The Central Hub for DSM Community and Information

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. This is where the DSM platform history is documented and archived. Log in to help us in our mission, and to remove most ads from the browsing experience.

Hypercoil Springs

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

DG-FNR

20+ Year Contributor
234
62
Oct 21, 2002
Geary, NB, Canada
This is really a response to a privately-posted question, but the answer is worth sharing in public.

So then, what's the deal with Hypercoil springs?

Hypercoil does not make application-specific "lowering springs" like some other spring companies. They probably would wind you a custom spring that would fit in an OEM DSM application, but it'd probably set you back a couple of grand.

Instead, Real Racecars use a couple of common spring sizes, and Hypercoil makes springs that fit. The only exception is NASCAR, which uses a spring based on some late 60's GM OEM fitment - but because NASCAR is so big and so rich, that OEM fitment is now considered a "Real Racecar" fitment :rolleyes:

Typically, a Real Racecar will use a spring that is either 2.5" in INNER diameter, or 2.25" inches in INNER diameter. An engineer has no real control over the OUTER diameter, because the OD is a function of the ID and the rate (stiffer springs use larger diameter wire)

They come in set lengths, from 3" to 12" by 1".

The big deal about Hypercoils is the extreme lengths they go to to ensure quality. A Hypercoil will never sag. A Hypercoil is guarenteed to be within 3% of it's indicated rate. Every single Hypercoil is rate-tested at the factory before it is shipped.

When you use a Hypercoil, you're using the exact same spring as an Indycar, CART car, WRC car etc etc etc. This is Real Racecar stuff. But because they have huge economies of scale, the springs are actually pretty inexpensive (in the ballpark of $60 per spring)

It's not very often that we get to use Real Racecar stuff and actually have it be CHEAP. Enjoy the opportunity! :)

To use Hypercoils on a DSM, you need to have a coilover kit - they will not fit without one. You also need coaxial upper spring hats (which has been discussed already)

What size depends on the ID of the spring hats, the length of the spring, and the rate you want. For Konis with native coilover hardware and appropriately-relocated snap ring grooves, that'll be 2.25" ID X 8" long by RATE. For most GC and EBAY-sourced stuff, 2.5" ID by 8" by RATE.

On my own car, I'm using 2.5" ID x 8" long X 900lbs/in on the front, and 2.5" x 8" long x 400lbs/in on the rear.

Something else that is helpful (heh) is the "helper spring" This is a flat-wound spring with minimal rate that is designed to collapse when the car's weight is applied to it, but expands when the wheel comes off the ground. It's useful to keep the spring from ratlling around on the perch in lowered applications. I use them on my car.
DG

Learn: http://autocross.dsm.org/books.html
 
EVERY spring manufacturer claims the same low variation in spring rate. Most of them are full of shite.

Agreed - that's why we test the rates of springs here. We have an electronic spring rate checker that we use to see what the rate of any given spring really is.

The only spring that I've never seen fail is a Hypercoil. The "red" springs are hit and miss, and the no-name springs on some of the Japanese suspension kits are garbage.

Your car has to push like a pig with those rates on there. Why are you running a 900# spring on the front?

Because that's what makes a 2G work.

2Gs are nose heavy, so the front suspension carries more of the car than the rear. The front spring/shock motion ratio is lower than the rear, so the wheel has more of a mechanical advantage over the spring in the front - the front springs are "softer" for a given rate. And the front roll centre is lower than the rear, which means the front roll moment arm is longer, which means the sprung mass has a larger mechanical advantage over the springs in roll.

All this adds up to make the front springs less effective, both in bump and in roll, than the rears, rate for rate. And the front springs need to carry more of the car. Generally, you want the front percentage of roll resistance to be equal to the front weight distribution - so that on a car that is 60% nose heavy, you want the front end to carry 60% of the roll resistance.

My weight distribution is 62/38, and my roll resistance distribution is 59/41, which should result in a car that is slightly LOOSE - and it is (at least on corner entry)

Hey DG-FNR does this meanThe spring rates on a certain suspension system would then be all generic or would could they be selected

Practically speaking, generic. There's a pair of great big monster variables over which we have little control and can find little about - the tire, and the driver.

The tire has tremendous influence on the setup, both in its construction and in its dynamic attitude. For example, if the front tires are rolling over on their sidewalls due to insufficiant inflation pressure or insufficiant static/dynamic camber, then the front end is going to come loose and the springs can't do much about it (they CAN help with dynamic camber by adding more roll resistance and changing the angle of the sprung mass at max roll)

The driver also has enormous influence, because he can induce handling problems, and he reports on the over/under steer balance - and might get it wrong. It takes a lot of experience to be able to REALLY feel what the car is doing - and AWD cars don't make it any easier, because the diffs crosstalk and hide balance nuances that are easier to feel on cars that only drive one end of the car.

