Designing and implementing a coilover suspension

[chevynut]

Over the years I have designed many coilover shock setups for my C4 suspensions and other applications, and learned a lot in the process…..I’m still learning.

I will confine this discussion to “universal” coilovers, those with eyes on each end with the spring captured by caps on the coilover. There are some vehicle-specific coilover kits where the spring rests in a pocket in the frame, but those are somewhat different. The “universal” coilovers are typically used with front frame clips or rear suspensions.

Some guys say they don't like coilovers, and I have never understood that. They are a nice, compact package containing the shock and spring in one unit. Spring rate changes are simple as are ride height adjustments, much simpler than with just about any other suspension setup. And there are lots of spring rates to choose from. Coilovers are used in a lot of high-end sports cars and exotic cars as well so they obviously have benefits. And in the end, a spring is a spring so one type of spring should perform pretty close to any other type.

I believe the reason some guys don't like coilover suspensions is that many coilover implementations are either improperly designed or they are heavily over-sprung resulting in a harsh ride. I’ve made the latter mistake myself, as I’ve learned. It's easy to slap a spring on a coilover, adjust the ride height, and call it done. On most applications you could probably put a solid tube in place of the spring and get the car to the correct ride height, if the coilover length is correctly chosen. But it would ride like a tank. Also, coilovers are fairly expensive, at a cost of around $1100-1500 for a set of 4.



Selecting the shock

The first step in implementing a coilover suspension is to select the specific shock for the application. Most popular coilovers use a 2 ?” spring ID so they’re very similar. The main difference is in the lengths offered, valving, and adjustability. There are single adjustable coilovers that are either rebound or compression adjustable, and there are double adjustable ones. The body of the shock is threaded to enable ride height adjustments to get to design ride height by turning a threaded nut on the body.

Once a brand and type of coilover shock is chosen, you need to choose the shock length. A longer coilover has a longer stroke which allows more suspension travel and will enable the use of softer springs to support the weight of the car. I’ll talk more about this later.

A coilover shock should sit at about 40-60% compressed when the car is at ride height per my discussions with QA-1 application engineers. You can adjust ride height up and down within this range, but you can't adjust ride height indiscriminately. If you need to make significant ride height adjustments, the shock mounts should be moved to keep the shock inside the desired ride height range. If operated outside of this range, the shock could top out or bottom out easily.

All coilover shocks have a recommended “ride height length” range specified by the manufacturer. One should look at available shock lengths before designing the shock mounts to get an idea of what length they should be shooting for. A rear coilover would typically have a longer shock than the front but not necessarily.

For example, a QA-1 DS401 coilover shock has an extended length of 14” and a fully compressed length of 10.125”, both measured from eye to eye. This gives a stroke length of 3.875”. The specified ride height length range is 11.5” to 12.5”. This means the mounting points should be at 12” plus or minus ?” apart for correct implementation. At 12.5” the shock is compressed 1.5” or 39% and at 11.5” the shock is compressed 65%. I would shoot for 12-12 1/4” in this case or 45-51% compressed as this leaves a bit more room for compression.

The proper way to implement a coilover shock is to set the suspension at the desired ride height and build the shock mounts such that the coilover shock is compressed at the desired amount at that ride height. The adjuster nut on the coilover body is then used to set the car at the designed ride height after a spring is installed.

If there's one drawback to coilover suspensions, it's that typically coilover shocks have a relatively short stroke, some as short as 3”. However, with a properly designed suspension and coilover mounts, the stroke is usually adequate for most street applications since the suspension’s “motion ratio” and shock angle allows more movement in the tire than at the shock.

For example, a typical front coilover with a 12" ride height length (eye to eye) will have a stroke around 4". That means you can set it to give 2" extension and 2" compression from ride height. Due to the shock angle and motion ratio of the front suspension, this gives you more than 2" of up and down travel at the wheel which is plenty as long as the shocks are properly set up. A typical C4 front suspension installation would result in over 5” of total travel at the wheel.

Now that the right ride height length is determined and the proper shock mounts installed, it's time to pick a spring.

