Direct Acting Spring Damper

[Claude Rouelle]

Trent,

About Z

Some good observations about Z here. There is long time I have decided to ignore his comments and respond to his useless sarcasms. Too bad he lacks some respect and basic manners because some of his engineering perspectives could be useful.....although to my knowledge these perspectives have never been confirmed by success on race car he built or teams he worked with. That is true that to successfully work in a racing team you need some positive attitude and some communication skills.....

To all,

About front and rear ARB

Ask your driver about the car balance; let's say he tells you he has is O/S. You need to increase your TLLTD by either stiffening your FARB or softening your RARB. Which solution will you choose? To make that decision, ask your driver a second question: is your car too nervous or too lazy? If it is too lazy you need to stiffen your FARB, if it is too nervous you need to soften your RARB. Now what do you do if the car oversteers and is too nervous and you do not have any RARB? You will need to soften your rear spring and that will destroy you ride and pitch targeted frequency. That is a reason why most race and passengers cars have frotn and rear ARBs

About high speed damping and blow off

Lest' be serious and practical: what do you call damper low speed and high speed? At what damper speed do your damper reach the blow off speed and where on the circuit? If you do a damper histogram you will see that on a FSAE you probably spend 80 % of the time between - 50 mm/s and + 50 mm/s. That is usually what damper guys call low speed. There are a few little bumps at FSAE tracks but these are not Sebring bumps, chicane curbs at Long Beach or WRC special stage junps....

Claude (in Le Mans; I have a blast!)

[Ahmad Rezq]

Claude
Excuse me. But many of Erik posts which I read are related to basic engineering skills.
Example : The time when many professional engineers offered complex ways to explain anti features or jacking Z offered Simple free body diagrams which were more useful.

I don't know if Erik has developed new theories in VD or not but his contributions in many topics always make sense.

[Z]

Trent,

Tough day at the office? Given that you now work in NascarLand, you have my sympathies.

I have yet to see your response to me in the moment diagram with weight transfer thread...

I was hoping more people would post on that thread with more specific details of how they use those MMMDs (which is what JP, Silente asked for). I have suggestions for how those diagrams could be made more useful, and I may get around to posting them... But should I bother?

You can get away with running no ARBs some of the time, you can't get away with running only springs all of the time.

I have NEVER seen a good off-road racecar with ARBs, and they are all about good suspension. Oh, yes, ... they are different to FSAE. And so is Nascar...

Horses for courses...

And your brown go-kart is the answer right?

See bottom of post...
~~~o0o~~~

MCoach,

The above discussions about "changed handling from changing MRs" is a good example of the SLOPPINESS of thinking that I was getting at. So, one more time, putting it as simply as I can...

Ritwik's rear suspension has a Kinematic layout of DASDs and suspension-arms that gives a 3% variation of instantaneous MR, over the 8 cm range from full-droop to full-bump. (This is my understanding from Ritwik's post top-of-p4, which is quite well written, given that English is probably not his first language.)

IMPORTANTLY, on Ritwik's car:
The MR at static-ride-height NEVER CHANGES.
The MR at full-droop NEVER CHANGES.
The MR at full-bump NEVER CHANGES.

It is simply that the MR at these different suspension positions is (very slightly!) different to the MR at the other positions.

But (!), on your car,

The bellcranks are already designed with the mounting holes in them, so we only had to take the down time of moving one bolt on each side of the car.

From which I assume, you are CHANGING THE MRs OVER THE FULL RANGE OF SUSPENSION POSITIONS, from one "car-set-up" to another "car-set-up"!!!

Unfortunately, you have not bothered telling us how much your MRs changed, so NO meaningful conclusions can be drawn!

Anyway, I hope you can see the huge difference in the above two cases (ie. they are NOT comparable!).
~o0o~

Bonus points for anyone who cares to post calcs for "How much handling-balance-change would come from Ritwik's above suspension, and in what circumstances?".

Hint: By my reckoning, and based on my Assumptions written on this scrap-of-paper, because LLTD would only change as peak cornering-Gs change (because of more body-roll, to spell it out), even a doubling of cornering-Gs would give a change in LLTD so small as to be imperceptible. This imperceptibility because the tiny change in LLTD would be against a background of the car going ~40% faster, having ~double the aero-drag (hence more R-load), ~double any aero-lift/downforce and associated shift in balance, ~double the toe/camber++ compliances, and massively changed loadings/temps++ of the tyres...

It is entirely possible that, because of all those "other" changes, the car would move from, say, mild-US to massive OS, even though the (tiny!) LLTD change would be in the opposite direction.
~~~o0o~~~

Back To DASDs.
=============

Posted by Claude:
... observations about Z...
... to my knowledge these perspectives have never been confirmed by success on race car ...

