Quantcast

suspension guru's: need help with final year project!

dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
hey guys, i need some help in finding information about dampers which are available today.

i am only in my 3rd year now, but have to submit a proposal for my final year masters project in about a month. as opposed to the normal aerospace research projects which are a bit boring unless you get the F1 one which all the hard core geeks go for, i want to do something bike related which i will have an actual interest in.

well, my proposed project is to have an electronically controlled damper of sorts. using electro-hydraulic servo valves to provide the damping and a position sensor on the piston shaft, then a standard PID controller to regulate the valves. the lay out i am thinking about is similar to a double barrel, so needing two EHSVs. making the damper having only favourable pressure gradients.

on paper it looks inviting as it will have a few advantages, it can be tuned with the P & D terms to make it position and speed sensitive. the only other damper which does that which i am aware of is the curnut control valve thing, but requires cooling for heavier applications. there is more freedom to play around with damping curves, adding lead or lag terms in the controller ect. and also having the advantage of having everything external, not needing to open the damper up for a different tune (of course as long as it lies with in the limitations of the EHSVs)

those advantages there are just pulled out of my ass really, and the reality of it could be very very different.

ideally you can refine it enough and make is small enough to fit it on a bike, but that's for later. right now i have to convince my lecturers that this is a project worth pursuing.


anyway, the main thing is that my lecturer needs to know how novel this idea is and what current research is done in this area which is or might be close to what i am proposing. i my self dont know a great deal of what is going on outside the world of bikes really, i know cars have some pretty smart systems but know nothing about them.

any help will be greatly appreciated!

Thanks,
Adam.
 
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
supafly - yea, steve (socket) told me about the cannondale set up. sounded like it was more of electronics just adjusting a conventional compression circuit. the noleen shock is pretty much spot on, thanks for that!

starting to wonder if there is any point to doing this as a final year project if its been done already.
 
ohio

ohio

The Fresno Kid
Nov 26, 2001
6,649
24
SF, CA
Noleen is a good place to look, but just because they tried it doesn't mean it's the only way to apply the tech.

I'd actually look to automotive... there are several vehicles that have adaptive damping. Rule out anything from GM, as they're using a ferro-magnetic fluid, but high-end imports (generally more luxury than sport) may offer some inspiration.
 
L

LMC

Monkey
Dec 10, 2006
683
1
biandon said:
though not implemented in the manner in which you describe, here's a system with similar functions and plenty of benefits. Bose, project sound.

http://www.bose.com/controller?event=VIEW_STATIC_PAGE_EVENT&url=/learning/project_sound/bose_suspension.jsp
Click to expand...
yeah thats the route i would go down dhkid. there was a video of the BOSE system posted here last year, the car remained pretty much level through a series of stutter bumps, round cones ect..

EDIT: google k2 smartshock for info about the noleen shock/fork
 
Last edited:
D

djamgils

Monkey
Aug 31, 2007
349
0
Holland
from what I understand dhkid doesnt want active suspension(really actuate the system, needs high energy levels) but just electronics to create a certain speed/position sensitive damping curve. From my point of view that would give semi-active suspension (needs only low energy levels to adjust things.
The bose system would be a active suspension from that point of view, and for dh I think you cant take big energy sources with you.

dhkid, do you really think there is a performance advantage possible over a custom tuned shock with a custom leverage ratio? the advantage of electronics would be adjustability over a mechanical linkage.
how do you see your system working, would you need a battery pack and actuators to change the shock curve during riding? you only want to make it shaft position sensitive or also keep track of your position on the course to anticipate for jumps/drops or rocks?

A classmate of mine remembers something about Jaguar that uses shocks that have motors attached to the adjuster knobs so it could adjust the damper while driving.
I guess you would need pretty high frequency adjusters to be able to make position sensitive adjustments.