[chevynut]

Choosing a Coilover Spring

All springs operate on the same principle: force increases with deflection whether you're talking about a leaf spring, torsion springs, or coil spring. Springs absorb and store energy when compressed, and release it back into the system when extended. An archery bow stores energy as it’s drawn, and that energy is imparted into the arrow and used to propel the arrow when it’s released. A vehicle suspension spring absorbs energy when compressed, and releases it on rebound.

Most simple springs operate linearly (force increases directly proportional to deflection) but there are progressive rate springs where force increases non-linearly with deflection. In this discussion I'm referring to "true"coilover shocks, where the top and bottom of the spring are both constrained by the shock itself, not the type used with a stock tri5 suspension. I’m also discussing linear springs, not progressive springs.

Let’s talk about spring rates. The spring rate is the number of pounds of force it takes to compress a spring 1”. So a 400 lb/in spring would require 400 pounds to compress it 1”. The wheel rate is the spring rate corrected by the geometry of the suspension. It is the rate observed AT THE WHEEL of the car. It is the force in pounds required to move the wheel up 1”. The wheel rate is always lower than the spring rate. This is due to the leverage, or motion ratio of the suspension and the angle of the shock. The wheel rate is what determines the ride quality, or stiffness, not the spring rate. So when guys try to compare spring rates with different suspensions, they’re talking apples and oranges.

I have typically picked a spring rate based on the calculated or actual weight of the car and specific suspension parameters. I calculated the spring compression at the designed ride height length, and added the additional preload needed by the adjuster nut to set the car at the correct ride height based on axial load on the spring. This calculation used the suspension geometry and static loads to determine the spring rate required to support the car. If a guy wanted a stiffer suspension, we'd use a higher spring rate and adjust the nut lower. If he wanted a softer suspension we would use a softer spring and set the adjuster nut higher. Both result in the same ride height, but with different wheel rates.

I mentioned previously that a longer shock allows the use of a softer spring. Let's look at an example: a coilover with 5” of stroke would be compressed at 2.5” at 50% compression. A coilover with a 3.5” stroke would be compressed 1.75” at the same 50%. With the same spring rate of 300 lb/in, the 5” shock would support 750 pounds of axial load, while the 3.5” shock would only support 525 pounds. If you needed to support 750 pounds, you would have to put a stiffer spring on the shorter stroke shock to do it at 50% compression, assuming no additional preload by the adjuster nut. It would require a 428 lb/in spring, resulting in a stiffer ride. It’s possible that one could adjust the nut higher with the shorter shock using the same 300 lb/in spring, but you might run out of adjustment. It depends on the shock as shorter shocks typically have less adjustment room as well.

I recently learned that many racers and pro-touring guys are using the natural frequency of the car's suspension to select spring rate. I had not heard of this and I think the concept is relatively new in most car circles. The natural frequency is determined from the wheel rate and the unsprung weight of the car.

The best way to understand this is to look at a light car versus a heavy car, both with the same wheel rate. The light car will tend to “bounce” over bumps, while the heavy car will “glide” through them. That’s one reason an old Cadillac drove so smooth on rough roads.

Suspension natural frequency is calculated as

F = PI*SQRT(WR/SW).

Where F is the frequency, WR is the wheel rate, and SW is sprung weight. That is 3.1416 times the square root of the wheel rate divided by the sprung weight. The frequency is in hertz or cycles per second. Sprung weight is the weight of one corner of the car, minus the weight of the wheel, tire, brakes, hub, spindle, half the a-arm, and half the shock. It’s the weight supported by the spring.

For a factory stock car, the frequency is typically 0.5-1.0, for a "lowering spring" car it's typically 1.0-1.5, and for a "rally car" it's typically 1.5-2. For all out road race cars it's over 2.

So, given the above, one can decide the suspension frequency you want, and knowing the sprung weight of the car you can calculate the wheel rate needed. From the wheel rate and the geometry of the suspension, you can calculate the necessary spring rate. This calculation is:

WR = F^2 * SW/9.869

Then you can calculate spring rate as

SR = WR/ (MR^2 * ACF)

Where MR is the motion ratio and ACF is the shock angle correction factor which is the COSINE of the shock angle from vertical. COSINE is easily found on a calculator with trigonometric functions. Motion ratio is how much the shock is compressed for every inch the wheel moves upward. It can also be calculated from the geometry of the lower a-arm and is less than 1.0.