My intention when first promoting DASDs was to give the smaller and less well resourced Teams a potential advantage. By tossing their previous year's Rockers+++, they could save time, money, and weight, and thus put more effort into the much more important parts of the project (eg. aero, driver-training, reliability, more aero+++). Oddly, it seems that it is mostly these lower-half-of-the-ladder Teams that still see DASDs as the work of the Devil.

On the other hand, and contrary to Claude's claim above about "... never been confirmed", it seems that the Teams towards the top-of-ladder think differently. Some of them like DASDs! Hmmm, maybe that is because these more successful teams are using that ancient approach of thinking things through, "Plato's Academy style"...???

Claude, check the ladder. And when you have time, try thinking through that Parallel Axis Theorem... You know, with clear Definitions, Assumptions, etc.

Z

(PS. Ahmad, Thanks. No new VD theories. I am still trying to learn what all those clever people from hundreds/thousands of years ago gave us...)

[Z]

... you need to give your self the opportunity to adjust your balance ... But you have no ARB? So what will you do? Change the springs probably...
... you will need to switch to a spring of let's say 125.0 lb/in to a 137.8 lb/in ... that doesn't exist. It will be 125 or 150 lb/in...
...
Up to you but in my opinion front and rear ARBs should have been part of your initial design.

Ritwik,

There was some discussion on the Lincoln-2013 thread (in Competitions section of Forum) about how to make fine adjustments to DASD-rates (see Matt Davis's posts bottom-p6 and p7).

Hopefully this link to Uni of Cincinnati's Facebook page will show you how it was done...

https://www.facebook.com/bearcatmoto...type=1&theater

The idea is simply to insert some rubber spacers between the spring coils, thus taking a half-coil or more "out of action", and thus slightly stiffening the spring-rate.

Also UoC's 2013 car was originally intended to use ARBs, but after initial testing they found them to be UNNECESSARY! And searching around for other recent competition winning FS/FSAE cars shows ... same result!

Z

[Claude Rouelle]

I uses bump rubbers a lot.

I fact I remember testing over 2000 of them on a specific test machine for Indycars several years ago. Nascar team play that game too. They are very useful fora very ride height sensitive heavy aerodynamic cars.

But their utilization can be misleading

Bump rubbers are not bad ideas but
a) It is an engineering competition; you will need to come with bump rubber curves Force Vs displacement
b) Bump rubber have huge hysteresis. Will you take this into account?
c) Bump rubber are not only stroke sensitive but also peed and temperature sensitive
d) In a perfect world, as their stiffness are not linear you need a damper which is stroke sensitive
e) If you use bump rubber with clearance at the moment the bump rubber is "engaged" the force variation is negligible and then progressive, that is what you want from a bump rubber. BUT the stiffness variation is HUGE. If you look at the SLOPE of the Force vs movement you realize that you suddenly add another spring. Your damper is not adequate anymore. I have been working on many race cars that worked well when the bump rubber was on ( we tired to do that in straights) or off (we tried to avoid using them in corners) but got crazy amplitude and frequency once the bump rubber was active Passengers can test driver can really feel that issue too: bumper rubbers on passengers cars are usually used for curbs "to avoid getting the tire in the inside face of the fender" but sometimes if the bump rubber become active in roll or in pitch the heave and roll or pitch frequency chance so much that the driver can have difficulties to control his car.

I don't say you will need to measure all this but you need to be aware of all this and ideally quantify bump rubbers advantage and limitations.

While you will have that finished you may will reconsider rockers with variable motion ratio. But, on the other hand, only a few FS/FSAE team have shown their ability to do so.

Make it simple yes but know how simplistic it could be.

