BMW M3 Forum (E90 E92)

BMW Garage BMW Meets Register Search Today's Posts Mark Forums Read


Go Back   M3Post - BMW M3 Forum > E90/E92 M3 Technical Topics > Suspension | Brakes | Chassis
 
Velos Designwerks
Post Reply
 
Thread Tools Search this Thread
      05-31-2008, 03:44 PM   #1
composed
Enlisted Member
 
composed's Avatar
 
Drives: E92 M3 Coupe, Alpine White
Join Date: May 2008
Location: Los Angeles

Posts: 41
iTrader: (0)

Why do some say changing springs destroys EDC shocks??

OK guys :

Let's hear the technical, intelligent answers to this! ie: Not my friend's car did this or I heard over on the M5/M6 forum that this happened. Let's talk physically and how it relates to the components involved about how lowering springs may or may not hurt or reduce the life of an EDC damper (ie: shock absorber).

My thought : The people saying damage may occur have some ignorance on the topic or if indeed dampers have had issues it's just a coincidence that it was on an EDC equipped car.

My reasoning : EDC is reading sensors in the car to automatically adjust the compression/rebound settings of the damper. The system has no effect/control upon the spring itself. In fact with lowering springs the travel of the damper is now less. Because the EDC is constantly making adjustments to the suspension of the car I assume it is reading the travel of the damper as one of its parameters in deciding what to do (the other sensors/parameters it's using are perhaps irrelevant in this discussion), and it should be able to account for whatever spring is in there within reason. I don't see how this could shorten the life of a damper within the realm of EDC - it is constantly changing and responding anyway. If a damper was designed for a specific spring rate, then the discussion seems to have merit (for example, typical aftermarket coilover setups).

Also I don't know the rates of the H&R or RD lowering springs vs. the OEM springs, can anyone comment specifically what those are? When H&R says that "the spring kit harmonises perfectly with the adjustable dampers" I assume it to mean the rates are similar.

What thoughts are out there?!
__________________


E92 M3 : Alpine White - Black Speed Cloth - Tech Package - H&R Sport Springs - Evosport 19" RG8 3-piece wheels - painted reflectors - Powerchip software - mObridge iPod integration kit - (scheduled for Evosport pulley install)
composed is offline   United_States
0
Reply With Quote
      06-01-2008, 12:11 PM   #2
GregW / Oregon
Commander-In-Chief
 
Drives: 2008 M3 Coupe, 2012 ML350
Join Date: Nov 2005
Location: Lake Oswego, OR

Posts: 7,254
iTrader: (1)

Garage List
Quote:
Originally Posted by composed View Post
OK guys :

Let's hear the technical, intelligent answers to this! ie: Not my friend's car did this or I heard over on the M5/M6 forum that this happened. Let's talk physically and how it relates to the components involved about how lowering springs may or may not hurt or reduce the life of an EDC damper (ie: shock absorber).

My thought : The people saying damage may occur have some ignorance on the topic or if indeed dampers have had issues it's just a coincidence that it was on an EDC equipped car.

My reasoning : EDC is reading sensors in the car to automatically adjust the compression/rebound settings of the damper. The system has no effect/control upon the spring itself. In fact with lowering springs the travel of the damper is now less. Because the EDC is constantly making adjustments to the suspension of the car I assume it is reading the travel of the damper as one of its parameters in deciding what to do (the other sensors/parameters it's using are perhaps irrelevant in this discussion), and it should be able to account for whatever spring is in there within reason. I don't see how this could shorten the life of a damper within the realm of EDC - it is constantly changing and responding anyway. If a damper was designed for a specific spring rate, then the discussion seems to have merit (for example, typical aftermarket coilover setups).

Also I don't know the rates of the H&R or RD lowering springs vs. the OEM springs, can anyone comment specifically what those are? When H&R says that "the spring kit harmonises perfectly with the adjustable dampers" I assume it to mean the rates are similar.

What thoughts are out there?!
Higher rate springs exert more maximum force on the dampers. In other words, the damper is working harder to control the motion. Whether there is a cause/effect relationship to failures on EDC cars with aftermarket springs I don't know. How the system deals with shorter travel and the higher rates I don't know either. It would be good to get a technical statement from H&R or some such on this subject.
__________________

Greg Lake Oswego, Oregon, USA
2008 M3 Coupe - Alpine/Fox Red/Carbon
2015 M4 coupe on order
GregW / Oregon is offline   United_States
0
Reply With Quote
      06-09-2008, 03:37 AM   #3
lucid
Major General
 
lucid's Avatar
 
Drives: E30 M3; Expedition
Join Date: Apr 2007
Location: USA

Posts: 8,034
iTrader: (0)

Quote:
Originally Posted by GregW / Oregon View Post
Higher rate springs exert more maximum force on the dampers. In other words, the damper is working harder to control the motion. Whether there is a cause/effect relationship to failures on EDC cars with aftermarket springs I don't know. How the system deals with shorter travel and the higher rates I don't know either. It would be good to get a technical statement from H&R or some such on this subject.
Actually, the spring does not exert any force on the damper at all. That would happen if the spring and the damper were to be connected in series. In the suspension, they are connected in parallel to the mass of the car. The force is exerted by gravity acting on the mass of the car, and is distributed between the spring and damper.