So it's tough to tell some random guy off the street that a given setup X is going to be perfect for him right out of the box. For example, I wouldn't recommend that just anybody put my setup on their car. There's so much spring in my car because I'm running 11" of great big sticky tires on each corner, primarily on concrete surfaces. The car sees transient cornering forces in excess of 1.6 G. It takes a lot of spring to keep the sprung mass under control when you've got that much lateral grip.

When we go to asphalt surfaces, it's not unusual to pull spring out of the car. The grip is less, so the cornering forces trying to roll the car are less, so we need less spring to keep the sprung mass under control.

For the general public who isn't racing against the Bob Tunnells, Vic Siases, Mark Daddios etc of the world, the concern is less "ultimate cornering power" and more about stance (ride height) and ride quality.

Ride quality is a direct function of the match between the springs and the shocks. My car, with all that spring, rides nice and smooth with no jarring or bouncing - because the shocks have been exceptionally well matched to the springs. OMG IT'S SO BOUNCY!!!! is almost always a shock/spring mismatch (sometimes because the car is sitting on the bumpstops, and bumpstops are really stiff springs)

So it's better for a street car to pick spring rates based on natural frequency, and ensure that the shocks match. If the natural frequency is higher than stock, the car will be both more responsive and be more tolerant to lowering (for bumps of a given amplitude, a lower ride height means more spring rate to keep the car from bottoming)

If a street guy can throw on our setup, and the ride quality is good and the car feels more responsive, we've done our job. If a race guy can throw on our setup, and he's got a good starting point for further development, we've done our job.

So the ATI setup that I'm developing (and yes I'm allowed to talk about it if specifiically asked, as I have been here - and if Management ever gets me a sponsorship quote I WILL be pitching my boss on actually sponsoring the forum) will be set up with a specific natural frequency and matching shocks.

You my friend are retarded. His car absolutely MUST HANDLE LIKE ASS.

ROFL!

Let's see... on these spring rates the car won:

- 2002 SCCA ProSolo Street Modified National Championship
- 2002 SCCA ProSolo Honda Street Challenge Championship
- 2002 MachV "Ultimate DSM" trophy at the Buschur Racing DSM Shootout
- 2003 SCCA CENDIV Divsional Championship
- 2004 FIA AutoSlalom Canadian National Championship
- and a host of individual event wins at SCCA Pro and National Tour events in 2002, 2003, and 2004

Yup.... handles like ass. :D

For more info, check out http://farnorthracing.com

DG

Learn: http://autocross.dsm.org/books.html
 
DG-FNR said:
Generally, you want the front percentage of roll resistance to be equal to the front weight distribution - so that on a car that is 60% nose heavy, you want the front end to carry 60% of the roll resistance.

I've seen something like this said in two ways that are subtly different. Sometimes I see this advice given as you did - viz. "you want the roll resistances to parallel the static weights" - and sometimes I see this in terms of weight transfer - viz. "you want the weight transfers to parallel the static weights." These are not the same thing, as, of course, you and I know.

If you are a fan of roll-resistance view, could you please say why. I understand the weight-transfer view (because I think that it's in Staniforth), but I don't quite get the roll-resistance view.

- Jtoby
 
If you are a fan of roll-resistance view, could you please say why. I understand the weight-transfer view (because I think that it's in Staniforth), but I don't quite get the roll-resistance view.

Strictly speaking, it really is "weight transfer" that we want to parallel the static weight distribution, and it is the weight transfer numbers that I use to determine what the expected final balance will be.

The roll resistance distribution and the weight transfer distribution are pretty close to each other - within a couple of percent - so I tend to use the terms interchangably.

The other issue is that there's a lot more involved in calculating weight transfer, so it takes a hell of a lot longer to explain. When you consider that the weight transfer via springs/bars is roughly double that of via roll centres, which is in turn roughly 5 times that of via the unsprung masses, it can be seen that roll resistance distribution is the dominant factor in determining weight transfer distribution.

So I simplify a little bit for the purposes of discussion.

Oh, it turns out I was reading the wrong cells for the quoted values given earlier - that was the RM rear sway bar numbers. The actual values are weight distribution 62/38, roll resistance 63/37, and weight transfer 59/41 - slighly loose. Raising the front springs to 1200lbs would exacly balance the car (at least on paper)

DG
 
DG-FNR said:
Strictly speaking, it really is "weight transfer" that we want to parallel the static weight distribution, and it is the weight transfer numbers that I use to determine what the expected final balance will be.