[ritwikdas18]

Thanks everyone for guiding me. I would like to put my final decisions, instead of reiterating over and over.
1.Z I couldn't go with using high stiffness springs because of the following reasons

a.My low motion ratio and high spring rate put my rear compression off the chart as is shown in the figure
attached.JPG
b.Therefore I had to change the frequencies to get it within the curve. Now my front and rear static deflections are 3.4cm and 3.9 cm with frequencies of 2.7Hz and 2.5Hz
https://docs.google.com/spreadsheets...it?usp=sharing. [ Please ignore the damping decisions section only . They are mentioned below]
Goost,I raised the roll centre to decrease the roll gradient but still roll is compromised due to low ride rate compared to former.All of the above 2 changes led to the following:
attached.JPG
2. The final damping decisions were taken as follows
Front Compression: Code 1
Front Rebound: Code 1

Rear Compression: Code 4
Rear Rebound: Code 2


attached.JPG
3.Due to the above frequencies there is now more travel, but it is still within the range I feel as Z pointed out[This is because when I increased by rear MR to 0.62 , the rod extension length increased as well, thus lesser change in MR with roll ].I would go through that forum page you told.
4.My spring rates are as follows:
Front: 29.8N/mm
Rear:~50N/mm
Custom Spring Manufacturing in India is more costlier than in US. Since I am getting the front springs from KAZ, and many people in this forum are from US,I would like if someone could post here a reliable custom spring manufacturer in the US.{The spring would be defined as follows:-

Material: Chrome Silicon

ID: 43.18mm (1.70 inch)
OD Max: 56mm (2.20 inch)
Free Length Max:127.00mm (5.00 inch)
Spring Travel-70mm (2.75 inch) (This may not be required)
End Coil Configuration- Closed(Squared) and Ground
}

On another note I would like to thank Z for saving so much time of my team with these DASD setup. He is right in saying that he wants to give small resourced teams a potential advantage.I may lose some points in design since I do not have ARBs and bellcranks, but hoping to cover them up in dynamic events or physical data from car testing.Some facts
In the recent FS India event ,there was/were
1 Car in Acceleration
3 in Skidpad
5 in AutoX
7 in Endurance(out of which only 2-3 completed)
out of a total of 40 cars.
Claude was there at the event as Design Judge. FSAE Michigan results also were more or less the same.

I will post more here when we go through the testing phase.

[Z]

Ritwik,

... I raised the roll centre to decrease the roll gradient but still roll is compromised due to low ride rate ...

This is a BAD DECISION, albeit quite common in FS/FSAE.

It is the type of decision made by students with limited understanding of the many issues involved, and usually made simply because of pressure from other people who perhaps have even less understanding, but who doggedly believe that certain meaningless metrics must be met (eg. "ride frequency", "roll gradient", etc.), regardless of the details. Nevertheless, and all things considered (ie. you are competing against other students), you might get away with it....

More important is to realise that it is VERY EASY TO INCREASE YOUR SPRING-RATES (see below). But do not worry about this now! Instead, get the car built and running as soon as possible. Use your above chosen spring and damper-rates to start with. Preferable is RCs close to ground, say 5 cm or less (especially so with soft springs!). DO NOT have RCs above 10 cms!.

Anyway, once you have worn out a set of tyres on your above roly-poly car (ie. with static deflections = ~3 - 4 cm), you can try stiffer springs. Please look again at the link I gave above to Uni of Cincinnati's Facebook page. See how they have pushed some spacers between the coils of the springs. All you need to do TO STIFFEN YOUR SPRINGS is PUT SPACERS BETWEEN THE COILS.

Alternatively, you can REMOVE COILS by cutting them off, and REPLACE WITH A SOLID SPACER (say, aluminium or even plastic). For example, cut the spring in half, put one half on shelf, add a spacer of appropriate length to the remaining half, and you have just DOUBLED your spring rate! Again, note that with your MRs you get the full ~5 cm suspension range with only about 3 cm of spring movement. So you only need about HALF-LENGTH of the springs you are currently specifying.

Also, I am sure that there are NNN millions of motorbikes in India that have springs about the right size for you. You know the inner-coil-diameter you need (ie. ~44+ mm, but can be bigger), so now work out suitable wire diameters and number of coils to achieve your desired rate. Then go looking through the bike stores and wrecking yards for suitable springs. A longer spring is fine, because you just cut it shorter to suit (and use oxy + angle-grinder for "closed and ground end"). Then, if necessary, machine some aluminium (or steel or plastic) seats so it fits neatly onto your QM dampers. <- All this is very easy to do, but do it AFTER first lot of testing...