Meaning,

Fspring + Fdamper = Ftotal.

or,

Fs + Fd = mass of car x acceleration

where,

Fd = - damping coeff x relative velocity of the two connection points of damper

Fs = - spring rate x displacement of spring

If you were to weld the damper rigidly, the spring would not experience any force as it would not experience any displacement. Conversely, in practice, the spring will experience most of the force and compress and support the weight of the car except when the mass starts to oscillate due to suspension inputs and dynamic loading due to acceleration/deceleration. Then the damper ports will move relative to each other and the damper will experience force as well, dissipating energy and damping the oscillations in the process.

That is a mass, spring, damper system in short. For a given value of mass, and predicted inputs to the system, engineers solve the differential equation relating those 3 variables and inputs, and pick spring rate and damping coeff values to control for the displacement of the mass. Although it is true that EDC can vary the damping coefficient, it can do so within a certain range, which is dictated by the physical characteristics of the damper. If one of the system variables changes, say the spring rate, the damping coefficient will need to change to achieve the same desired mass displacement, or for EDC, vary within a different range than stock. A couple of things can go wrong:

1. The new damping coeff range might not be physically attainable depending on what the damper is capable of producing.
2. The new damping coeff range is physically attainable, but within a range that will produce higher forces on the damper than anticipated on average, resulting in premature wear.
__________________
lucid is offline   United_States
0
Reply With Quote
      06-09-2008, 11:11 AM   #4
MDCTFTW
Lieutenant Colonel
 
Drives: E92 DCT
Join Date: Apr 2008
Location: Norcal

Posts: 1,974
iTrader: (0)

Quote:
Originally Posted by lucid View Post
Actually, the spring does not exert any force on the damper at all. That would happen if the spring and the damper were to be connected in series. In the suspension, they are connected in parallel to the mass of the car. The force is exerted by gravity acting on the mass of the car, and is distributed between the spring and damper.

Meaning,

Fspring + Fdamper = Ftotal.

or,

Fs + Fd = mass of car x acceleration

where,

Fd = - damping coeff x relative velocity of the two connection points of damper

Fs = - spring rate x displacement of spring

If you were to weld the damper rigidly, the spring would not experience any force as it would not experience any displacement. Conversely, in practice, the spring will experience most of the force and compress and support the weight of the car except when the mass starts to oscillate due to suspension inputs and dynamic loading due to acceleration/deceleration. Then the damper ports will move relative to each other and the damper will experience force as well, dissipating energy and damping the oscillations in the process.

That is a mass, spring, damper system in short. For a given value of mass, and predicted inputs to the system, engineers solve the differential equation relating those 3 variables and inputs, and pick spring rate and damping coeff values to control for the displacement of the mass. Although it is true that EDC can vary the damping coefficient, it can do so within a certain range, which is dictated by the physical characteristics of the damper. If one of the system variables changes, say the spring rate, the damping coefficient will need to change to achieve the same desired mass displacement, or for EDC, vary within a different range than stock. A couple of things can go wrong:

1. The new damping coeff range might not be physically attainable depending on what the damper is capable of producing.
2. The new damping coeff range is physically attainable, but within a range that will produce higher forces on the damper than anticipated on average, resulting in premature wear.
Physics 101 ... thanks for the technical explanation. So lucid, would you swap springs and keep your stock EDC dampers?
__________________


MDCT ... so you can drive and eat a hamburger at the same time.
MDCTFTW is offline  
0
Reply With Quote
      06-09-2008, 04:27 PM   #5
MDCTFTW
Lieutenant Colonel
 
Drives: E92 DCT
Join Date: Apr 2008
Location: Norcal

Posts: 1,974
iTrader: (0)

I was thinking about this over lunch ... let me know what you think lucid.

I'm questioning whether the damping coefficient really matters or not when we swap out springs. When we swap in springs they're typically shorter in length and have a higher spring rate. By higher I'm guessing a higher spring rate means it's requires more force for the spring to travel x distance than it would for the stock spring to travel x distance.

Anyway ... say we have 2 identical spring lengths with different spring rates (stock springs A and higher spring rate B). When a spring has a higher spring rate why would that cause more wear than the lower spring rate? Say A & B drive over the same bump at the same speed resulting in the same amount of total force exerted on the system and X amount of travel for A and Y amount of travel for B. Is it safe to say that X will be greater than Y? I would have thought yes.

