Roll Spring + "3rd Spring" but no single wheel springs

[blister]

What do you think about using two springs/dampers on each axle, but interconnect them such, that one unit acts in heave/pitch and the other only in roll?

I know that "warp" is not decoupled from the other modes with this layout by any means, but could it be desirable to split ride/pitch control from roll control? I could imagine that it will be quite hard to calculate the single wheel rates. Also a standard layout with single wheel spring/dampers and ARBs could possibly achieve the same. But i like the idea of using seperate damping in pitch/ride and roll.

The setup has to be determined but for easier understandig think about a monoshock with a low friction and good dampened slider...

[RobbyObby]

I believe the Dartmouth Formula Hybrid team has run a similar monoshock system with separate ride and roll springs the past few years.

[MalcolmG]

yes yes yes! IMO it would be very worthwhile doing (although I have no feel for how it would actually affect lap times), largely due to the ability to have better control over your damping in roll, which I think would be very useful on an FSAE car that spends a significant amount of time in cornering transients.

I think the trickiest part of achieving this is the physical implementation of it, the pitch/ride control isn't too hard, you can basically just use a regular spring and damper unit in a typical monoshock-type layout. The hard part is the roll damping - most methods I was able to come up with for actuating a linear damper in roll meant the damper would be compressed rolling in one direction, and extended rolling in the other direction - unless you used a through-rod damper with the same bump and rebound performance, that would lead to having different characteristics depending on which direction the car is cornering. You would also probably need to have a separate ARB for springing because the standard spring over the damper will also only work in one direction.

You could have a linkage at its toggle point so roll in either direction will cause the damper to compress, but there are further difficulties with this with combined pitch and roll causing potentially undesirable effects.

So really, I think you need to find/make a good rotary damper. I've never got as far as investigating what is out there - motorcycle steering dampers are an option that gets mentioned a bit, but I have no idea what their performance characteristics are like.

[brettd]

Perrsonally, if you can work the adjustment into your remaining springs I think getting rid of the corner springs is a good idea but I don't know about dampers.

Main reason is you'd be losing your adjustability for road holding (single wheel or warp) or having to change other damping modes to adjust it. This pretty much puts you back to where you started, just on the other side of the fence.

Having roll damping that is adjustable front to rear would be pretty interesting and probably very useful though.

Also, in regards to warp stiffness, with so many transients and sharp turns in FSAE, low warp stiffness (or very low warp stiffness) is not always the fastest way around the track.

[Neil_Roberts]

It's worth doing a weight and volume trade study of corner springs vs ride and roll springs so that you will have an objective basis for your decision, and also a way to justify your decision during the design event. There have been F1 cars designed in the semi-recent past with separate ride and roll springs, mostly for the purpose of producing rising heave stiffness with constant roll stiffness. Such things are necessary to make flat bottom ground effect cars fast.

However, there are profound benefits to an individual damper on each corner. Read Chapter 13 of Think Fast for the full story about that.

http://thinkfastbook.com

[murpia]

Originally posted by brettd:
Perrsonally, if you can work the adjustment into your remaining springs I think getting rid of the corner springs is a good idea but I don't know about dampers.
There was a car at FSUK like this, it had conventional 'corner' dampers but no 'corner' springs instead a heave spring and roll spring each end.

I forget which team though.

Regards, Ian

[blister]

The problem with keeping the corner dampers is that normally the roll mode is already overdamped. So if you add a roll damper you are even more overdamped. But the main reason to have seperate roll damping are clearly the amount of transients we have.

A reasonable sized/tunable ARB can solve this problem. Maybe adding a heave spring will add lots of freedom on your damping coefficients in all modes already

A nice setup of corner dampers, heave spring, roll spring (and corner springs):

Google: "peugeot 905 suspension"

[woodsy96]

I found a picture a while ago looking at specific roll dampers, I belive that this one is off of ETA's 2006 car: h t t p : / / img223.imageshack.us/i/monoshock2resized5iz.jpg/

[exFSAE]

Well.. if you're going to go down the road of intellectual masturbation, why limit yourself?

With corner springs, 3rd spring, and ARBs.. you basically have 3 (coupled) ways of tuning the ride, roll, and pitch response of the sprung mass.

Take one of those out, and you only have 2. Could remove the 3rd spring too and have a "normal" suspension.

Of course.. IMO sprung mass force transmissability and body attitude control are a non-issue in this series... and there are much bigger players in transient yaw response. Roll response is secondary at best for turn-in.

If I were doing this again I'd aim to tune the car well enough with springs and not have to bother with ARB's and 3rd springs.. and I bet it could be tuned to handle better than 90% of these cars.

Just my opinion.

[Silente]

In my opinion the main problem with this solution is that rotational speed you will have on the rotating roll damper you is too low to actually control this movement with such a device

[HenningO]

During my last two years in FSAE I got bitten by the Monoshock bug. Both designs were what we called "sliding bellcranks" as can be found on for instance some F3 cars, however ours had a roll damper. Here are some of my thoughts:

- It was a whole lot more fun designing a monoshock system compared to a traditional system. Does "fun" help you win a race, probably not. In any case it is a very good engineering exercise.

- We had to develop our own software to calculate motion ratios and geometries (this was before OptimumK was released).

