Progressive vs Dual Rate Coils - JeepForum.com
Search  
Sign Up   Today's Posts
User: Pass: Remember?
Advertise Here
Jeep Home Jeep Forum Jeep Classifieds Jeep Registry JeepSpace Jeep Reviews Jeep Gallery Jeep Clubs Jeep Groups Jeep Videos Jeep Events Jeep Articles
Go Back JeepForum.com > Models > Jeep Wrangler Forums > TJ Wrangler Technical Forum > Progressive vs Dual Rate Coils

End of Summer Sale, 20% OFF!2007 - 2011 Jeep JK Long Arm Lift KitsIntroducing MONSTALINER™ UV Permanent DIY Roll On Bed Line

Reply
Unread 05-20-2013, 02:05 PM   #1
DivineKaos
Registered User
2000 TJ Wrangler 
 
Join Date: Jul 2009
Location: Saint John, New Brunswick
Posts: 506
Progressive vs Dual Rate Coils

Hey All,

Trying to decide on what to swap out my coils for. I'm looking at Savvy (Currie) vs MetalCloak at the moment. What is the difference between dual rate and progressive? Does one provide a better ride over the other?


Any other coil suggestions?

Thanks!

__________________
"A man has to know two things in life, limitations and priorities. You let your priorities get in the way of your limitations." - MrBlaine
DivineKaos is offline   Reply With Quote
Unread 05-20-2013, 07:47 PM   #2
Jeepineer
Registered User
 
Join Date: Aug 2009
Location: Sacramento
Posts: 186
Linear rate springs

I am one of the engineers at MetalCloak. The following three posts were originally posted by myself on another forum but are applicable to your question.




The diagram shows three types of springs commonly used in the offroad industry. I will do my best to describe each in the following posts. I think it is important to describe linear rate springs and their short comings before discussing progressive and dual rate springs.

I am going to provide specific values for spring rate, load at ride height etc. for discussion purposes only. These are not actual numbers because most manufacturers have proprietary spring rates and loads that they believe are ideal for a quality ride or are tuned for their specific systems (so does MetalCloak). All numbers are examples only.

Linear Rate Springs

Stock springs are linear rate springs. Many of the larger suspension manufacturers use Linear Rate Springs. Linear springs cost less, are simpler to design, simpler to inspect and provide fairly consistent ride heights.

Unfortunately they are not good for long travel suspensions for the following reason. I will use an example to explain.

Assume the front driver side spring of a JK has to hold a load of 600lbs at ride height.

Assume the spring must compress to a length of 10” at ride height for a given lifted JK. (Lift height does not matter for this discussion)

Assume our example is a lifted average travel suspension system and the Free Length (fully extended length) of the Spring is 14”

The spring rate of this spring is equal to the Load at Ride Height (600lbs) divided by the Length the Spring must compress (4”).

Actual Calculation

600lbs / (14” - 10”) = 150lbs/in

Let’s assume that 150lbs/in is the perfect spring rate for the perfect ride quality of a JK. Not too squishy (move all over the place) not too stiff (feel every little bump on the road).


Now, let’s look at what happens when we need a spring that will remain seated in a long travel suspension at the same lift height.

So let’s assume that we need the Free Length (fully extended length) of our example spring for this long travel suspension to be 18”.

The 10” compressed length of the spring at the same lifted ride height remains the same.

Again The new spring rate is equal to the Load at Ride Height (600lbs) divided by the length the spring must compress (8”) .

New Calculation

600lbs / (18” - 10”) = 75lbs/in

The new example spring rate of 75lbs/in will feel too soft to driver of the vehicle. The vehicle will role or bounce at the slightest motion.

This is the problem with Linear Rate Springs when you want to make a longer travel suspension that requires longer free length springs....the spring rate just gets too low by the time the spring compresses to the desired lifted ride height.

In other words, if we want to maintain the example perfect ride quality of 150lbs/in in a long travel suspension it becomes impossible with a Linear Rate Spring.

Therefore long travel suspension companies that want their springs to remain seated throughout the travel of the suspension and provide a ride quality at least in the same realm as the stock vehicle must use a spring other than a Linear Rate Spring.