Here is some short info on Mercedes Active Body Control
http://img83.imageshack.us/img83/7013/mercabc.jpg
 
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
biandon & LMC -like djamgils said, a bose system would be way too complicated.

djamgils - i dont see it as trying to replace a conventional well tuned shock. even if by some miracle i can get it to perform somehow like a shock, its complexity would rule it out from being on a bike. its purely from a research point of view, trying to make an idea work, and learning something from it. i guess if you are looking for an application, it could be used as to find a 'base line' tune. in theory, it will be more tunable then a conventional shock with everything being external and in the software.

when i said position sensitive, i mean the shock, the damping in relation with the shaft position. like what a 5th (nearly fully position sensitive if you ignore the lsc and hsc adjusters on the piggy back) and dhx is. you can take shaft position data, use it as your speed input and position. speed will just be the derivative of the position and the damping can be tuned to react accordingly to both inputs. either that or use a pressure sensor to handle the speed sensitive side of things.

i have talked to steve (socket on here) about it briefly, its definitely is going to be a challenge as getting the valves to open up initially from closed on a high speed hit will be a nightmare. so most likely the valves will be slightly open to start with or a shim stack to handle the very low oil volumes, then the EHSVs handle the higher oil flows.
 
D

dw

Wiffle Ball ninja
Sep 10, 2001
2,943
0
MV
Just doing 1/10th of the math that you need to characterize the input / output of a servo hydraulic system is going to be a massive undertaking. I'd suggest looking in detail at a smaller part of the system. It's so easy to bite off more than you can chew in a project like this, I see students do it all the time. I did it myself many times before I learned my lesson. Concentrate on one area and do a really through job of it. I bet that your professors will recognize the maturity of an approach like this, and if you do a great job, I'm sure your grades will be your reward.

In my senior design I developed an entire water package for our mini baja East car. It was extensive, built a water tunnel for testing float shapes, designed "scrapers" to increase thrust, developed tire tread patterns, and a load of other stuff. We did a ton of calculations and backed them up with testing. We got SECOND PLACE. We were bent.

The guy that won did a theoretical analysis of an anti-matter injection nozzle.

Case in point.
 
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
cheers for the input dw, am worried about that too. the project is only 6 months long and worth a 1/3rd of final year marks.

i am really just trying my best to stay away from the standard projects that are given out, but seems like i might have to.
 
D

djamgils

Monkey
Aug 31, 2007
349
0
Holland
its purely from a research point of view, trying to make an idea work, and learning something from it. i guess if you are looking for an application, it could be used as to find a 'base line' tune. in theory, it will be more tunable then a conventional shock with everything being external and in the software.
Click to expand...
Being a scientist is funn, you are allowed to do "useless" research. but yeah it might be usefull to have a shock with a wide range of easy to do adjustment. But on the other hand people already know what a good effective wheel damping would be and you can calculate the desired shock curve trough the leverage ratio.
with your idea you could decouple the position sensitivity of the spring and damper. So that the effective spring force is governed by the leverage curve and the effective damper force is changed trough its travel by the controller.
 
FCLinder

FCLinder

Turbo Monkey
Mar 6, 2002
4,402
0
Greenville, South Carolina
If you do go through with it, make sure to use the motion of the shock piston as your energy source to run whatever electrons that are involved. I will tell you trying to get circuit boards to run off limited power is a project in its own. Think of energy transfer. You know like those Kids shoes that light up every time they put down their feet on the ground. Same principles.

Good luck as I think this is the next evolution for coil over shocks among a few other Ideas I know of,

Cecil
 
Last edited:
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
^ in an ideal world, yea. but i have doubts that i can get it to work that well to the point that you can just bolt it on to a bike, play around on a computer for a bit, get it programmed and ride of into the sunset and live happily ever after. :)

dw said i might be biting off more then i can chew, and i'll take his word for it. on paper it seems all quite nice a simple, and the controller part of the project doesn't scare me too much, i have worked on them a fair bit. our last control lab was getting a helicopter to take off and hover at a set height, no overshoot and with low rise time ect. but still shafts speeds that real world dampers see is nuts, the precision needed to tune that well might be a bit daunting for a 6 months project.

guess you could mount it up to a dyno and tune the gains in the controller through a bit of trail and error till you get in the ball park of how you want it to react. but then so many other things have to have gone just right along the way with no glitches. the design of the damper ect.
 
ohio

ohio

The Fresno Kid
Nov 26, 2001
6,649
24
SF, CA
Not sure if you're a software guy or happy working in theory, but an interesting project is to develop the algorithms to create the damping curves you want with a given damping setup. For example, take a given shimstack and assume you *could* vary the preload dynamically, what would the preload curve look like. Or vary the port size. But I wouldn't mess with more than 2 variables.