I have not found much information as to whether the frequency of the front and rear of the car should be close to the same, or whether one should be higher than the other for best results. I also have not found any information on how a swaybar affects any of this.

[chevynut]

Here’s a real-world example:

When I assembled my Nomad chassis I picked a 450 lb/in front spring because my static force calculations gave me that spring rate based on the suspension geometry and where I thought I wanted the adjuster nut. My BBC is fairly heavy and I wanted the shock to be around 40% compressed. A heavy spring will feel softer in a heavier car with the same suspension due to inertia of the heavier weight so I wasn’t too concerned about over-springing it. Note that the frequency will decrease as weight is added to the car.

At that 450 lb/in spring rate and my C4 suspension motion ratio and shock angle correction, I calculate a wheel rate of 261 lb/in and a frequency of 1.68 Hz based on my expected front corner sprung weight. The frequency is a little high (in the rally car range) but I plan to drive the car to see how I like it before I make and changes. Using that spring rate on the front coilovers, it looks like the car is going to sit at the near the perfect design ride height when the rest of the front end sheetmetal, interior, and windshield is installed. I have plenty of adjustment left up and down and the spring is compressed about 2.5" total. If the front feels too stiff for my liking, I will use the natural frequency calculation to determine what spring rate I want.

In the rear, I picked a 500 lb/in spring. Due to the fact that I got little or no feedback from my more recent customers, I used an early implementation with more of a shock angle to “calibrate” my spring rate calculator. The reason is that it’s more difficult to calculate the motion ratio of an independent rear suspension, or IRS by taking measurements, or at least I don't know how to do it . It turns out that the increased shock angle resulted in me calculating a higher spring rate than needed with my newer chassis.

After putting 95% of the weight on the rear of the car I realized the springs are probably too stiff. So I decided to try to take some measurements, and calculate the wheel rate and frequency to see where I am. I actually have two variables I don’t know, one being the IRS motion ratio and the other the sprung weight of the rear corner of the car. But I estimated the latter and “calibrated” my calculator to derive the motion ratio.

The result is that my wheel rate is 291 lb/in and the frequency is 1.96 Hz. I suspect this is too high, and I found that I will have to adjust the nut way too low to get the correct ride height. So I will probably need a softer spring to get where it needs to be. But again I will probably drive it to see how it feels when the car is done if I can get the ride height correct. It’s fairly easy to change springs especially in the rear.

I had always thought intuitively that you could go up or down 50-75 pounds in spring rate without running out of adjustment on a coilover but never spent much time doing calculations to prove otherwise. After doing a bunch of recent additions to my calculator, I have found that the range is much larger than that so there’s more room to get a stiffer or softer suspension than I realized.

[TrifiveRichard]

Thanks for sharing your knowledge and experience, Laszlo. This is useful. The only suspension system I’ve built was from kits for the 1949 Ford pickup we built a number of years ago. It is a mustang II type front and triangulated 4-bar rear, both with coil overs.

392E29C3-2692-438F-A8A5-7D6AD02FFAA7.jpeg

[chevynut]

Thanks for sharing your knowledge and experience, Laszlo. This is useful. The only suspension system I’ve built was from kits for the 1949 Ford pickup we built a number of years ago. It is a mustang II type front and triangulated 4-bar rear, both with coil overs.

Thanks Richard. Are you happy with the ride of your truck? Do you recall what spring rates you used in the front and rear? I think the motion ratio in the rear of a solid axle car is pretty easy to measure. I'm pretty sure you measure from the center of the tire to the bottom of the opposite lower shock mount and divide by the tire track. On the front you measure the lower a-arm from the a-arm pivot point to the lower shock mount, and divide by the length from the pivot to the lower balljoint. If you're interested, I'm curious what natural frequency you ended up with.

[Custer55]

Good Information. So how does the shock angle affect total travel and spring rate? On my C4 set up I angled the rear shocks more inboard than stock. Also I used the stock 90 Vette springs as my 55's weight was only about 190 lbs more than our 90 Corvette. Front to rear balance was nearly identical. My shocks have the 3 settings which you could adjust with a switch on the console (I just adjust them manually). I have the fronts set on the softest setting and the rear on the stiffest setting to get the best ride quality. The rear seemed to soft and bouncy with the shocks on the softer setting.
Overall I am very happy with how mine rides and drives. My ride height also came out right where I wanted in the front. The rear was a bit high but was easily fixed with longer spring bolts. Coil over shocks would sure make it a lot easier to dial everything in just right though, they just weren't within my budget.
Brian

PS, welcome back Cnut.