Cool. And I agree that stopping at the point where you have calculated roll resistance is probably easier, but as soon as we all have the data and spreadsheets to play with, no-one on this list will have to simplify anything. I can't wait. You just threw me when you didn't flag this as an approximation or short-cut.

You're not seriously thinking of 1200# fronts, are you? I thought the idea was to reduce the rear sway-bar below stock, instead.

- Jtoby
 
You're not seriously thinking of 1200# fronts, are you?

At some point, other parts of the car start to be more elastic than the springs....

But the natural frequencies with 1200lbs aren't outrageous. It might be worth trying just to see what happens.

DG
 
DG-FNR said:
At some point, other parts of the car start to be more elastic than the springs....

You know, I was working on this a few days ago. Here was the approach that I took:

Instead of calculating a roll angle for the entire car, I calculated separate roll angles for the front and rear. The first thing that jumped out of this was why having a sunroof is, to put it mildly, sub-optimal. It also explained all the damned noises coming from the hatch in hard turns, because the roll angles that my front and rear ends want to adopt at 1g are quite different.

When I did Velocity (for those of you not into autocrossing and/or not from Vancouver, this is a driving school put on by the builder of the fastest autocrosser ever), Joe Cheng stressed that the roll stiffness of the car should always be less than one tenth of the torsional stiffness of the chassis. This point is ignored by all texts that I've seen, since they all seem to make the simplifying assumption that the torsional stiffness of the chassis is infinite.

Well, we all know that the torsional stiffness of a 2G DSM is not infinite and that 2Gs with sun-roofs are not only not infitine but probably measurable with three jack-stands and a frickin yardstick. How bad it is in actual numbers, I don't know, but I can see why raising the fronts much more might not get the balance that a simple spreadsheet claims.

- Jtoby
 
I can't remember if I've asked this here, but does anyone have the "spring"-rate curves for any modern R-compound. At a minimum, it needs to be for a reasonable air-pressure and cover weights of 500-2000#. If you have a three-D plot with rate as a function of both air-pressure and weight, that would be even better. Or I'll settle for an approximating equation. Anything other than just assuming that a Kumho V700 has a "spring"-rate of 800#/inch under all conditions (which is what I've been doing).

- Jtoby
 
Sunroof vs solid roof - I don't know what it is, but I can tell you that it is dramatic when dealing with SP+ prepped 2Gs. As you mention it should be a simple matter to calculate it. I have seen the difference just putting my two cars up on jackstands - no matter what different heights they're set to, my 98 gladly accomodates the discrepancy :-(

The only place I think you'll get specific tyre spring rate data from is the manufacturers themselves. There is generic data in Paul Haney's book, and somewhere else too. I'll dig out the Haney if this would be useful ?

Charles
 
Well colour me stoopid !

It seems that a tyre's spring rate is 80% air pressure, so it changes significantly with pressure.

It also appears that the spring rate of a tyre is nothing more than the deflection created by weight, so we can determine it very simply !

Measure from a point on the wheel to the ground, then slowly jack the car up whilst trying to rotate the wheel. As soon as the wheel can be rotated measure the same point to the ground - that's the deflection caused by that corner weight, and that's the tyre's spring rate at that pressure.

So I guess we don't have worry about all the exotic stuff, all we care about is the end result, and that can be determined as above :)


Well duh !

Charles


P.S. From what I've just read, 12-1600lb/in is a more reasonable estimate to use, 800 would be much too low.
 
ACM said:
It seems that a tyre's spring rate is 80% air pressure, so it changes significantly with pressure.
Agreed. That is why I want curves that have air pressure as one of the independent variables.

.

ACM said:
It also appears that the spring rate of a tyre is nothing more than the deflection created by weight, so we can determine it very simply !
I don't agree. The problem is that the effect of weight on a tire's deflection is non-linear (unlike the springs that you, Dennis, and I use). And one of the most important parts of the curve that we need is where the weight on a given tire is much higher than the static weight, so it isn't just a case of measuring what we can measure with the car in our garages. As I see it, our options to get the information we want (other than getting it from Kumho) are these: 1. add weight to the corners as we do our measuring, using, for example, the bags of water-softener salt that Hey, Culligan Man! just dumped in my driveway, or 2. measure the curve for weights less than the corner weight and extrapolate.

Since I teach my students to avoid extrapolation, this is not my preferred option.

- Jtoby
 
Sorry you don't like it - I'm simply regurgitating my morning's reading matter. You'd have to ask Hoosier, Kumho et al for tyre-specific plots.