A colleague of mine did his thesis on a theoretical drive system with 4 independent electric motors, one for each wheel. He built a 1/24th scale model, treated the motor and battery tech as fixed, and just concentrated on developing algorithms for active steering and braking using the motors. Very cool work, and I believe a few patents came out of it.
 
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
cecil - this is right up your alley: http://web.mit.edu/newsoffice/2009/shock-absorbers-0209.html
it will be too complicated to work on that along with active damping. its only a final year project, not a phd!

UiUiUiUi - thanks for the link, dont have the time to read it right now, but will do asap.

ohio - i am more of an hands on guy, but wouldn't mind doing a few hundred lines of C++ coding just as long as the code is going to bring on some useful/interesting. well, this project i would most likely be in matlab with simulink, which is a lot more tolerable then C++

ohlins actually has a programme like that, for their ttx shock. you can actually download it online, quite fun to play with, but the novelty dies quite quickly. you can change the shims, settings and the springs in there and it shows you the damping curve. whether its done through testing or theory is another question.

well, modelling a shim stack is going to be nuts. i dont even know where to start. working with ports on the other hand would be more realistic.
 
go-ride.com

go-ride.com

Monkey
Oct 23, 2001
548
6
Salt Lake City, UT
dhkid said:
ohlins actually has a programme like that, for their ttx shock. you can actually download it online, quite fun to play with, but the novelty dies quite quickly. you can change the shims, settings and the springs in there and it shows you the damping curve. whether its done through testing or theory is another question.
Click to expand...
dhkid, where can if find this program?
 
B

blackspire

Monkey
Jul 19, 2007
115
0
dw said:
Just doing 1/10th of the math that you need to characterize the input / output of a servo hydraulic system is going to be a massive undertaking. I'd suggest looking in detail at a smaller part of the system. It's so easy to bite off more than you can chew in a project like this, I see students do it all the time. I did it myself many times before I learned my lesson. Concentrate on one area and do a really through job of it. I bet that your professors will recognize the maturity of an approach like this, and if you do a great job, I'm sure your grades will be your reward.
Click to expand...
I agree, honestly I don't think there is any chance that you are gonna succeed in making what you described in your first post. It's likely you need to be maybe 5-6 students to accomplish that. In my third year me and another student did our bachelors thesis on suspension & springs on a small racing car, and there was no chance at all in that time frame to do a damper from scratch. I've completed my master thesis now and afterwards all I can say is, half a year goes by very fast. You need to narrow it down. Maybe if you can just make some kind of concept where you don't design and dimension every part, maybe just a few, that's more realistic. I imagine just designing the damper will take a long time, and then you have to write the code to control the damper. Especially if you don't do your master thesis at a company, if you would do at Öhlins you'd have alot of help but if not you're bound to make a lot more rookie mistakes.

It would be hard to do alone too, since you need various expertises, such as machine design, mechatronics, vehicle engineering etc.
 
Last edited:
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
tomac_racer - thanks for that offer, but i am still far far away from even starting the project. just seeing if its worth it at the moment.

scott - unfortunately, the link doesn't work anymore. maybe you can email ohlins about it if you are interested. page with the broken link: http://www.ohlins.com/test_ny_artikel/Automotive/TTX40-MkII/

blackspire - thanks for you input, will have to think about a smaller project then.
 
