Z,
You think the main reason that 4WD is better is that it yields more regenerative braking?
I.....I......I......
Disagree.
Printable View
Z,
You think the main reason that 4WD is better is that it yields more regenerative braking?
I.....I......I......
Disagree.
Those transient issues which hurt turbo performance also hurt NA performance. The engine is doing the same job, so why treat the peripherals differently? If the rules were to place the throttle body after the resitrictor for all types of induction, then turbocharging would be a more difficult choice. The current proposed rumor highly biases a turbo choice for powertrain in the future.Quote:
Originally Posted by apalrd
No, Bemo. But that is kind of the point of creating a dialogue on the forum, correct? The unofficial FSG rules were removed after they were leaked onto the forums and the community was able to respond, both on the forums and in formal letters of complaint.Quote:
Originally Posted by Bemo
I agree. But the Americans are who suffer more than the Europeans with the late rules. If you are an American team wishing to compete in Europe and at Michigan or Lincoln, your car needs to be running 1 to 2 months earlier than the Europeans. That is a lot of time, which is why we are well into our development for the 2015 season. We are also involved in providing feedback for some of the proposed rules.Quote:
Originally Posted by Dunk Mckay
Yup. And what happens when the oil supply runs out? Like in a high-G situation on an engine originally developed with a wet sump? If the turbo is always removing the oil from itself with vacuum by dumping it into the intake, if it isn't supplied with oil it will overheat.Quote:
Originally Posted by GXP_Matt
I am not worried about turbocharging an FSAE car. I have access to resources which makes it possible for me to try different things and brake a few engines getting to a good turbocharged system. For a smaller team, however, creating a good working turbo system becomes yet another barrier preventing upward motion in the FSAE classes.Quote:
Originally Posted by GXP_Matt
I like the risk in racing. Go to the pits of any circletrack race and you will quickly get run over by a 3500# car driving 20mph with no clutch. The organisers of the event are worried that fresh people to racing won't know how to operate dangerous devices like race cars in a safe manner. Guess what FSAE is full of? Thousands of 20 year olds who have never gone to a race track and built their first car. If you are concerned with litigation than you have to cover your ass. Safety rules and other preventative measures make sense when you stand in the organizer's shoes.Quote:
Originally Posted by Mumpitz
Thanks for the photos Matt. Everything looks pretty wedged in. Do you have any data about the ambient temps in the engine compartment? We put peak temperature stickers everywhere to track the hotspots for cheap.Quote:
Originally Posted by Mbirt
GFR's simulations give us a lot of guidance. I have been speaking for myself and not for the team. If turbos happen, it just means that I get a bigger budget and new toys and more people dedicated engine development and engine sponsor resourcing. I have had potential sponsors turn me down for sponsorship because I was not running a turbocharger.
I think that changing up the turbo rules would need to be couple with a rule creating separate classes. A small and underfunded team working against their university is never going to be able to compete against turbo-charged hyper-aero cars. But as an organiser you can't simply remove all of the fun and learning possibilities surrounding aero developement and engine development. It would turn FSAE into Baja, which is just a parade of the same car driving around the track testing to see who has the higher quality construction.
It could, depending on the wording of the rules. I think that's what the plenum volume limit tries to fix. But I don't like how difficult plenum volume is to measure. A turbo also doesn't just have the time delays of filling the manifold (which can be helped with transient controls and ETC, NA or turbo), they also have time delays of spooling the turbo, and building boost to whatever high pressure the team wants to run. A supercharger does not usually have this issue as it can build boost without significant exhaust mass flow.Quote:
Originally Posted by Swiftus
Maybe the rules should allow the throttle to be downstream of the restrictor and compressor but require it to be close to the compressor? Or the plenum volume restriction should only apply to the space between the compressor and throttle (allow additional, unmeasured plenum volume between the throttle and engine?)
We currently run software/calibration changes to help with the filling delays of a 4.5L plenum (on a 450cc engine).
Uh, if you can't supply pressurized oil to the turbo then you'll have bigger problems on your hands very quickly, like a spun rod bearing and after that some unintended engine block ventilation...Quote:
Originally Posted by Swiftus
I don't think you understood my point that the proposed turbo rules make it EASIER on smaller teams, since the installation is simplified. A small team could slap a turbo running a few lbs of boost on an un-optimized engine package (maybe even intentionally compromise the engine for weight savings) and be making more torque than a fully built and CFD optimized NA engine of a well-funded larger team.