- A substantial amount of money can be saved by not having to get 4 coilover dampers. We ran 2 coilovers and 1 steering damper - total savings were about 1000 USD.

- The systems ended up weighing about the same as traditional systems.

- Reducing friction and hysteresis in the system is quite crucial. In the end we ran hard anodized Al shafts in ceramic bushings. With a thin film of lubricant, friction during load was found to be "very low".

- We ran a linear steering damper on the front axle to add damping in roll. It worked really well and 2 clicks could be felt. I think our motion ratio for the roll damper was something like 12 mm/1 deg roll. So getting enough displacement on the damper was not an issue. First year the roll motion ratio was constant with heave, the second year it was allowed to vary some with heave which simplified the mechanical design.

- As exFSAE mentions, transient yaw response is crucial for FSAE cars and it was a dream to tune that with the roll damper. After having had the steady state balanced figured out on the skidpad, finding the right roll damper setting was done instantly after a couple of laps on the track. The range of the roll damper went from ultra lazy turn in to snap oversteer turn in.

- Being able to run a progressive MR in heave allowed us to run softer heave springs without bottoming out. The softer setting made the car easier to drive and feel for our drivers.

- In the rear we ran a very simple solution with two rockers that both attached to a single coilover. An U-type ARB worked as the roll spring.

- In my mind the best implementation of a monoshock has driver adjustable front roll damper in combination with a driver adjustable rear anti-roll bar. With two dials you'll be able to adjust both transient and steady state balance of the car. Just make sure the driver knows which way to turn the knobs!


First year design (Dartmouth), pullrod actuated. The belleville/spring washers were used as roll springs.
h t t p : //img818.imageshack.us/i/monoshock07.jpg/

Second year design (KTH), pushrod actuated. Belleville washers are inside the black tubes, this design simplified changing/re-stacking the washers.
h t t p : //img299.imageshack.us/i/monoshock08.jpg/

Second year design (Dartmouth)
h t t p : //www.flickr.com/photos/thayerschool/3196552985/sizes/l/in/photostream/



Conclusion:
Will monoshocks automatically win the competition for you, no.
Will it challenge you as an engineer, yes.

[MalcolmG]

Originally posted by murpia:
There was a car at FSUK like this, it had conventional 'corner' dampers but no 'corner' springs instead a heave spring and roll spring each end.

I forget which team though.

Regards, Ian
RMIT had a similar system on the front in '07, they had a T-type ARB with the centre attached to a 3rd spring/damper unit which acted in heave, and corner dampers with no springs on them. They only ran the system one year though, from memory in '08 they were back to a 1 spring/damper per corner + ARB, and last year in the front they had a heave spring but no ARB (I think, never had a close look)

[flavorPacket]

Originally posted by blister:
The problem with keeping the corner dampers is that normally the roll mode is already overdamped. So if you add a roll damper you are even more overdamped. But the main reason to have seperate roll damping are clearly the amount of transients we have.

Check your math. Your critical roll damping should be several times more than your critical single wheel damping or critical heave damping.

[Zac]

Cinci ran an interesting setup last year. They had monoshocks front and rear operating in heave and two additional linear shocks that operated in roll through a mechanical linkage that connected the front and rear suspensions. But I don't know that they ever really got it dialed in. It seemed like they were fighting their spool quite a bit.

[murpia]

Originally posted by HenningO:
During my last two years in FSAE I got bitten by the Monoshock bug. Both designs were what we called "sliding bellcranks" as can be found on for instance some F3 cars, however ours had a roll damper.

- Reducing friction and hysteresis in the system is quite crucial
I just thought I'd highlight this point, as it's easy to overlook...

Regards, Ian

[Brian Barnhill]

Originally posted by murpia:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by HenningO:
During my last two years in FSAE I got bitten by the Monoshock bug. Both designs were what we called "sliding bellcranks" as can be found on for instance some F3 cars, however ours had a roll damper.

- Reducing friction and hysteresis in the system is quite crucial
I just thought I'd highlight this point, as it's easy to overlook...

Regards, Ian </div></BLOCKQUOTE>

+1 to this!

WPI ran monoshocks on a few of our cars (and it's on the car that I bought off the school and run as an A-mod) The first year did not have roll damping, the car was driveable, but did suffer from lack of damping in some condtions.

The next two iteration used motorcycle steering damper as a roll damper, very similar to HennigO's setup.

It is quite a different step as far as engineering the vehicle. Performance wise I am not sure I can comment either way. I've drive both traditional spring per corner cars and monoshock cars, and it seems to be varying which one's drive better. There are so many other factors at play here.

That said, a monoshock can handle well, when properly dialed in, and MAY find those few extra design points. If your system is not 100% though the design judges will cruicify you on it. Our initial design lost us a lot of points, espically for the lack of damping and friction/hysterisis in the system.

[2BWise]

Originally posted by Zac:
Cinci ran an interesting setup last year. They had monoshocks front and rear operating in heave and two additional linear shocks that operated in roll through a mechanical linkage that connected the front and rear suspensions. But I don't know that they ever really got it dialed in. It seemed like they were fighting their spool quite a bit.

They did. It was very interesting and I don't know if they ever fully proved the concept, although I do know the car is close to being back together for the end of this summer. Maybe soon they'll be to comment.I drove it and the spool was so awful that I never got a feel for how the suspension performed.