Two solutions are Progressive Springs and Dual Rate Springs which I will discuss in the following posts.
Jeepineer is offline   Reply With Quote
Unread 05-20-2013, 07:49 PM   #3
Jeepineer
Registered User
 
Join Date: Aug 2009
Location: Sacramento
Posts: 186
Progressive rate springs



Progressive Rate Springs

The diagram shows three types of springs commonly used in the offroad industry. This post will discuss Progressive Springs. I am going to provide specific values for spring rate, load at ride height etc. for discussion purposes only. These are not actual numbers because most manufacturers have proprietary spring rates and loads that they believe are ideal for a quality ride or are tuned for their specific systems (so does MC). All numbers are examples only.

Progressive Rate Springs- Many top end long travel suspension manufacturers use Progressive Rate Springs. There are two ways to produce a progressive rate spring.

1) Continuously Variable Rate (Pitch)....or the winding of the spring coils such that the distance between each subsequent active coil (coil pitch) of the spring increases as you progress from the top coil to the bottom coil. In other words, the coil pitch continouosly changes throughout the entire length of the spring. Production of this type of progressive rate spring is rare as it requires specialized CNC spring winding machines.

2) Combining 3 or more Spring Rates (Pitches)....most Progressive rate springs are produced using a combination of 3 or more linear spring rates (coil pitches) wound into a single spring coil. Some manufactures use marketing terms like Triple Rate or Quad Rate springs, however these are really just progressive rate springs with a fancy name. Let me explain what I mean...when one uses three spring rates (coil pitches) in a single spring, each spring rate must also have a transition rate (coil pitch) between each of the three spring rates, therefore a spring with three spring rates must effectively have a total of at least 5 spring rates . Once you have that many different spring rates in a single spring coil it is effectively an estimate of a progressive rate spring. In other words, Progressive Rate Springs, Triple Rate Springs, Quad Rate Springs are all the same thing with just different names. I will refer to all of these as Progressive Rate Springs in this post.

Progressive Rate springs are generally more expensive to make, harder to design, harder to inspect and do not have the ride height consistency of linear rate springs. They are excellent springs for long travel suspension systems. Look at the Progressive Spring in the image above. Notice that the distance between the active coils (Pitch) steadily increases as you progress from the top coil to the bottom coil. This is how you can identify a true progressive spring from a marketing progressive spring.

A true Progressive Rate Spring is progressive by RATE. Meaning the spring rate increases the further the spring is compressed.

Unfortunately a Linear Rate Spring is progressive by FORCE. Meaning the load force of the spring increases the further the spring is compressed. I have seen some manufacturers claim there springs are progressive and technically they are progressive by force but not by rate.

Notice how the last two active coils are the furthest apart (I will return to this later in the post).

So I am going to use the same spring assumptions I used in the last long travel suspension example:

Assume the desired spring rate for the perfect ride is again 150lbs/in
Assume the compressed length for the Progressive Spring at ride height is again 10”
Assume that the Free Length for the Progressive Spring is again 18”

So in summary we need our spring to compress a total of 8” (18” - 10”). When it compresses this 8” and reaches the desired length of 10” it needs to handle a load of 600lbs and have a rate of 150lbs/in at that specific point.

The following is a chart of what is happening with the progressive spring as we compress it to the desire ride height. I hope this chart works on your screen.

Spring Compression___Spring Rate at Compression__Force at Compression___Length of Spring
___0_________________________0____________________ _0_________________18”
___2”_____________________20 lbs/in________________40 lbs______________16”
___4”_____________________40 lbs/in________________120 lbs_____________14”
___6”_____________________90 lbs/in________________300 lbs_____________12”
___8”____________________150 lbs/in________________600 lbs_____________10”

So this is an example of a progressive spring with a continually changing rate that will work in the example long travel suspension used above.

As you can see the Spring Rate is changing as the spring is compressed and aproaches the desired ride height. This is what makes it more difficult to provide a consistent ride height because the Rate and the Force of the spring are both changing at the point that the user wants his/her ride height. Remember I said it makes it more difficult not impossible.

In addition, as the Progressive Spring in the image above, compresses every coil of the spring will move the same distance resulting in the top coils slowly collapsing on top of one another before the lower coils collapse on one another. As a coil collapses on a previous coil it becomes fully supported or solid and no longer acts as an active part of the spring at all. This collapsing of coils onto one another is the exact mechanical feature that causes a Progressive Rate Spring to act as a Progressive Rate Spring.