Wherever a volume is set for the plenum, it is going to add a step and therefore more time to tech inspection. At every competition without a dedicated scrutineering day, it will be adding more time onto a process which already cannot be completed within the allotted amount.Quote:
Originally Posted by apalrd
...which happens quite often in FSAE?... How many cars complete Endurance? Autocross? Skidpad?Quote:
Originally Posted by GXP_Matt
I did understand what you said. And for a no more effort or time but a lot more money, a well funded team could make more power for less weight without any compromises in safety or performance. Turbo manufacturers would be having a field day because of the huge demand for tiny turbos! 'Custom turbo sized for a 250-610cc engine you say? How much are you willing to spend?'Quote:
Originally Posted by GXP_Matt
This point's intent was to show that making turbos easy will simply widen the gap. If it is currently a lot of effort for small gains, than imagine the same amount of effort for huge gains? Does the small team have the resources to try and keep up with the big teams?
I am all for making the competition more interesting with the easier implementation of complex technologies like E-throttles and turbochargers.
BTW, what do you do at BorgWarner Turbo Systems?
My thoughts exactly. For someone who comes on here and calls all current Formula students, and I quote,Quote:
Originally Posted by stever95
Mr. Z would understand the simple Dynamics problem comparing the traction loads available for FWD, RWD, and 4WD configurations on a given vehicle.Quote:
Originally Posted by Z
Thinking out loud, with the electric cars I've seen running a Motor-on-Hub arrangement, wouldn't the inertia losses on a 4WD Electric car be lower than on a 2WD Combustion car? In the electric car there's another set of wheel/tire/hub to turn in addition to the four electric motors, however that is offset by not having to operate an engine/transmission/drive chain/differential/driveshafts, intuition would suggest that combining lower inertia losses and a higher tractive potential for an equivalent power level would give higher acceleration, all other things being equal, no? Someone please correct me if I'm wrong.
Z,
We've gone down this road before a while ago on a different thread, discussing drag strip methods for the FSAE acceleration event.
Considering your third question, you may be 'confident [we] will all fail', but I remember your explanation concerning inertial resistance to pitch motions about the rear-wheel.
However, in order for resistance to pitch motions to come in to play, there must be pitch motions. In other words, there must be enough grip off the line to lift your front wheels in the first place.
You basically agreed to that, by saying we should assume a start line acceleration of 2.5g.
Like I said before, I don't see that happening with 'normal' tires on a 'normal' track. 1.5g might be doable.
You still haven't told me where you do your tire shopping.
I'm sure there are things we can learn from drag racing, but I feel you're a bit overeager to draw parallels.
This is how sticky a drag strip is:
https://www.youtube.com/watch?v=x5wQajhGaeI
so this:
http://bit.ly/Uj6YRP
comes as no suprise.
On the other hand, I don't remember seeing cars in a circuit race class lifting their front wheels.
Again, no surprise, they'll just have their rear wheels spinning instead.
If you want to make a point about how slow and boring current FSAE car designs are, and how stupid 4WD is, please point out a single race car that can go around corners (i.e. one that doesn't do its runs on a duct tape track, using tires with 9.5" walls at <0.5 bar that last 3km at most), and that accelerates quicker to 60km/h than the fastest 4WD electric FSAE cars.
Regarding the weight distribution on 2WD (combustion) cars. If you only look at acceleration then yes, you want to design a static weight distribution as close to the driven axle (usually rear). That's why dragracers look the way they do. However, there are 4 dynamic events in formula student, where 3 out of 4 require some sort of lateral force creation. With minimal load on the front axle, minimal total lateral force is created needed for cornering. I guess every team that has some sort of lap simulation will agree that the average speed during these 3 dynamic events is significantly increased by increasing cornering speed. I won't mention F1, ok maybe I did. Therefore increasing points obtained significantly. So yes, you take a slight hit on 75 points from acceleration, but you gain a lot more on 50+150+300=500 points. Hmm... hard decision.
Then you have the issue of balancing the car for these 3 dynamic events. Even if you can get close to a good setup, a slighty hickup will cause the car to oversteer with a very rearward CoG (ask the people from the DUT11 for example). Yes drifting is a lot of fun, but it's also not very quick around a corner. As added bonus you'll wear out your tires quite rapidly, which is costly at ~150 dollars a pop.