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
been thinking about how to simplify the project,

how about just concentrating on just the ESHVs and getting a specified amount of damping from them. ie, have a hydraulic reservoir as a fixed pressure and run the oil through the valve, and try to get an amount of damping that would be in around the right ball park for a damper.

basically it will be the whole thing in a more controlled environment, just focusing on using the EHSVs to provide damping, which really is the main idea in the first place. that way you wont have to worry about transients and it will be pretty much steady state. then start changing the oil pressure, and fine tune the system so that it will work over a range of pressure like in a damper at different shaft speeds.

is that still too complex?
 
Last edited:
B

blackspire

Monkey
Jul 19, 2007
115
0
dhkid said:
been thinking about how to simplify the project,

how about just concentrating on just the ESHVs and getting a specified amount of damping from them. ie, have a hydraulic reservoir as a fixed pressure and run the oil through the valve, and try to get an amount of damping that would be in around the right ball park for a damper.

basically it will be the whole thing in a more controlled environment, just focusing on using the EHSVs to provide damping, which really is the main idea in the first place. that way you wont have to worry about transients and it will be pretty much steady state. then start changing the oil pressure, and fine tune the system so that it will work over a range of pressure like in a damper at different shaft speeds.

is that still too complex?
Click to expand...
Might work, you should speak to your professor or similar and get some input from them. You should make some kind of specification on the project and what you intend to do, make some kind of estimate on how much time each part will take and discuss and bring that to your professor.
 
dhkid

dhkid

Turbo Monkey
Mar 10, 2005
3,358
0
Malaysia
yea, the problem is that the lecturer i am talking to works in the field of control and knows nothing about hydraulics. will probably have to talk to a few different people in the dynamics lab.

anyway, right now i just want to get things in the right ball park and know as much as possible about the topic and what other research is going into this area. then i will have to talk to him and come up with a proposal. from there, they will decided if its worth pursuing.
 
H

HaveFaith

Monkey
Mar 11, 2006
338
0
Wish I would have seen this sooner. My company specializes in Magneto Rheological damping for off-road vehicles and this would have been a great project to collaborate on. Let me know if you still have any interest in it for a school project. We could for sure get something to work awesome for a DH bike, probably better performance than most dampers out there first go. Just need a bit of motivation from a young mind like yourself. (Though I guess Im still pretty young myself...)

Let me know!
 
M

mike425

Monkey
Apr 16, 2004
105
0
HaveFaith said:
Wish I would have seen this sooner. My company specializes in Magneto Rheological damping for off-road vehicles and this would have been a great project to collaborate on. Let me know if you still have any interest in it for a school project. We could for sure get something to work awesome for a DH bike, probably better performance than most dampers out there first go. Just need a bit of motivation from a young mind like yourself. (Though I guess Im still pretty young myself...)

Let me know!
Click to expand...
Thats the kind of offer you want! If its not too late for you I would jump at that.

Where abouts are you studying? I can not remember the details of it, but there is an engineer in Sheffield who just recieved quite a large grant to develop a mountain bike shock controlled but magnets. Kind of like what some of the modern cars use, adjusting the magnets strength changes the suspension characteristics.

May be worth trying to look him up?

Mike
 
Bldr_DH

Bldr_DH

Monkey
Aug 8, 2003
762
0
NO BO CO
^^ Agreed, I think a magneto-rheological damper is the more fool-proof way to go. Fewer moving parts, fewer points of failure.

It's hardly an original idea since a number of car manufacturers are using them, but it would certainly be innovative in the cycling world. I'd love to see what could be achieved with active suspension on bikes.
 
essenmeinstuff

essenmeinstuff

Chimp
Aug 11, 2009
71
0
halifax
I have often pondered moving magnet/coil as a damper. Not sure if it could be packaged very small/light, but controlling the current in the coil would give you pretty wide control of the shock characteristics...

For example:
- Sense amount of torque on the pedals and movement in suspension it could determine the difference between pedal bob and real suspension action = smart lockout/propedal type thing
- nearly infinite control of damping curve vs velocity/position/time of day
- since its essentially being used as a generator in compression and rebound, it could self power.

If it could be packaged, you have the most adaptive damper, and no fluids/seals etc to worry about...
 
You must log in or register to reply here.
Top Bottom