Your selected dampers are good enough for now. Any damper issues you might have in testing can be addressed later (also easy to fix).

Z

[ritwikdas18]

I can change the RC's back again till the suspension clevises are put during manufacturing , not an issue but what shall i do with the increase in roll gradient that comes with it? What value of roll gradient do you think that must be met?
Thanks for the advice on springs . My current RC heights are around ~2.7 inch and ~3 inch Rear currently with previous heights being around ~1.7 inch and ~2 inch.

[Z]

Ritwik,

My current RC heights are around ~2.7 inch and ~3 inch rear ... with previous heights being around ~1.7 inch and ~2 inch.

Your "previous" RC heights are preferable (ie. 4 and 5 cm). Your higher "current" RC heights are bordering on DISASTER, especially with soft springs. Because..., when cornering fast the rear of the car will suddenly "jack-up" in mid-corner, the inner-wheels will suddenly lift off the ground, the driver will suddenly soil his nice new racesuit..., and the driver will certainly never try cornering that fast again!

The standard solution to this problem is simply to lock the suspension up rigid, because "...any suspension will work, if you don't let it...". But there are better ways...
~~~o0o~~~

...but what shall i do with the increase in roll gradient that comes with [lower RCs]? What value of roll gradient do you think that must be met?

"Roll Gradient" is one of the meaningless metrics I mentioned last post. In FS conditions a rigidly sprung car (ie. RG = 0) can go quite fast. A well suspended car (ie. with non-constant RG!) goes even faster.

IMO, "well suspended" means quite soft and (relatively) overdamped in the small range around ride-height, say, +/- 1 cm. Outside that soft range bump-rubbers add increasing (non-linear!) spring-rate in the bump direction, and there is not much need for any extra droop movement, so the suspension can be droop-limited below -1cm (though this depends on circumstances).

So, rather than thinking about some magical, constant, roll-gradient, it is more helpful to think about the peak-roll-angles you expect to see. These are obviously related to the peak-camber/inclination-angles your tyres will see, via the suspension's "Camber-Compensation", etc. That is, with 0% CC you want small peak-roll-angles (ie. stiff springs), else you simply wear the outside shoulders off your tyres. With 100% CC, say from beam-axles or short-lateral-swing-arms, you can have large roll-angles (ie. soft springs) with no adverse effects on camber-angles, or resultant tyre-grip.

Peak-roll-angle can be controlled by:

1. Spring stiffness over its linear range. Obviously, softer springs = more peak-roll-angle, and vice versa. But nothing stays constant, or "linear", forever...

2. Non-linear springing. For example, soft springs in centre of range (= good for grip over small ripples), then bump-rubbers and droop-limiting to set peak-roll-angle. The right sort of bump-rubbers are important if the track has many bumps ... so NOT important in FS! Overall, in FS, the droop-limiting is more important, because otherwise narrow-car + large-inner-wheel-droop leads to -> jacking -> CG rise -> rollover! Or driver too scared to drive fast.

BTW, droop-limiting is usually set by spring-seat-height (on the threaded-damper-body) together with the ride-height-adjuster that you should add to your DASD (see bottom of U-Cincinnati's DASDs for hints). Typically, in FS conditions, the wheels should be allowed to droop downwards NO MORE than ~2 - 3 cm from "static", at which point the dampers reach full extension and set the "droop-limit". Droop-limiting can also be done independently of the dampers, say with adjustable seat-belt-like nylon-straps. These are absolutely essential in off-road racing, else (standard) dampers get ripped to shreds.

3. Stiff low-speed damping. This can be enough to limit peak-roll-angle in short corners such as slaloms and chicanes. In FS, or general road-racing conditions, it is usually best to put most of the low-speed damping stiffness on the rear-wheels (ie. fronts are kept soft). This "loosens" the car (ie. moves handling balance to OS) on corner entry, making it "turn in" better. It also "tightens" the car (more US-ish) on corner exit, allowing better acceleration out of corners.

Again, all of the above is "fine tuning". Build the car and do a lot of driving with whatever SDs you choose to start with. Then think about above issues, and try a bit more-or-less stiffness, here-or-there... Always easy to add more stiffness...

Z

[mdavis]

It's good to see UC's 2013 car still being relevant. As the suspension guy for that car, it makes me happy.

Having skimmed this thread, I think there are a couple of good view points in here, Claude and Z, who are approaching this from 2 separate angles, and I think they could end up in a similar space (faster cars and better engineers).