In other words, as spring coils are removed from being active the fewer remaining active coils result in the spring becoming stiffer. It is the same concept as a long rod being easier to bend (softer) than a shorter rod of the same diameter (stiffer).

This is another way to identify if you truly have a progressive rate spring or not. If the coils of your springs do not progressively stack directly on top of one another (touching) during a good portion of the compression cycle of the spring you likely do not have true progressive rate springs. Or your spring has not reached its progressive rate stage within the compression cycle that you viewed.

Progressive Rate Springs have a very nice feature in that they substantially increase in rate as you compress beyond ride height. This helps to reduce the force of impact of the axles on the bump stops under extreme off road conditions. Nice feature over linear springs.

Now for the number one reason MC did not choose to design our springs as Progressive Rate Springs.

We use a lot of up travel in our suspension systems therefore our springs get very close to the fully compressed (all coils touching) condition. Progressive Rate Springs do not like to be fully compressed as they are susceptible to sagging. If you can picture an almost fully compressed Progressive Rate Spring, all of the top coils will be collapsed on one another and inactive while only the very last active coil will be active. In this condition the largest amount of spring stress is concentrated in the final active coil. Substantially more stress than what is in the first coil that collapsed and became solid long before the fully compressed condition. This large concentration of stress in a single coil can cause the spring to fatigue and sag over time.

Therefore MC chose to use Dual Rate Springs. That write up is next.
Jeepineer is offline   Reply With Quote
Unread 05-20-2013, 07:51 PM   #4
Jeepineer
Registered User
 
Join Date: Aug 2009
Location: Sacramento
Posts: 186
Dual rate springs



Dual Rate Springs (used in MetalCloak suspensions)

The diagram above shows three types of springs commonly used in the offroad industry. This post will discuss Dual Rate Springs specifically shown in action in the diagrams below. I am going to provide specific values for spring rate, load at ride height etc. for discussion purposes only. These are not actual numbers because most manufacturers have proprietary spring rates and loads that they believe are ideal for a quality ride or are tuned for their specific systems (so does MC). All numbers are examples only.

Dual Rate Springs – are just another type of progressive rate spring, however a Dual Rate Spring’s progression is from a first rate (Flex Rate) to a second rate (Ride or Road Rate) as opposed to the continually changing rate of a traditional progressive rate spring. I think very few long travel suspension manufacturers use Dual Rate Springs. Dual Rate Springs are generally more expensive to make, harder to design, and harder to inspect than linear rate springs. But Dual Rate Springs do have the similar ride height consistency of linear rate springs. Dual Rate Springs are excellent springs for long travel suspension systems.

Look at the Dual Rate Spring in the image above. Notice that the distance between the top approximateLY 4 active coils (Pitch) is smaller and identical to one another. Notice that the distance between the bottom approximately 4 active coils (Pitch) is larger and identical to one another. This is how you can identify a true Dual Rate Spring.

Because of the unique appearance of a Dual Rate Spring it is easily identifiable from a Linear Rate Spring. In other words it would be very difficult to market it as anything but a Dual Rate Spring.

Using the same spring assumptions from the previous two posts for the same example long travel suspension:

Assume the desired spring rate for the perfect ride is again 150lbs/in
Assume the compressed length for the Dual Rate Spring at ride height is again 10”
Assume that the Free Length required for the Dual Rate Spring is again 18”

So in summary we need our spring to compress a total of 8” (18” - 10”). When it compresses this 8” and reaches the desired length of 10” it needs to handle a load of 600lbs and have a rate of 150lbs/in at that specific point.

The following is a chart of what is happening with a Dual Rate Spring as it is compressed to the desired ride height.

Spring Compression___Spring Rate at Compression__Force at Compression___Length of Spring
___0_________________________0____________________ 0________________18”
___2”_____________________25 lbs/in________________50 lbs______________16”
___4”_____________________25 lbs/in________________100 lbs_____________14”
___6”_____________________25 lbs/in________________150 lbs_____________12”
___7”____________________150 lbs/in________________300 lbs_____________11”
___8”____________________150 lbs/in________________600 lbs_____________10”

So this example of a Dual Rate Spring meets the goals for the design of the example long travel suspension. In this case the transition from Flex Rate (1) to Road (Ride) Rate (2) occurs at about 7” of compression or 1” above the desired ride height.