You can still dtermine the effect of pressure by repeating the exercise for different pressures. Cross-check your results by performing the same exercise on the other 3 corners with the same tyre, 4 different weight measurements should allow you to validate the data.

Oh, aside from the fact the corner weight will change as you move it...Poo, that will foobar the results. Oh well...

Charles
 
Hoosier won't give you the information - they don't have it to give.

You can indeed get an estimate of tire spring rate using the weight of the car, your scales, and a dial indicator. It's a rough estimate, but ballpark.

A better way is a fixture that can be put in a press, the same way you test suspension springs. It's even better if the fixture is adjustable for camber. It's still better if the tire is warm.

The best way is to send a tire sample to Miliken's test facility, and get the rolling spring rates.

DG
 
ACM said:
Oh, aside from the fact the corner weight will change as you move it...Poo, that will foobar the results. Oh well...


Just goes to prove I'm as dumb post-coffee as I am pre...

I will go an read a bunch more, see what else I can find, but so far this appears to be the consensus.

Charles
 
DG-FNR said:
You can indeed get an estimate of tire spring rate using the weight of the car, your scales, and a dial indicator. It's a rough estimate, but ballpark.

That's where I got 800# for 245/45/16 V700s at about 37 psi. About 7/8" when the fronts come down off the jack. But it was done in my garage on a lark, so the tire was cold. If temp makes a difference, too, then we need at least independent variables: weight, psi, and temp. Plus there's camber. Buggers.

- Jtoby
 
After the test fit, we decided we wanted to do one more protoype run before we released the parts. Ideally, these prototypes will be the final release configuration, and will need no further revisions.

Customers are not beta testers.

That means we're back in the machine shop queue.

Once the final prototypes have been cut and verified as OK to release, they'll go on sale.

Hopefully we'll be able to do a full release announcement once they're ready. I'm trying to negotiate forum sponsorship even as we speak.

DG
 
Minor change to the front bolt circle spacing. Slightly larger change to the rear bolt circle spacing. Minor change to the OD of the plate. Change to thickness of spacers to pull top of shock rod away from hood.

Little stuff; nothing major at all. We just don't want to make a bunch of changes to the drawings, and then do a big production run and find that something got overlooked.

The machine shop credo is "if it matches the print, it's correct" so it behooves those of us making the drawings to make sure that what's on the print is what we want.

DG
 
That makes sense. As you know, I've been worried about your design hitting the hood since early December. Glad it's all coming together. Congrats.

Have you considered giving a set to Sean (who is sponsored by this site) as a way to celebrate their release?

- Jtoby
 
A little late on the thread but I'll throw my 2 cents.

Somewhere I have a picture of a set of used 8" eibachs compared to a brand new set of 8" Hypercoils. The eibachs were a good 1/4" shorter freestanding and this was after 2 years of autocross use. If you are throwing together a "custom" set of coilovers, use Hypercoils. After my eibach experience and talks with Dennis, I won't be using anything but unless forced to. :)
 
jtmcinder said:
That's where I got 800# for 245/45/16 V700s at about 37 psi. About 7/8" when the fronts come down off the jack. But it was done in my garage on a lark, so the tire was cold. If temp makes a difference, too, then we need at least independent variables: weight, psi, and temp. Plus there's camber. Buggers.

- Jtoby

Don't forget the dynamic effect on the skins from centripetal acceleration due to angular velocity of the wheel/tire...

...I suppose it is only a force and can be resolved to cornering velocity for the critical section(s) of any given event.

Come to think though, tire growth, as it is apparently termed, will also reduce the effective pressure in the tire to a degree due to Boyle's Law. This may be pertinent (to tire spring rate) esp. because of the non-linear nature of pneumatic springs.

Good luck,
LB

I liked the world better before I knew that roll center/couple, wheel rates, SAI, scrub radius, Akerman and bump existed. It's a crying shame that all the development and talent is focused on the SLA suspension of the 2g platform. Where's my guiding light?
 
laggin3 said:
DG, PLEASE tell me you have something new on your developments?! :sosad:

First off, he's in Ohio right now getting his butt kicked by The Alien in an Evo and our own Tevenor -- see http://www.dsmtuners.com/racing/2005program.php . Second, it is against the rules to make product announcements in the tech forums, so why don't you take this to PMs or continue the conversation in the vendors' forum, OK?

- Jtoby
 
Add Value - Be Respectful - No Trolling - No Misinformation - Participate Often!
Support Vendors who Support the DSM Community

Build Thread Updates

Latest Classifieds

Back
Top