[chevynut]

So how does the shock angle affect total travel and spring rate? On my C4 set up I angled the rear shocks more inboard than stock.

The more the shock is off of vertical, the more wheel travel you get for a given shock movement. It's the shock travel divided by the cosine of the angle. So if the angle is 20 degrees from vertical, the cosine of 20 is .94. Then you get 1/.94 or 1.064 inches of wheel travel for every inch of shock travel. If the shock was at 30 degrees, the cosine of 30 is .87 so you'd get 1/.87" or 1.15" of wheel travel for every inch of shock travel. Also, you'd need a stiffer spring as the angle increases to support the weight of the car at the same spring compression.

Also I used the stock 90 Vette springs as my 55's weight was only about 190 lbs more than our 90 Corvette. Front to rear balance was nearly identical. My shocks have the 3 settings which you could adjust with a switch on the console (I just adjust them manually). I have the fronts set on the softest setting and the rear on the stiffest setting to get the best ride quality. The rear seemed to soft and bouncy with the shocks on the softer setting.

So you have the FX3 electrically adjustable C4 shocks? Are you using the stock C4 computer for them? The engine in a tri5 would have to sit further forward than in the Corvette so I would assume more weight would be over the front tires. There is some adjustment in the front C4 spring with shims between the spring and k-member, as you know. But that adjustment is pretty limited. With a heavier engine you would have a hard time maintaining ride height.....I have some customers that have had some issues with it.

There are a few spring rate choices for the C4s but they are all designed to put the car at the same height. So installing a stiffer spring doesn't always result in a ride height change, it just changes the wheel rate.

Overall I am very happy with how mine rides and drives. My ride height also came out right where I wanted in the front. The rear was a bit high but was easily fixed with longer spring bolts. Coil over shocks would sure make it a lot easier to dial everything in just right though, they just weren't within my budget.

Yes the coilovers definitely give more flexibility and make adjustment easier. Do you know the spring codes on your front and rear springs? I have a chart that gives all the spring rates and wheel rates. It would lbe interesting to see where GM put the natural frequency on these cars. I do know that the '84 Z51 had a super stiff springs and I've sold some of them to guys wanting them for autocross.

[chevynut]

I just looked at the FX3 suspension spring codes on the charts I have. Here's the codes and the listed spring rates.

Front
89-91 FHA 532 lb/in
92-93 FSR 418 lb/in
94-95 HA 343 lb/in

Rear
89-93 NYR 228 lb/in
94-95 RR 149 lb/in

It's clear to see that as time went on GM kept softening the C4 springs. The 84 Z51 had a front spring rate of 583 lb/in and 500 lb/in in the rear. By 96 the base front spring was 343 lb/in and the stiffest one was 414 lb/in.

The charts also have the wheel rates listed, but they seem soft to me. Maybe it's because the spring contacts the a-arm further inboard than the coilover mount would be.

For example, it shows the front FHA spring with a 532 lb/in spring rate and a 153 lb/in wheel rate. The HA spring at 343 lb/in spring rate supposedly gives a wheel rate of 109 lb/in, about the same as a stock Tri5.

Looking at the overall ratio between spring rate and wheel rate for the FHA spring I get 153/532 or .2876. There is no shock angle correction for the transverse spring so to get the motion ratio you would just take the square root, which would be .5363. This is the ratio between the a-arm pivot to the spring pocket center, divided by the length of the a-arm from the pivot to the balljoint which is 15" on a late C4. So multiplying .5363 by 15" one would get 8.044" from the a-arm pivot to the spring pocket or a little more than halfway past the center of the a-arm. Maybe that's about right, but I haven't measured it.

So let's assume the WR of 153 lb/in is correct. A 1990 C4 weighs 3223 pounds and the weight distribution is almost exactly 50/50 so 1611 pounds would be on the front suspension. One corner would be 806 pounds, and subtracting about 150 pounds of unsprung weight (maybe a bit high) you get 656 pounds of sprung weight. This gives a front frequency of 1.52 Hz.