Note that the Spring Rate is not changing as the spring compression approaches the desired ride height, only the force is changing. That makes it much easier for the manufacturer to maintain consistent ride height. I am not knocking progressive rate springs or saying they do not maintain consistent ride heights I am only saying it is more difficult to do so with progressive rate springs.



As the Dual Rate Spring in the image above compresses each coil will move the same distance relative to one another resulting in the approximately top 5 active coils collapsing on top of one another and therefore are fully supported or solid. At this point those 5 coils become inactive which facilitates the transition from the Flex Rate to the Ride Rate. In the example spring this transition occurs at approximately 7” of compression. The remaining bottom approximately 8 active coils will not collapse until the spring reaches a fully compressed condition at which time they will all touch each other at effectively the same time. In other words the bottom 8 coils will remain active throughout the rest of the compression cycle.

The fact that the bottom 8 active coils remain active throughout the full compression cycle of the spring is the main reason we chose Dual Rate Springs for the MC suspension. As stated in the previous Progressive Rate Spring Post. MC uses a lot of up travel in our suspension systems therefore our springs get very close to the fully compressed (all coils touching) condition. But even when our suspension is close to this fully compressed condition the stress in the spring is distributed equally across all of the 8 bottom active coils. In other words, we spread the heavy lifting out to multiple coils. This feature of Dual Rate Springs helps to prevent spring fatigue and sagging which MC believes is a major concern of customers.



Properly designed Dual Rate Springs and some Progressive Rate Springs (depending on design) have a unique feature that results in added spring stability and therefore added suspension stability. MC calls this feature “dropping the virtual spring bucket”. This feature is illustrated in the image above.

In order to discuss we have to agree on the following assumption.

If one compares a LONGER spring to a SHORTER spring both having the same wire diameter, same number of coils and same spring diameter, the LONGER spring is more likely to buckle and demonstrate lateral instability than the SHORTER spring.

As you can see in the image above when the Flex Rate Coils collapse they become solid and fully supported by one another. In effect…it is like the spring bucket that is retaining the top of the spring moves down to a new location or “virtual spring bucket”. This in turn results in the spring effectively becoming SHORTER and therefore more stable and less likely to buckle.

I hope this information is helpful.
Jeepineer is offline   Reply With Quote
Unread 05-20-2013, 11:15 PM   #5
chmo
Web Wheeler
2006 TJ Wrangler 
 
Join Date: Nov 2009
Location: Abu Dhabi, United Arab Emirates
Posts: 1,240
Quote:
Originally Posted by Jeepineer View Post

...

I hope this information is helpful.
YES ... very!

Thanks a LOT!
__________________
"The more I know, the more I realize how much I don't know"
my LS1 swap build thread
THE cheap progressive bump stop solution
chmo is offline   Reply With Quote
Unread 05-20-2013, 11:58 PM   #6
Oman Jeep
Web Wheeler
 
Oman Jeep's Avatar
 
Join Date: Aug 2001
Location: Dubai, United Arab Emirates
Posts: 5,133
Jeepineer - thanks for that Already ordered over a week ago but good to know more of the concept behind it
Oman Jeep is offline   Reply With Quote
Unread 05-21-2013, 04:30 AM   #7
Jerry Bransford
Do it right or not at all
 
Jerry Bransford's Avatar
2004 TJ Wrangler 
 
Join Date: Jan 2003
Location: Escondido, California, California
Posts: 60,367
Undoubtedly the best description of the various types of spring designs I have ever seen in any of the Jeep forums I read, nice work & thank you very much Jeepineer!
__________________
Getting Savvy...

Coolest offroad magazine ever! CRAWL Magazine

When you have a choice, buy American.