And then I haven't even mentioned tire load sensitivity yet...
Note: if anyone has 2.5G tires I will trade them for ours :)
Rear weight bias in fsae has definateley been done quite well by maryland
https://fbcdn-sphotos-g-a.akamaihd.n...81080726_n.jpg
as far as lateral force generation goes, if I remember correctly, there weren't many SAE cars faster around a skid pad, or scca autocross. (even though it was heavier than most)
transient response and the length required to get the weight distribution could be improved with a custom, short engine/driveline as discussed by other in this thread.
One problem with the rear weight distribution which will be getting worse with the proposed rules is trying to get the aero balance to where it should be. In my non-aero-person opinion, with current packaging "maximum" downforce on a minimum length car would likely occur with an aero balance pretty close to 50:50. With these new rules who knows? Maybe some low-powered aero teams might show up with 65% front weight distribution and skinny rear tires to maximise total balanced downforce? (i am choosing to keep the questionable claims of creating massive underbody downforce in their other threads)
It will interesting to see if any of the top teams come up with something radically different in these new rules
Daniel,
If your theory on axle loading held true, there would be no point to reducing mass. Minimal axle loading may create minimal lateral force, but there is less mass to move. See the Deltawing concept.
Having done quite a bit on turbo tuning with the UoW cars, I agree with GXP_Matt that turbo implementation would be easier. Drainage was a frustrating problem and on some cars resulted in much victory smoke. Also, the amount of work required to make the turbo provide a gain was huge. This doesn't seem to be appreciated by many, who think it's just a matter of whacking on a turbo and boosting like hectic. With the proposed intake path, it would be possible to do just that. I don't see this as a problem, because it forces teams who go down that path to realise the other requirements for successful turbo implementation (assuming the oiling and some lag issues have been resolved by the new layout). I can go through my take on these requirements if anyone is interested, but suffice to say they are applicable to real world situations.
To reiterate my main point here, there is a strong suggestion in the proposed Rules changes, and in many students comments here, that E-cars have an "unfair advantage" over the C-cars. This is supposed to be particularly so with the 4WD E-cars in the Acceleration event, and the recent FSUK-14 results might (!?) be used to support this view.
FSUK-14 results:
#. Team ......... O/A ... Acc.
===================
1. Delft(E) .... 856 ... 73
2. Stutt(C) .... 837 ... 38
3. KIT(C) ..... 828 ... 37
4. Zurich(E) . 827 ... 75
5. Monash(C) 821 ... 21
My argument is that if the above C-car teams learn how to do Acceleration properly, then there would be NO E-cars on the podium!
I will spell out how to do this in more detail in a few days, but meanwhile...
~~~~~o0o~~~~~
Stever95,
"I.....I......I...... Disagree."
The appropriate saying here is that "Opinions are like a...holes. Everyone has one."
Would you care to support your opinion with some quantifiable reasoning?
~~~~~o0o~~~~~
Troy,
"... intuition would suggest that combining lower inertia losses and a higher tractive potential for an equivalent power level would give higher acceleration, all other things being equal, no?"
Your intuition is wrong because your premisses are wrong. Your main mistake above is the second one. I'll give the numbers later.
~~~o0o~~~
Thijs,
"In other words, there must be enough grip off the line to lift your front wheels in the first place.
You basically agreed to that, by saying we should assume a start line acceleration of 2.5g."
Yes, as all the Toothless-Hillbillies know! And the "front-wheels lifting" kind of makes your front-motors redundant, doesn't it? (Except, as noted before, for regen-braking, and also good for very low grip conditions where there is little rear weight transfer.)
"You still haven't told me where you do your tire shopping."
I buy whatever is readily available, but then use it correctly. I note that Delft has some rather special looking Apollos? They look good. :)
"This is how sticky a drag strip is:
https://www.youtube.com/watch?v=x5wQajhGaeI"
And on that sort of drag strip (which is only like that near the start) a Top Fueler takes ~3.8 seconds for ~400m (Edit: STUPID Z!!! See later post). That is an AVERAGE of more that 5.5 Gs, over the full distance! I am asking you to aim for LESS THAN HALF that number, and only for a very brief period at launch (numbers later....).