So if X > Y the dampers travel less for the higher spring rate. How does this cause premature wear on them? I can understand how it could if the springs are shorter and the average travel location on the dampers is different. Maybe I'm missing something.
__________________


MDCT ... so you can drive and eat a hamburger at the same time.
MDCTFTW is offline  
0
Reply With Quote
      06-09-2008, 05:30 PM   #6
gadget
Major
 
gadget's Avatar
 
Drives: 09 e92///M SSII DCT
Join Date: Nov 2007
Location: US

Posts: 1,162
iTrader: (0)

damn, you guys are smart, most of the stuff above went right over my head. But in any case, good information-
gadget is offline   United_States
0
Reply With Quote
      06-10-2008, 03:09 AM   #7
lucid
Major General
 
lucid's Avatar
 
Drives: E30 M3; Expedition
Join Date: Apr 2007
Location: USA

Posts: 8,034
iTrader: (0)

Quote:
Originally Posted by MDCTFTW View Post
I was thinking about this over lunch ... let me know what you think lucid.

I'm questioning whether the damping coefficient really matters or not when we swap out springs. When we swap in springs they're typically shorter in length and have a higher spring rate. By higher I'm guessing a higher spring rate means it's requires more force for the spring to travel x distance than it would for the stock spring to travel x distance.

Anyway ... say we have 2 identical spring lengths with different spring rates (stock springs A and higher spring rate B). When a spring has a higher spring rate why would that cause more wear than the lower spring rate? Say A & B drive over the same bump at the same speed resulting in the same amount of total force exerted on the system and X amount of travel for A and Y amount of travel for B. Is it safe to say that X will be greater than Y? I would have thought yes.

So if X > Y the dampers travel less for the higher spring rate. How does this cause premature wear on them? I can understand how it could if the springs are shorter and the average travel location on the dampers is different. Maybe I'm missing something.
The transient response of the system is a function of both k (spring rate) and c (damping coeff). Just to clarify, the transient response is how the system would react to an input after it has reached a steady state. So, in the scenario you described, where spring rate kB > is kA, spring A will compress more and than spring B when the springs are installed and the suspension is loaded, and car A will sit lower than car B. That is the steady state. Then you hit the same bump at the same manner with both cars. The transient displacement response of the suspension, how much more the suspension will move, will not only depend on k, but also on c at that point. To understand why that is you need to solve the differential equation I set up above (you substitute -kx for the spring force and -cv for the damper force, and then express acceleration and velocity as time derivates of distance). The nature of the solution (real or complex) depends on a ratio that is termed "the damping ratio", which is:

damping ratio = c / (2 sqrt(k x m))

another parameter is the natural frequency, w0, which is:

w0 = sqrt(k x m)

If the damping ratio is 0, then the system is critically damped, and will return to a steady state the soonest without oscillating. If it is > 0, the system is over-damped, and will take longer to return to steady state (how much longer depends on how overdamped it is), but will still not oscillate. If it is < 0, the system is under-damped, and will overshoot (the spring would compress more than a critically damped system) and oscillate until it returns to steady state at its damped frequency (a function of wo and the damping ratio), but not as fast as a critically damped system.

So, to answer to your question about X and Y, one would need to know not only kA and kB, but also cA and cB. In other words, if car B is under-damped, and car A is critically damped, it is "possible" for Y < X despite kb > kA (car B having stiffer springs).

Now, coming back to why replacing stock springs with aftermarket springs that have a higher k might end up resulting in higher damper forces. If you look at the damping ratio equation above, you will see that if the EDC wants to maintain a specific damping ratio in a given situation--say, it wants critical damping and therefore a damping ratio of 1--and if k is now higher because of the aftermarket springs, then c needs to be higher as well.

Now, we know that Fd = - c x v, and we also know the c of the aftermarket system will be higher. Then, the question is if the velocity of the aftermarket system is higher, equal or lower than the velocity of the stock system. To find that out you can differentiate the displacement response of the critically damped solution, but there might be an easier way. I think you can compare the natural frequencies of the aftermarket and stock systems since that solution is a function of the w0 and time only. So, if the aftermarket system has a higher w0, and if both systems are critically damped, the aftermarket system should experience higher velocities, and therefore, higher damper forces. And since higher k will result in higher w0, that seems to be the case.

I don't do this for a living (not my area of expertise), and am thinking through what I learned in a basics dynamics course 15 years ago. So, you or anyone else is welcome to chime in and critique my thinking on this.

Also, this is a simplified "model" of what really happens. In reality, there are 4 wheels experiencing different inputs but are connected to the same mass, the wheels and suspension parts have mass, stiffness and damping, and inputs vary in size and duration. So, to figure out a real situation, you need to turn to people like Swamp who codes computer simulations, and use those tools. Even then, simulations go so far and sometimes do not explain what needs to be explained.
__________________
lucid is offline   United_States
0
Reply With Quote
      06-10-2008, 03:37 AM   #8
lucid
Major General
 
lucid's Avatar
 
Drives: E30 M3; Expedition
Join Date: Apr 2007
Location: USA

Posts: 8,034
iTrader: (0)

Quote:
Originally Posted by MDCTFTW View Post
So lucid, would you swap springs and keep your stock EDC dampers?
I guess that depends on the motivation. 3 cases:

1. Want better performance than stock on the street. Can't do it. Can't beat M engineers by swapping springs and shocks.

2. Want better performance at the track. Possible since the car, despite being performance oriented is not optimized for the track. However, I personally do not want to commute in a car optimized for the track.