For the rear we have a similar wheel rate of 147 lb/in for the NYR spring. Since the rear corner unsprung weight should be about the same, we can do the same calculation and end up with a frequency of 1.49 Hz. The softer springs will give a lower frequency in the 1.3 range using the above numbers.

[Custer55]

The shock angle I have explains why I need the rear shocks on a stiffer setting then. I am not using the console switch to change the shock settings. I just do it manually with the knob at the top. They seem to stay where you set them fine without the factory electrical adjusters on them. Probably wouldn't make sense to use the factory adjusters anyway as I have the front and rear at different settings. I don't have the spring codes handy but what you found is most likely correct. I do remember reading what you found that they kept going to softer spings on the later C4's. My guess would be they did that for better ride quaulity and adjusted the shocks as well to match the spring rates.
Thanks for all the great info.
Brian

[chevynut]

Si we can now see that GM was shooting for around 1.5 Hz natural frequency with the C4 Corvette. That seems to fit with this info:

"To properly design a suspension for a car’s weight, aerodynamics, and tires, we often use suspension natural frequencies as a starting point for picking wheel rates and spring rates.


Common suspension frequencies:
0.5-1.0Hz Passenger cars, typical OEM.
1.0-1.5Hz Typical lowering springs.
1.5-2.0Hz Rally Cars.
1.5-2.5Hz Non-Aero racecars, moderate downforce Formula cars.
2.5-3.5Hz Moderate downforce racecars with up to 50% total weight in max downforce capability."

While my car is not a race car, I do want good, balanced performance and I want a good ride quality. I don't want it to ride like a Cadillac, but more like a sports car. If you know someone with coilovers on their car that is unhappy with them, I would recommend they do these calculations to see where their frequency is. I'm betting their car is over-sprung.



Here's a pretty good read too...

http://www.tracktuned.com/feed/2016/11/27/overspring-spring-frequency

[chevynut]

Here's some info I found on front vs rear frequency....

"The rear ride frequency must be greater than the front. This is due to the fact that there is time delay between hitting an obstacle by front and rear wheels. The rule that the rear suspension should have a higher spring rate (higher ride or natural frequency) rationalized the observation that vehicle bounce is less annoying as a ride motion than pitch.
To minimize pitching tendency of vehicle one must plot the front vs rear ride frequencies in an amplitude vs time graph by considering the time lag. By performing various iterations of frequencies one can identify which two frequencies matches early so that there will be only bounce motion."

"One common idea is to design for “flat ride,” which is to have the rear suspension frequency roughly 15% stiffer (or faster) than the front. The idea is that since the front hits a bump before the rear, the rear should be a little stiffer in order to settle at the same time. Flat ride was initially designed to improve comfort on vehicles with soft dampers, but it’s also useful for sports cars. Many racecars do not use flat ride and there can be benefits to higher front frequencies. Ultimately, the best spring rates for a race car are the ones that allow it to go fastest around a track. This rarely comes from plugging numbers into a formula, but suspension frequencies and flat ride are useful starting points. Many variables are not accounted for with these methods, including sway bars and alignment settings."

"With equal or higher suspension frequency in the front, the pitching motion and out of phase front to rear motions will be more pronounced and the vehicle will have a tendency to pitch back and forth and not remain level after encountering some bump."

[567chevys]

Hello Laszlo for Posting this stuff , How do you find the time to do all the stuff you have been doing lately ?

This stuff will help lot of people.

Thanks Sid

[55 Rescue Dog]

I took a chance on my C4 55 and am running the softer stock transverse leaf springs and shocks front/rear from a 96 Grand Sport and don't think I could have picked better spring setup. GM has been using them through 4+ generations of Corvettes for a reason. The car just feels totally planted and balanced without being overly harsh. It absorbs the dips in the road better than anything else I've driven and stays flat front to rear with no pitch or double bounce from the rear like many cars. My stripped down 55 weighs more than I thought it would at 3400 pounds. Roll bar added around a 100lbs, and I still need to put the front bumper on and install door windows. I'm not using the rear sway bar yet, and don't think I really need to since the car can be driven hard into a corner without getting loose. I took it to one autocross to find out how the car felt pushed to the limits and was very happy how the car felt, and nothing broke or fell off. I could make it faster, but it is more than enough fun the way it is. It's like driving a tall Corvette on a stretched wheelbase.
1955 Chevy autocross - YouTube

[WagonCrazy]

Good info. Thanks for diligently going thru it Laszlo.