"I don't remember seeing cars in a circuit race class lifting their front wheels.
...
If you want to make a point about how slow and boring current FSAE car designs are, and how stupid 4WD is, please point out a single race car that can go around corners ..., and that accelerates quicker to 60km/h than the fastest 4WD electric FSAE cars."
You really should get out more. Or use Google?
Last time we had this discussion all the student "experts" were saying that no racecar engine could ever make the 30 kW/sq.cm of restrictor area that I said was a reasonable upper limit. I then had to spend five minutes on Google to find the countless engines that do, in fact, make close to that number, with some claiming quite a bit higher.
You are being LAZY. Please do your own research.
~~~~~o0o~~~~~
Daniel,
"Regarding the weight distribution ...
[Premisses ->]
If you only look at acceleration then yes, you want to design a static weight distribution as close to the driven [rear] axle ...
However, there are 4 dynamic events in formula student, where 3 out of 4 require some sort of lateral force creation. ...
With minimal load on the front axle, minimal total lateral force is created needed for cornering. ...
Then you have the issue of balancing the car for these 3 dynamic events. ... slight hickup ... oversteer ...
[Conclusion ->]
So yes, [with 50:50 WD] you take a slight hit on 75 points from acceleration, but you gain a lot more on 50+150+300=500 points."
Try not to take this personally, because the majority of your fellow students also think the same way, but the above reasoning is really STUPID.
Firstly, your "theoretical" reasoning is based on vague, qualitative, hunches. There are NO NUMBERS! If you do your calculations/simulations correctly, with NUMBERS, then you will see how wrong the above reasoning is.
Secondly, you have clearly not done any "empirical" research (ie. checking the "prior art", aka "benchmarking"). If you did, then you would be aware of the countless racecars that have a significant rear weight bias, with larger rear-tyres than fronts, and can BOTH accelerate hard, AND go around corners fast. See below for examples.
Sadly, your sort of reasoning is all too common amongst people who like to call themselves "Engineers" (I have heard it too many times in the past). NO correct use of theory, and NO checking of the empiricism. But worse yet, NO shortage of arrogance along the lines of "I are an Engineer, so I know all about these things. And if I can't do it, then it can't be done!".
~~~~~o0o~~~~~
Steven,
"[Maryland car]... as far as lateral force generation goes, if I remember correctly, there weren't many SAE cars faster around a skid pad, or scca autocross.
... transient response and the length required to get the weight distribution could be improved with a custom, short engine/driveline ..."
Yep. It works well, and squashing it all up and making it a bit lighter would work even better...
~~~~~o0o~~~~~
Jay,
"Minimal axle loading may create minimal lateral force, but there is less mass to move. See the Deltawing concept."
Again, yes.
Ben Bowlby's DeltaWing concept is driven primarily by the goal of half-mass + half-power = half-fuel-consumption + same-laptimes. To a large degree this is achieved via better aerodynamics. But along with the aero (or as a result of it?) the DW has ~28F:72R weight-distribution (with 75% rear aero), and it goes around corners just fine!
Interestingly, before it ran all the Engineering experts said "IT WILL NEVER WORK!!!". They said this even though the earlier 1980's AAR Eagles that inspired the DW worked really well! Ah, ... experts! :)
~~~~~o0o~~~~~
Bottom line for now, next year's FSUK C-cars should aim to win Acceleration outright, and push the E-cars off the Overall podium. So stop carrying on like sooky little girly-boys and wingeing that the E-cars are not playing fair... :)
The empirical evidence is out there, and numbers coming soon...
Z
Z, it is noteworthy that both the electrical cars in the top 5 had problems getting a good time in endurance. Zurich got 6 laps in which they lost 30 secs per lap (3 minutes, equal to about 90 points on FSUK) and Delft also lost about 2.5 minutes (equal to about 70-75 points) due to some problems.Quote:
Originally Posted by Z
I can't get this image out of my head when visualizing Z's weight distribution concept
http://images.thesamba.com/vw/gallery/pix/1025010.jpg
Low CG (boxer motor)
High rear weight dist.
No unnecessary technology
Don't take that as an insult to the idea by any means either. As an FSAE alumni with a few rotary motors laying around and an adapter kit for a VW bug box... craigslist is a dangerous place for the post FSAE junkie.