3. Lowered look. I would like to see the car about an inch lower, but I don't yet feel comfortable doing this. I would like to understand why the car is riding as high as it does stock. There might be flow issues. Clearence does not seem to be issue as the 335 seems to ride lower (although I have not confirmed this for sure).
__________________
lucid is offline   United_States
0
Reply With Quote
      06-10-2008, 10:40 AM   #9
Gearhead999s
Major General
 
Gearhead999s's Avatar
 
Drives: E93 335is(wifes)E92 M3
Join Date: Apr 2007
Location: Toronto

Posts: 6,677
iTrader: (0)

Quote:
Originally Posted by lucid View Post
I guess that depends on the motivation. 3 cases:

1. Want better performance than stock on the street. Can't do it. Can't beat M engineers by swapping springs and shocks.

2. Want better performance at the track. Possible since the car, despite being performance oriented is not optimized for the track. However, I personally do not want to commute in a car optimized for the track.

3. Lowered look. I would like to see the car about an inch lower, but I don't yet feel comfortable doing this. I would like to understand why the car is riding as high as it does stock. There might be flow issues. Clearence does not seem to be issue as the 335 seems to ride lower (although I have not confirmed this for sure).
My wifes 335 Cab sits quite a bit lower than my M3 and it is stock with the sport suspension and 19's as delivered.I also do not understand why the M3 has a higher ride height.We do have issues with the 335 touching down the odd time and it is way more an issue in the winter.
Gearhead999s is offline  
0
Reply With Quote
      06-10-2008, 04:51 PM   #10
lucid
Major General
 
lucid's Avatar
 
Drives: E30 M3; Expedition
Join Date: Apr 2007
Location: USA

Posts: 8,034
iTrader: (0)

Quote:
Originally Posted by Gearhead999s View Post
My wifes 335 Cab sits quite a bit lower than my M3 and it is stock with the sport suspension and 19's as delivered.I also do not understand why the M3 has a higher ride height.We do have issues with the 335 touching down the odd time and it is way more an issue in the winter.
If you get a chance, can you measure the chassis ride height difference between the 335 and the M3? Thanks.
__________________
lucid is offline   United_States
0
Reply With Quote
      06-10-2008, 07:27 PM   #11
MDCTFTW
Lieutenant Colonel
 
Drives: E92 DCT
Join Date: Apr 2008
Location: Norcal

Posts: 1,974
iTrader: (0)

Thanks for the post lucid. I'll need to jot some of that down on paper and ponder some more. I do agree that the M sits a tad high for my taste. I originally thought it was perfect, but I was walking to my car the other day and said "could be a tad bit lower".
__________________


MDCT ... so you can drive and eat a hamburger at the same time.
MDCTFTW is offline  
0
Reply With Quote
      06-10-2008, 11:54 PM   #12
cosmos515
Second Lieutenant
 
cosmos515's Avatar
 
Drives: I find what I keeps
Join Date: Feb 2008
Location: So Cal

Posts: 215
iTrader: (0)

The M3 rides higher because of two reasons.

1) The front diffuser is lower to the ground and BMW/NA requires a certian height for curbs, since the front diffuser is lower and the curbs are the same height the car needed to be raised a bit.

2) The cars are also a certian height not for ulitmate shock travel or for some other suspension issue (control arms being in a neutral position) but for the simple purpose that the DOT requires headlights to be a certian height level as well.

So if you want to lower it 1" or so, it will be fine, anything more than that, you simply change too much in the engineering of the ssupension.
__________________
"Motion does not equal progress"
"I love the smell of cosmoline in the morning"
cosmos515 is offline  
0
Reply With Quote
      06-11-2008, 07:59 AM   #13
Gearhead999s
Major General
 
Gearhead999s's Avatar
 
Drives: E93 335is(wifes)E92 M3
Join Date: Apr 2007
Location: Toronto

Posts: 6,677
iTrader: (0)

Quote:
Originally Posted by lucid View Post
If you get a chance, can you measure the chassis ride height difference between the 335 and the M3? Thanks.
I should be able to do that tonight.
Gearhead999s is offline  
0
Reply With Quote
      06-11-2008, 08:57 AM   #14
booforty
Major General
 
booforty's Avatar
 
Drives: E92 M3
Join Date: Mar 2005
Location: London, UK

Posts: 8,813
iTrader: (0)

Garage List
2008 E92 M3  [0.00]
Quote:
Originally Posted by cosmos515 View Post
The M3 rides higher because of two reasons.

1) The front diffuser is lower to the ground and BMW/NA requires a certian height for curbs, since the front diffuser is lower and the curbs are the same height the car needed to be raised a bit.

2) The cars are also a certian height not for ulitmate shock travel or for some other suspension issue (control arms being in a neutral position) but for the simple purpose that the DOT requires headlights to be a certian height level as well.