Having hung around with a bunch of autocrossers for the last couple years, I can definitely say that a pretty significant majority have no idea how to make their car faster (via setup tweaks), or why their current setup is fast. FSAE teams would be wise to seak out the few that do know how to make a car fast and pick their brains. An FSAE course at Lincoln (MIS is a little different, no real comment about European courses, since I've never driven them) has a similar number of driver inputs that your local autocross has for a big car. Go find the top 5 guys in PAX at your local SCCA events (for US teams, teams from other parts of the world, find whatever local motorsports applies to you), and try to co-drive with them. We did this in 2011, and it was very eye opening. We put a fast driver in our car, and he put 3 seconds on our drivers on a ~30 second course. Until you can keep up with nationally competitive autocrossers, your drivers are slow. There are only a few FSAE drivers that I think I've ever seen that I would consider "fast". Plenty drive fast cars, but few drivers are actually fast. This is going to be a key component to a good dynamics score. PS, it's fun to drive. Get out there and do more of it. It's much better than working in the shop or on the computer... I would also suggest going to a few dirt go-kart races, if you want to learn to be smooth with steering inputs.

Now, if I were still involved in FSAE (namely, the last 2 years after our 2013 campaign), I would be working on the following things: correlation of simulation to on-car results, finding the weak points in the design of our car (there were several in 2013, and anyone on our team will admit it) to improve for the next iteration. We started doing some simulation in '13, and it went alright. We didn't have any idea how the graphs from the computer tied to the car's performance. That was our issue. Given 2 more years (hell, 1 summer), we would have had a much better idea of why the car did what it did on track. Someone asked me if it was car or driver that had us ~6-8 seconds back of GFR in autocross. I think it was 50/50 car and driver. There was plenty we could have improved on both fronts...

Our 2013 team leader told Claude that if we had our team together for 2 more years, we would be challenging for top 5 overall at MIS. I think this is a 100% factual statement. Here's how the timing breaks down:

Year 1:
1. Build a reliable car
2. Tune the hell out of it (we were sort of on the right track, and the Lincoln pace showed it)
3. Train your drivers
4. Develop some kind of MMM (or similar) simulation to show what the car is doing on track

This gives good dynamics scores, for a reasonable overall finish. This is important for building sponsor base for the next 2 years.

Year 2:
1. Continue 2 and 3 from above
2. Correlate the simulation from Year 1 to on track performance
3. Simulate a variety of set-up options and validate them on track
4. Make very very small tweaks from Year 1 car, improving only the biggest 2-3 errors

This starts down the path of predictive simulation, and closes the loop on the simulation developed in Year 1. With the good drivers and a faster car, you're looking at a better dynamics finish, and with better understanding of what the car is doing, you're probably doing better in Design as well.

Year 3:
1. Use simulation that was designed in Year 1 and correlated in Year 2 to design new car, using plenty of sense checks along the way
2. Build car early
3. Test often
4. Repeat 3, a lot

By this point, your drivers should be very capable, and your car should be quite fast because you're on your 3 iteration of the package.

This takes time, and is probably why it's not followed very often. Please note, if you make any major changes (engine, tires, aero/no aero, etc.) you basically start this process over again... This is what we would have done, had our 2013 team been back for 2014 and 2015.

If anyone has any specific questions about our 2013 car, post them up, I'll be happy to answer them.

-Matt

PS, Z is giving away some autocross setup secrets in his last post. These were learned by running softly sprung cars on DOT race tires. The cars would be very close to dragging their mirrors from all of the body roll, but they were very very fast around an autocross course. Those cars were running DASD's with stock (generally small) or potentially 1 large anti-roll bar. If I were to build an A-Modified for SCCA use (basically, a ~180hp, 900lb FSAE car with lots of sprung mounted aero), I would use DASD's, and potentially no ARB's.

edit: Having watched all of the afternoon session of Endurance at FSAE-M this year, the GFR car was incredibly fast. Did DASD's cause this? No, they have been fast for years now. Did DASD's make them slower? It sure didn't look like it on track. Good job to the GFR team, your car was very well put together. As mentioned by the Kansas adviser in your paddock, I wonder how many teams show up with DASD's next year, because you guys showed they are not holding a car back... We only showed that it was feasible. GFR showed the real speed potential.