the rear is firmer. We are going for a drive over some windy, hilly country roads on the weekend, so it will be a good test.

There’s one small doubt. The GD Fit does have a surprising amount of travel, but run-of-the-mill coil springs like the Teins don’t have a lot before the coils go solid. These ones will coil bind at about the same time as the bump stops are compressed all the way down to about 10 mm of their height. And this is certainly possible on very big bumps. Not often – maybe never – but if it ever does, it will bend the spring mounts on either the body or the rear beam. So the permanent solution would be something like the Eibach 1000.2530.0325, which has a 20 mm safety margin even with zero bump stop left.

None of these improvements are easily available to the ordinary owner. I’d also be surprised if any Honda franchise would contemplate altering the suspension from the original. But if you can do it yourself...

I’ll see what the weekend reveals!

 

The trip away revealed a lot about some drivers' inability to notice a long queue of cars building up behind them, so offered few opportunities to explore he limits of the new setup. Understeer is certainly even further reduced, and body roll is now completely acceptable. But as I feared, the 330 lb/in (6 kg/mm) Tein springs are a just little too stiff for comfort, unless there is quite a heavy load in the back of the car. The best way, as I knew all along, would be to cut out the rear 19 mm anti-roll bar and weld in a proper, 25 mm one. But that's not an easy task - nor necessarily legal.

Instead, adding a 22 mm Progress bar would certainly be right in the ball park size-wise. But I have reservations about the attachment points. The steel of the spring seat (or perch) isn’t very thick. (Maybe I will build a bar myself one day – you can buy spring steel bar and bend it by heating it red hot with a gas torch to soften it where you want the bend. When it is all finished, and welded to some very strong brackets, it needs re-tempering in a furnace by a spring maker.)

 

I've just managed to get the Fit away by myself over a few pieces of road I know well. It's certainly a whole new car compared to what it used to be. Not so worried about passenger comfort this trip, the stiff rear springs didn't bother me and I enjoyed the drive a lot. It sits flat enough and you can be very confident entering corners fast, knowing it won't understeer, and if the corner tightens it happily tucks in tighter the instant you lift off. But it's not perfect.

A sudden swerve when a car changed lanes on me without looking or signalling didn't feel great and showed up that the damping in roll isn't quite as good as I'd hoped. So the OE shocks seem to be the limiting factor here.

One of the sections of the journey was an unsealed road. Fine gravel over a very hard, smooth base. Wide, but plenty of bends and crests. A typical Northland New Zealand road of the type loved by rally drivers. The tyres aren't well suited to this type of road, with rounded shoulders and no open tread grooves - and high pressures. So at any sort of pace the car was always moving around on the little rolling stones it was running on. Mostly it behaved itself very well - no particular understeer and reasonable traction - but in a couple of places where there was a trace of corrugation (washboarding) it very definitely suffered a significant loss of rear grip even with power on. Not enough to snap sideways but enough to demand my attention. A bit disconcerting with a wishy-washy CVT, and for the very first time I was glad to be able to select "Manual 7-speed" mode so I had some positive control.

I think there are two factors here - once again, the limited ability of the rear shocks, but this time coupled with the unnecessarily heavy 15" wheels. The damping just can't control them under such conditions. I've got some much lighter wheels that will fit it, both 14" and 13", and next time I get a chance to drive it I'll see how they change things.

On returning home I jumped into one of our trusty mid-90s Amon-suspension Corollas. Softer, but stiffer. I mean softer over bumps, yet stiffer in roll, and very well damped in both. So I guess that despite the fact that I'm now perfectly happy to drive the Fit on long trips, and even enjoy it a lot, there's still a way to go. I'd still choose the Corolla if driving pleasure was the only criterion...

 

not to say that there isn't good info here... but it seems like you're using a sticky more as your own personal blog on your suspension setup...

 

Yes, a fair point, but I think I'm applying a lot more theory, measurement and practical science than many of the others...

And I'd like to know more about the differences between the American, European, Australasian and Japanese versions of the various models. It would be pretty pointless if we were actually all talking about quite different cars.

 

WOW there is some serious information here! Honestly I just skimmed through most of it but I am curious to the set-ups people are doing on competitive Fits. This would be AutoX or Road Race.

Personally what I have found in the last 2 years in AutoX on a National competition level is:

- The 22mm Progress RSB doesn't provide the rotation needed. Unfortunately the crappy H-Beam rear is a tough one to figure out and will require some custom application for a nice hollow 26mm+ RSB design.

- Springs are a mother fucker for this car. I am currently at F 560lbs (10K) / R 784lbs (14K). Since the Fits have high center of gravity i could use even more spring rate. My co-driver and I think the car would respond well with a F 650lbs spring without a FSB and a R 1100-1300lbs spring. I know this sounds crazy but many STF SCCA cars are running "HIGH" rates. If you look at David's SI (Yes a different suspension type but high center of gravity vehicle like the Fit, he is running F 1000lbs, R 2000lbs. YEAHHHHH.

Anyways just wanted to throw some food for thought and see what others have tried.

 

^Is it an EP3?

 

granted this will take another route to the same thing, but doing a fwd-stagger (wider front wheels/tires) will also help... there's a point when a higher spring rate won't do you any favors...

 

The vehicle is an GD.

I run 15mm spaces in the front to give a false reverse stagger. Unfortunately Bridgestone doesn't make a 225/45R15 RE71R yet so the 205's have to stay for now. I understand that springs will at some point do nothing else but wheres is that point? I think with 800lbs now theres more to go.