So if you want to lower it 1" or so, it will be fine, anything more than that, you simply change too much in the engineering of the ssupension.
I'll measure the ground clearance of my Euro M3 too as from photos the Euro M3 looks lower in stock set-up than the US M3. Maybe this is due to US regulations?
booforty is offline   United Kingdom
0
Reply With Quote
      06-15-2008, 11:20 AM   #15
jm1234
Lieutenant
 
jm1234's Avatar
 
Drives: E90 M3 Jet Black/Tan/SA 6MT
Join Date: Mar 2008
Location: Alpharetta, GA

Posts: 431
iTrader: (0)

Quote:
Originally Posted by lucid View Post
A lot of good stuff...
I'd guess that the EDC in "normal" mode is a closed loop control system and thus the damping ratio is the controlling variable and not a constant. In comfort mode I would assume a slightly underdamped system and in race a critically damped system (or perhaps even slightly overdamped). In normal mode though it's a question of what is being controlled? Is it ride height at each wheel (to control compression/extension); is it roll/pitch frequency/rate; is it lateral/longitudinal acceleration; is it a combination; or something else entirely? Any portion of the control system based on ride height is going to be negatively impacted by changing the ride height. The control system will see a constant downward offset which could (just guessing here) result in a constantly overdamped system. Portions of the control system based on roll/pitch/acceleration should incorporate spring rates and by changing the spring rates you will impact the system response but how would be more difficult to predict.

I don't see how you can take a controlled system and swap parts to change the response values and not negatively impact the performance. It's comparable to open/restricting the intake and exhaust on the engine but not changing the computer and fuel maps. The car might "look" cooler with lowered springs but unless you change the dampers; control system; and/or keep the car in a static suspension setting (comfort or race) then I think you'll suffer a variably compromised ride with difficult to predict consequences. I'm not predicting dire things just crap suspension feel (too stiff, too soft, bouncing, bottoming out, excessive roll, etc...) some portion of the time while in "normal" mode.

Does anyone know the details of how BMW's adjustable dampers work? I'm still curious and some of you seem to have a lot of engineering knowledge about the car.
jm1234 is offline  
0
Reply With Quote
      06-15-2008, 01:08 PM   #16
OC ///M
Dr. Octagon
 
OC ///M's Avatar
 
Drives: Moonstone e92 M3
Join Date: Feb 2007
Location: So Cal

Posts: 1,811
iTrader: (0)

Quote:
Originally Posted by jm1234 View Post
I'd guess that the EDC in "normal" mode is a closed loop control system and thus the damping ratio is the controlling variable and not a constant. In comfort mode I would assume a slightly underdamped system and in race a critically damped system (or perhaps even slightly overdamped). In normal mode though it's a question of what is being controlled? Is it ride height at each wheel (to control compression/extension); is it roll/pitch frequency/rate; is it lateral/longitudinal acceleration; is it a combination; or something else entirely? Any portion of the control system based on ride height is going to be negatively impacted by changing the ride height. The control system will see a constant downward offset which could (just guessing here) result in a constantly overdamped system. Portions of the control system based on roll/pitch/acceleration should incorporate spring rates and by changing the spring rates you will impact the system response but how would be more difficult to predict.

I don't see how you can take a controlled system and swap parts to change the response values and not negatively impact the performance. It's comparable to open/restricting the intake and exhaust on the engine but not changing the computer and fuel maps. The car might "look" cooler with lowered springs but unless you change the dampers; control system; and/or keep the car in a static suspension setting (comfort or race) then I think you'll suffer a variably compromised ride with difficult to predict consequences. I'm not predicting dire things just crap suspension feel (too stiff, too soft, bouncing, bottoming out, excessive roll, etc...) some portion of the time while in "normal" mode.

Does anyone know the details of how BMW's adjustable dampers work? I'm still curious and some of you seem to have a lot of engineering knowledge about the car.
jm:

I think the questions posed in your first paragraph are important ones. I do know that piezoelectric technology incorporated into shock absorbers is not new, there must be something published on this. Hopefully someone can enlighten us. Just to confirm though, from what I understand, comfort is adaptive as well...Can someone validate this?
__________________
OC ///M

2013 E92 M3 Individual
Sold: 2011 E92 M3 SSII | Sold: 2008 E92 M3 SSII | Into a Wall: 2008 E92 M3 SSII | Sold: 2007 E92 335i JB
OC ///M is offline   United_States
0
Reply With Quote
      06-15-2008, 01:29 PM   #17
GregW / Oregon
Commander-In-Chief
 
Drives: 2008 M3 Coupe, 2012 ML350
Join Date: Nov 2005
Location: Lake Oswego, OR

Posts: 7,254
iTrader: (1)

Garage List
EDC

Quote:
Originally Posted by jm1234 View Post
I'd guess that the EDC in "normal" mode is a closed loop control system and thus the damping ratio is the controlling variable and not a constant. In comfort mode I would assume a slightly underdamped system and in race a critically damped system (or perhaps even slightly overdamped). In normal mode though it's a question of what is being controlled? Is it ride height at each wheel (to control compression/extension); is it roll/pitch frequency/rate; is it lateral/longitudinal acceleration; is it a combination; or something else entirely? Any portion of the control system based on ride height is going to be negatively impacted by changing the ride height. The control system will see a constant downward offset which could (just guessing here) result in a constantly overdamped system. Portions of the control system based on roll/pitch/acceleration should incorporate spring rates and by changing the spring rates you will impact the system response but how would be more difficult to predict.