Jerry Bransford is online now   Reply With Quote
Unread 05-21-2013, 06:10 AM   #8
mike_dippert
hodophobic
 
mike_dippert's Avatar
2005 LJ Wrangler 
 
Join Date: Jan 2006
Posts: 3,027
Wow. I had a good comprehension of the difference, but that just made it crystal clear.
__________________
You can't argue with ignorance.
Quote:
Originally Posted by HurricaneRZA View Post
JF taught me that the 2.5L, Ax-5 and D35 together are so powerful that angels weep when I shift into 4LO.
Quote:
Originally Posted by mrblaine View Post
The only thing a bicycle inner tube is good for, is tying a knot in the end of when you run out of condoms.
mike_dippert is offline   Reply With Quote
Unread 05-21-2013, 06:55 AM   #9
TheRealZ
Registered User
 
Join Date: Mar 2013
Posts: 94
Quote:
Originally Posted by Jeepineer View Post

The new example spring rate of 75lbs/in will feel too soft to driver of the vehicle. The vehicle will role or bounce at the slightest motion.

This is the problem with Linear Rate Springs when you want to make a longer travel suspension that requires longer free length springs....the spring rate just gets too low by the time the spring compresses to the desired lifted ride height.

In other words, if we want to maintain the example perfect ride quality of 150lbs/in in a long travel suspension it becomes impossible with a Linear Rate Spring.

Therefore long travel suspension companies that want their springs to remain seated throughout the travel of the suspension and provide a ride quality at least in the same realm as the stock vehicle must use a spring other than a Linear Rate Spring.
Not necessarily true.......especially for typical on road applications where a "long travel" suspension is not going to be using much of its suspension travel to begin with.

Also, sway bar stiffness and shock valving/damping, etc are also part of the equation. Even the definition of "long travel" itself and at what length spring does a standard linear rate spring really become ineffective? When you COMBINE these together the picture can change significantly.

Also, not mentioned its coil bind length of your springs. Since part of the spring is acting as a secondary "virtual spring bucket" your in effect just adding another potential element to limit uptravel and require more bumpstopping...........in the case of the TJ/LJ this would affect the rear significantly.....possibly the front too? I would assume since the factory coil bucket out back can compress down to around 3 1/2-4"..............how tall are these springs at full bind? The way I look at it, these in essense are still "linear rate" springs for all intensive purposes........the 2nd set of coils are basically just helping a bit with axle droop/extended length when unloaded on the trails, but when under load at ride height act as a "virtual" spring bucket as you stated, which under normal conditions adds nothing(other than limiting uptravel).

Its not really as cut and dry as your making it out to be. There can be marked disadavantages to using these type of springs.

Last edited by TheRealZ; 05-21-2013 at 07:09 AM..
TheRealZ is offline   Reply With Quote
Unread 05-21-2013, 06:58 AM   #10
jeepinhokie
Web Wheeler
 
jeepinhokie's Avatar
2004 TJ Wrangler 
 
Join Date: Apr 2010
Location: MD, Maryland
Posts: 2,359
How does the fully compressed length of a dual rate spring compare to a progressive rate spring?

ie. Currie 4" spring vs the 3.5" MC spring fully compressed and fully extended?

I assume the MC spring is longer in both instances. which would leave me to assume the 4" spring has more up travel potential due to a shorter compressed length.

lets not get into a pissing match here. both companies provide top notch products to their customers. I happily run products from both companies and am contemplating the same thing the OP is.
__________________
HOKIES UNITED!

A Hokie build

check out my jeepin videos!
jeepinhokie is offline   Reply With Quote
Unread 05-21-2013, 07:10 AM   #11
Unlimited04
This post/info=my opinion
 
Unlimited04's Avatar
2004 LJ Wrangler 
 
Join Date: Dec 2005
Location: Colorado
Posts: 23,308
Good info Jeepineer.

Quote:
Originally Posted by TheRealZ View Post
Also, sway bar stiffness and shock valving/damping, etc are also part of the equation. When you COMBINE these together the picture changes.
true.