I need to go back and figure out the spring rates I put on the ViKing coilovers on my Nomad. Vague memory says 400# fronts (LS1 aluminum block so lighter weight than steel block) and 500# on rear. But in order to get the ride height I wanted, the rears adjuster nuts are near the bottom of the shock. So I suspect those may be too stiff and was going to buy a set of 450# to just try out and see.

There's another thing to consider here. The changing overall vehicle weight.
Weight of Driver (on left side)
Weight of full tank of fuel vs. near empty. Fuel tank on side of vehicle or centered.
Weight of battery (on one side or the other)
Etc.

Not sure how that would all factor in, but it's probably not enough to worry about when you're first setting up the vehicle. Your calculations should put you "in the sweet spot" as a starting point when you get it on the road and do the final suspension tuning.

Another thing...my ViKings are double adjustable and I've done test drives after having played with compression and rebound settings. Big difference in feel when they both are set all the way to the extremes. The car rides just stiff enough now for my liking, but I still may put a 450# set of rears on just to play with it.

[chevynut]

Good info. Thanks for diligently going thru it Laszlo.

I need to go back and figure out the spring rates I put on the ViKing coilovers on my Nomad. Vague memory says 400# fronts (LS1 aluminum block so lighter weight than steel block) and 500# on rear. But in order to get the ride height I wanted, the rears adjuster nuts are near the bottom of the shock. So I suspect those may be too stiff and was going to buy a set of 450# to just try out and see.

There's another thing to consider here. The changing overall vehicle weight.
Weight of Driver (on left side)
Weight of full tank of fuel vs. near empty. Fuel tank on side of vehicle or centered.
Weight of battery (on one side or the other)
Etc.

Thanks for the feedback Paul. I included all the weights of the things you mentioned, and my spreadsheet is flexible for added weight, but they're really pretty small when you're talking about a corner sprung weight of around 800+ pounds. Notice that the frequency calculations aren't affected by the type of shock or length used. All you need to know is the corner sprung weight and the wheel rate.

At the time we did your build, I was still unsure about how to calculate the effect of motion ratio and when it should be squared in the calculations in my spreadsheet. It's now very clear to me and I modified my spreadsheet accordingly. Also, Wade gave me feedback that a 600 pound spring wouldn't work on the rear of his car and it sagged badly. So I used his information to develop my first pass spreadsheet. As I mentioned, I had a lot more rear shock angle on his frame than on subsequent frames. I think I moved the top rear shock mount outboard right after his build. I'm not sure where yours is but I think it's the new orientation.

At 400 lb/in in front you should be at a wheel rate of 232 lb/in with a late C4 suspension. I have actual weights from Wade's 57 Nomad with LS/4L60E and his was 1870 pounds in front, or 935 per corner. I just went through the numbers and made a better estimate of unsprung weight and came up with 103 pounds in front with my wheels and tires. So your front corner un-sprung weight would be about 832 pounds. This results on a frequency of 1.64 Hz. Adding 100 lbs. to the front corner weight for a driver drops the frequency to 1.57 Hz. A 350 lb/in spring would drop that to 1.47 Hz so that gives you an idea of the ranges and sensitivity.

Going by the suggestion of 15% stiffer in the rear, assuming you like the front as-is, you should be running about 1.89Hz in the rear (no passengers). Wade's Nomad's rear weight was 1816 pounds and I come up with an un-sprung weight of 97 pounds with my wheels and tires. Using those numbers, you should run a 400 lb/in rear spring to get to 1.88 HZ.

I don't think dropping to 450 in the rear is worth doing. I'd go to 400 if you like the front where it is. If you want the front softer, I'd do that before I picked a rear spring and go through the calculations again. It would be better if you could actually weigh your car front and rear to get an actual weight but I'll bet these are close.

One more note....the tires also act as a spring so a wider tire with a larger sidewall will ride softer than one with a shorter sidewall. This doesn't affect the static wheel rate, but it does affect the dynamic wheel rate as the tire compresses over bumps or going into turns.