I don't see how you can take a controlled system and swap parts to change the response values and not negatively impact the performance. It's comparable to open/restricting the intake and exhaust on the engine but not changing the computer and fuel maps. The car might "look" cooler with lowered springs but unless you change the dampers; control system; and/or keep the car in a static suspension setting (comfort or race) then I think you'll suffer a variably compromised ride with difficult to predict consequences. I'm not predicting dire things just crap suspension feel (too stiff, too soft, bouncing, bottoming out, excessive roll, etc...) some portion of the time while in "normal" mode.

Does anyone know the details of how BMW's adjustable dampers work? I'm still curious and some of you seem to have a lot of engineering knowledge about the car.

"EDC-K is available for the first time in the E92 M3. EDC-K is an option and is based on the EDC-K in the E65.

Both dampers of one axis are always activated in parallel. The valve is installed internally in the damper in the damper oil system.

The driver can choose between three settings, the controlled programs "Comfort" and "Normal", or the uncontrolled fixed setting
"Sport". The program is selected using the EDC-K button on the center console or preset via the MDrive menu and activated using the M button on the steering wheel.

The input signals come from two vertical acceleration sensors in the front wheel arches and a third sensor in the rear right-hand wheel arch.
The steering column switch cluster sends the steering angle to the F-CAN. This is transmitted together with the wheel speeds from the DSC to the PT-CAN and evaluated in the EDC-K control unit.

The longitudinal, lateral and vertical accelerations calculated as a result are used as a basis for regulation. The EDC-K button signal enters the junction
box and is transmitted to the EDC-K on the PT-CAN." -- BMW Aftersales Training - Product Information

The only thing that is controlled is the shock damping, both compression and rebound. The following graph illustrates the damping force on the vertical axis "A" (compression below & rebound above) related to the damper piston speed on the horizontal axis "B". Controlling ride height would take some sort of air suspension or such. Also, the M3 does not have active anti-roll bars like the Range Rover and some other vehicles. Still, it is quite effective.
Attached Images
  
__________________

Greg Lake Oswego, Oregon, USA
2008 M3 Coupe - Alpine/Fox Red/Carbon
2015 M4 coupe on order
GregW / Oregon is offline   United_States
0
Reply With Quote
      06-15-2008, 02:57 PM   #18
swamp2
Major General
 
swamp2's Avatar
 
Drives: E92 M3
Join Date: Sep 2006
Location: San Diego, CA USA

Posts: 9,939
iTrader: (1)

Quote:
Originally Posted by jm1234 View Post
Any portion of the control system based on ride height is going to be negatively impacted by changing the ride height. The control system will see a constant downward offset which could (just guessing here) result in a constantly overdamped system.
I disagree with this statement. Well not the stament itself but the ASSUMPTION. The sensors used for the primary inputs for EDC are the three accelerometers, two front and one rear. Typically acceleromters are good at measuring ACCELERATION. Of course you can numerically integrate that signal and get velocity as well, if you have great equipment and/or DSP stuff and are very careful you can even get position information from an accel but it is tough. What is the point? The system is very likely position independent, i.e. meaning that the damping changes are based on instantaneous accelerations (or perhaps including some recent buffer/time history of what the suspension has been doing "recently"), nothing else. So the offset of the damper travel caused by a lowering spring should be fairly transparent to the control system.

A small thought experiment can help convince you that this is correct. What about ride height variation caused by max/min load cases (i.e. fuel and 1 vs. 3 or 4 passengers and possibly cargo). There will absolutely be a large noticeable change is ride height based on min/max loads and guess what EDC system will work just fine with a damper "offset". It has to.

BMW may or may not tell us the truth on this topic. As well the spring makers like to do BS like call springs "EDC compliant". I have no solid proof but strongly suspect a spring that would work and feel reasonable on the car (i.e. slightly shorter and somewhat stiffer) will very likely not adversely affect the EDC systems function nor longevity.

On the ride height issue: I am pretty pleased with with the cars ride height and looks. The rear wheels (19's) fill the wells beautifully and perfectly IMO. The front look slightly less "right" to me but definitely not nearly enough of an issue for me to muck with it, decrease ride quality, increase scraping and generally make the car more track like and less daily driver like.
swamp2 is offline   United_States
0
Reply With Quote
      06-15-2008, 10:52 PM   #19
jm1234
Lieutenant
 
jm1234's Avatar
 
Drives: E90 M3 Jet Black/Tan/SA 6MT
Join Date: Mar 2008
Location: Alpharetta, GA

Posts: 431
iTrader: (0)

Quote:
Originally Posted by swamp2 View Post
I disagree with this statement. Well not the stament itself but the ASSUMPTION. The sensors used for the primary inputs for EDC are the three accelerometers, two front and one rear. Typically acceleromters are good at measuring ACCELERATION. Of course you can numerically integrate that signal and get velocity as well, if you have great equipment and/or DSP stuff and are very careful you can even get position information from an accel but it is tough. What is the point? The system is very likely position independent, i.e. meaning that the damping changes are based on instantaneous accelerations (or perhaps including some recent buffer/time history of what the suspension has been doing "recently"), nothing else. So the offset of the damper travel caused by a lowering spring should be fairly transparent to the control system.
Agreed, but not for the reason stated. Getting position offset from an accelerometer is trivial. Speed and distance sensors for running are based on foot mounted accelerometers and give fairly accurate results in a very complex environment (running feet as opposed to a rolling chassis). Dynamic position offset is not the same as static position offset and accelerometers wouldn't be up to detecting static offset and thus wouldn't respond to lowered springs.