Quote:
Originally Posted by jeepinhokie View Post
I assume the MC spring is longer in both instances. which would leave me to assume the 4" spring has more up travel potential due to a shorter compressed length.
my suspicion as well. based on looking at the whole product line, MC seems to be focusing on maximizing downtravel, while sacrificing some uptravel for coil bind and longer shock lengths w/ BPE's...requiring longer bumpstop extensions. I think the other post here, showed 3" bumpstop extension w/ the 3.5" coils.

it's not wrong, it's just a different approach. I prefer a ~50/50 travel ratio for general purpose use, but it's not uncommon for the slow speed rock crawlers to maximize droop.

but still haven't seen any full bump pics with the MC track bar, 6-pak shocks or any info on coil bind lengths yet. would be helpful to compare.
Unlimited04 is offline   Reply With Quote
Unread 05-21-2013, 07:25 AM   #12
TheRealZ
Registered User
 
Join Date: Mar 2013
Posts: 94
Quote:
Originally Posted by Unlimited04 View Post

it's not wrong, it's just a different approach. I prefer a ~50/50 travel ratio for general purpose use, but it's not uncommon for the slow speed rock crawlers to maximize droop.

but still haven't seen any full bump pics with the MC track bar, 6-pak shocks or any info on coil bind lengths yet. would be helpful to compare.
Thats what I think too. Sure downtravel, is maximised.......but at a MAJOR cost to uptravel. And losing uptravel means more direct body roll and instability in many cases.

Also not sure if you want to maximise downtravel especially when using a SA setup like whats on the TJ/LJ. Your arm arc centerpoint is now well below the parallel line in the horizontal plane point from stock. So your starting point is nearly 4" below that. Having major downtravel means those axle are going to be pulled under the body of the Jeep quickly........potential for major axle steer and climbing ability and stability can be compromised, and then add in the excessive body roll you could get due to limited uptravel.........Im sure there are some advantages........but the disadvantages are quite obvious too.
TheRealZ is offline   Reply With Quote
Unread 05-21-2013, 07:30 AM   #13
WKdeuce
Registered User
1998 TJ Wrangler 
 
Join Date: Sep 2006
Location: Malvern PA
Posts: 2,203
I ordered the 3" progressive rate from Currie.
Not sure if it's that last turn being tightened that makes it that or what.
Good info Jeepineer.

__________________
2011 5.7L WK2 | 5.3L TJ work in progress
Rust in Pieces Blue
JEEP FIEND CLUB
WKdeuce is offline   Reply With Quote
Unread 05-21-2013, 01:19 PM   #14
Jeepineer
Registered User
 
Join Date: Aug 2009
Location: Sacramento
Posts: 186
Compressed length of springs

The compressed length (or solid length) of a constant coil diameter spring whether it is a linear, progressive or dual rate spring is calculated using the same equations:

If all coils have the same diameter the solid length is:


[(total number of coils) + 1] x (wire diameter)


If ONE end of the coil spring is "pig tailed" (wound smaller) the solid length is:


(total number of coils) x (wire diameter)


If BOTH ends of the coil spring is "pig tailed" (wound smaller) the solid length is:


[(total number of coils) - 1] x (wire diameter)


The above equations are a simple and handy way to check manufacturers compressed length specifications without having to fully compress the spring. As you can see the type of spring (linear, progressive or dual rate) has absolutely nothing to do with the compressed length of the spring. It is a function of total number of coils and wire diameter.

The compressed length of constant coil diameter spring is only based on the engineers design relative to the system in which the spring will be utilized.

In other words one cannot assume that a progressive spring has a shorter compressed length than a dual rate spring. It is design dependent. I will be working on a post this afternoon relative to the specific design characteristics of the MC dual rate coils as an example.

I hope this information is helpful.
Jeepineer is offline   Reply With Quote
Unread 05-21-2013, 01:35 PM   #15
jeepinhokie
Web Wheeler
 
jeepinhokie's Avatar
2004 TJ Wrangler 
 
Join Date: Apr 2010
Location: MD, Maryland
Posts: 2,359
Quote:
Originally Posted by Jeepineer View Post
In other words one cannot assume that a progressive spring has a shorter compressed length than a dual rate spring. It is design dependent. I will be working on a post this afternoon relative to the specific design characteristics of the MC dual rate coils as an example.
understood.

I was specifically curious in comparing the 3.5" MC spring to the 4" Currie springs.

i would like to keep my build in the 3.5" lift range, but am concerned about the possible lack of uptravel with the 3.5" MC springs.
__________________
HOKIES UNITED!

A Hokie build

check out my jeepin videos!
jeepinhokie is offline   Reply With Quote




Jeep, Wrangler, Cherokee, Grand Cherokee, and other models are copyrighted and trademarked to Jeep/Chrysler Corporation. JeepForum.com is not in any way associated with Jeep or the Chrysler Corp.