Quote:
Originally Posted by swamp2 View Post
A small thought experiment can help convince you that this is correct. What about ride height variation caused by max/min load cases (i.e. fuel and 1 vs. 3 or 4 passengers and possibly cargo). There will absolutely be a large noticeable change is ride height based on min/max loads and guess what EDC system will work just fine with a damper "offset". It has to.

BMW may or may not tell us the truth on this topic. As well the spring makers like to do BS like call springs "EDC compliant". I have no solid proof but strongly suspect a spring that would work and feel reasonable on the car (i.e. slightly shorter and somewhat stiffer) will very likely not adversely affect the EDC systems function nor longevity.

On the ride height issue: I am pretty pleased with with the cars ride height and looks. The rear wheels (19's) fill the wells beautifully and perfectly IMO. The front look slightly less "right" to me but definitely not nearly enough of an issue for me to muck with it, decrease ride quality, increase scraping and generally make the car more track like and less daily driver like.
Bear with me here, I might have something wrong. The point of the suspension is to keep the car level (with respect to the road for ideal tire contact) yet still allow the tires to track the pavement (not too stiff, so the tires do not hop and/or the passengers don't spill their coffee). Controlled suspensions do this by adjusting damping as little as possible to achieve a level ride. If I have this right so far then to suggest this task can be done dynamically without knowing the spring rate isn't plausible. Spring rate, distance of spring from the c.o.g. and lateral rotational inertia are the three key system variables and changing any will cause the control system algorithms to be wrong. Those algorithms are undoubtedly based on a system model containing little more than those three system variables. The system would likely be wrong by a little bit for small changes in the spring rate but wrong is wrong and will have a negative impact on ride as considered "ideal" by BMW. There are no objective measures so we can't say how wrong or if it's detectably wrong. An owner might very well like the ride better and even drive faster with the suspension but that doesn't mean the EDC system is operating as designed. If the EDC works by changing duty cycles then new springs could conceivably wear out the system early. Conceivably isn't the same as definitely but so far no one here has posted any knowledge of how the dampers work. Somehow the normal setting results in damping that isn't the same as either sport (can be stiffer) or comfort. I would say that there is more than binary variability and thus the possibility that the system cycles (to effectively emulate multiple settings) or contains multiple settings. If it is switching damping rates more often because the spring rates changed the dynamics then that impacts long term reliability. Again no comment on the degree of impact, just that it exists. I'm not trying to scare anyone into not lowering their car. Just realize that it's an expensive (just guessing here) closed loop system and you are changing the variables and the warranty isn't going to cover any damage. Personally, I think that for small changes in the spring rate you'll be fine and not have a noticeable change in the life of the system or ride. But, I acknowledge that my opinion is not based on any real knowledge of the system and I've done enough controls system design to know it's difficult to predict what might happen when you don't know the details of the controlled parameters (three accelerometers positioned how, measuring what), the control surfaces (how do the dampers work) or the control algorithm (PID, sliding mode, fuzzy logic, yada, yada, yada).
jm1234 is offline  
0
Reply With Quote
      06-16-2008, 01:13 PM   #20
swamp2
Major General
 
swamp2's Avatar
 
Drives: E92 M3
Join Date: Sep 2006
Location: San Diego, CA USA

Posts: 9,939
iTrader: (1)

Quote:
Originally Posted by jm1234 View Post
Agreed, but not for the reason stated. Getting position offset from an accelerometer is trivial. Speed and distance sensors for running are based on foot mounted accelerometers and give fairly accurate results in a very complex environment (running feet as opposed to a rolling chassis). Dynamic position offset is not the same as static position offset and accelerometers wouldn't be up to detecting static offset and thus wouldn't respond to lowered springs.



Bear with me here, I might have something wrong. The point of the suspension is to keep the car level (with respect to the road for ideal tire contact) yet still allow the tires to track the pavement (not too stiff, so the tires do not hop and/or the passengers don't spill their coffee). Controlled suspensions do this by adjusting damping as little as possible to achieve a level ride. If I have this right so far then to suggest this task can be done dynamically without knowing the spring rate isn't plausible. Spring rate, distance of spring from the c.o.g. and lateral rotational inertia are the three key system variables and changing any will cause the control system algorithms to be wrong. Those algorithms are undoubtedly based on a system model containing little more than those three system variables. The system would likely be wrong by a little bit for small changes in the spring rate but wrong is wrong and will have a negative impact on ride as considered "ideal" by BMW. There are no objective measures so we can't say how wrong or if it's detectably wrong. An owner might very well like the ride better and even drive faster with the suspension but that doesn't mean the EDC system is operating as designed. If the EDC works by changing duty cycles then new springs could conceivably wear out the system early. Conceivably isn't the same as definitely but so far no one here has posted any knowledge of how the dampers work. Somehow the normal setting results in damping that isn't the same as either sport (can be stiffer) or comfort. I would say that there is more than binary variability and thus the possibility that the system cycles (to effectively emulate multiple settings) or contains multiple settings. If it is switching damping rates more often because the spring rates changed the dynamics then that impacts long term reliability. Again no comment on the degree of impact, just that it exists. I'm not trying to scare anyone into not lowering their car. Just realize that it's an expensive (just guessing here) closed loop system and you are changing the variables and the warranty isn't going to cover any damage. Personally, I think that for small changes in the spring rate you'll be fine and not have a noticeable change in the life of the system or ride. But, I acknowledge that my opinion is not based on any real knowledge of the system and I've done enough controls system design to know it's difficult to predict what might happen when you don't know the details of the controlled parameters (three accelerometers positioned how, measuring what), the control surfaces (how do the dampers work) or the control algorithm (PID, sliding mode, fuzzy logic, yada, yada, yada).
Running accels work based on the periodicity of an acceleration signal not based on position. I stick to my claim that is it tough and inaccurate to get absolute position from an accel. Also think you are confusing yourself (and me) with terms like "dynamic position offset". The damper has an initial static set point and the accelerometer can not really measure that. Once it oscillates you can integrate acceleration to get velocity and position but it is not very accurate. I've done it, have you?

Your second paragraph indicated some serious misconceptions about how a suspension system works in term of the goals and effects of both the spring and damper (be it a static or dynamic damper). Yes the EDC system is probably "aware of" the spring constant, but the system is going to be robust to changes in spring rate, damping and ride height, virtually guaranteed.
swamp2 is offline   United_States
0
Reply With Quote
      06-16-2008, 07:28 PM   #21
jm1234
Lieutenant
 
jm1234's Avatar
 
Drives: E90 M3 Jet Black/Tan/SA 6MT
Join Date: Mar 2008
Location: Alpharetta, GA

Posts: 431
iTrader: (0)

Quote:
Originally Posted by swamp2 View Post
Yes the EDC system is probably "aware of" the spring constant, but the system is going to be robust to changes in spring rate, damping and ride height, virtually guaranteed.
Perhaps I should be more clear, I think it is fair and valid to conclude that the "system is going to be robust to changes in spring rate" (it is actually guaranteed because even stock spring have some variability) but you do so without attempting to put bounds on the statement. I'd bet the N54 engine is robust to changes in boost pressure too. Surely you don't believe that was a system design criteria and/or that any spring rate or type will work just fine? It's a question of degrees and nobody seems to want to put any numbers to any of the changes discussed.

The conclusion that the system is robust to changes in damping when damping is the controlling variable is nonsensical. The damping rate is clearly designed to change and to do so over a much greater range than a traditional damper so the comment doesn't make sense when placed in the same context as the spring rate which is not designed to change.

I wish you luck with your modification and if you say that the specific springs installed do not ruin the ride I'll take your word for it. I still think that the system will not take any spring rate and that some spring rates will cause a crappy ride regardless of the system. I also think that the effect on damper life is an open question as no one can describe any of the pertinent design elements of the system (like how the dampers or control systems work) let alone analyze their interaction.

By dynamic system offset I'm simply referring to the offset over a known period of time. In other words, the system can calculate with a certain degree of error and over a fixed period of time what the relative (from the start of the time period) position offset is based on accelerometer readings.

By static system offset I'm referring to a measurement of the system offset with a certain degree of accuracy without regard to time. Basically you wouldn't be able to do this with an accelerometer because the buildup of position error over time would overwhelm the position measurement.
jm1234 is offline  
0
Reply With Quote
      06-17-2008, 12:11 AM   #22
OC ///M
Dr. Octagon
 
OC ///M's Avatar
 
Drives: Moonstone e92 M3
Join Date: Feb 2007
Location: So Cal

Posts: 1,811
iTrader: (0)

After almost 5k miles on RD Sport - No noticeable degradation of shock functionality.

subscribed.

Perhaps we can get an inquiry by email to an engineer at Racing Dynamics, H&R, or even Bilstein and have the responses to jm's questions copied to this thread?
__________________
OC ///M

2013 E92 M3 Individual
Sold: 2011 E92 M3 SSII | Sold: 2008 E92 M3 SSII | Into a Wall: 2008 E92 M3 SSII | Sold: 2007 E92 335i JB
OC ///M is offline   United_States
0
Reply With Quote
Post Reply

Bookmarks

Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 06:04 PM.




m3post
Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
1Addicts.com, BIMMERPOST.com, E90Post.com, F30Post.com, M3Post.com, ZPost.com, 5Post.com, 6Post.com, 7Post.com, XBimmers.com logo and trademark are properties of BIMMERPOST