Z,
May I suggest spending more time around the drag strip when the track is being prepared. I have worked at a multiple-time record-holding (both ET and MPH) drag racing facility for 8 years, the track surface is sprayed the entire quarter mile in PJ1 Trackbite, per NHRA guidelines.
I'm not buying the Top Fuel comparison. I am personally interested in seeing your math that says an FSAE car on an unprepared surface can accelerate at half the rate of a Top Fuel dragster with a power/weight ratio >5200 hp/ton on a much higher bite track surface.
ADDED: Also, the quarter mile record is 4.58 seconds in a wheel-drive car as seen here (*Language Warning* http://youtu.be/57-DDGblOJw), you are most likely thinking of the 1,000-foot record of 3.7 seconds (http://www.nhra.com/points/national-records.aspx). Unless you're thinking of the 3.58 quarter mile set by a Hydrogen-Peroxide rocket car at Santa Pod a number of years ago (http://youtu.be/0lnEjOuQGBc), that really isn't relevant 'cause rockets aren't FSAE-legal.
To quote someone I highly regard on these forums:Quote:
Originally Posted by Z
I would like to see support from your statement that the combustion cars are NOT doing accel "properly". On top of that, support that the electric cars ARE doing accel "properly".Quote:
Originally Posted by Z
For the people that are not swinging their lower members around here....
With reference to the proposed rules I don't think anyone has bothered to reference to the 2014 rules. Throttle only seems to affect forced induction applications.
Check page 66: http://students.sae.org/cds/formulas...fsae_rules.pdf
T15.9 Throttle Body – The committee is considering changing the position of the throttle body to place it
downstream of the compressor on turbocharged and centrifugally supercharged engines. The restrictor would
remain upstream from the compressor. Naturally aspirated engines would not be impacted by the change.
Drive by wire includes switch (hopefully not, because we already have a driver kill switch)
T15.7 Drive by Wire Throttle – The Committee is considering that drive by wire throttles can be used on
Formula SAE cars if they include a form of the brake plausibility device which is currently required for
EVs. Feedback on this topic and whether you would like to adopt a throttle by wire throttle would be
appreciated.
As a student from Kettering, and having by far the most cost efficient car, this is in my interest.
T15.10 Design Event – The committee is considering including the objective of value in the design event
objectives. This is to make it clear to all participants that a cost effective car which is well executed should be
able to score well in the design event. This will result in the design event being judged on the three main
objectives of Design for Performance, Design for Value and the knowledge of the team members. The
committee hopes that this will change the perception of the design event so that it is clear that a large budget is
not a prerequisite to winning the design event.
T15.11 Cost Event – The committee is considering a major revamp of the cost event such that it addresses
product / component engineering issues including design for cost, design for manufacturing, design for
sustainability and the life cycle of the product. Students will be expected to have an appreciation of all areas of
relating to product / component engineering which will be important in their engineering careers. The
committee would appreciate feedback and prop
I'll bite on on this one MCoach.
Would a kind person please explain why engines with different types of induction should be treated differently with regards to how their power is being limited?
"T15.9 Throttle Body – The committee is considering changing the position of the throttle body to place it
downstream of the compressor on turbocharged and centrifugally supercharged engines. The restrictor would
remain upstream from the compressor. Naturally aspirated engines would not be impacted by the change."
Jay, glad you've asked. I went looking and dug up another relevant document.
From this 2013 document, this is the exact reasoning from the rules committee:
http://www.fsaeonline.com/content/Ch...tion%20web.pdf
"B8.6.1:
Moving the throttle body downstream of the centrifugal boosting devices will avoid
creating a vacuum in the compressor housing when the throttle is closed which draws oil
past the seal resulting in visual engine smoking and potential plug fouling. No
performance benefit will be gained by centrifugally boosted engines as any air leaks in
the throttle body will reduce the amount of air going into the engine at wide open throttle.
The restrictor will continue to choke the flow through the compressor limiting power as
with the current configuration.Placing the throttle body downstream from the turbocharger will put it in the normal
position used in current production vehicles thus making the turbo a realistic design
option for the teams. Turbocharging of production vehicles is widely expected to grow
significantly worldwide in the near future so providing an opportunity to incorporate this
technology into FSAE performs an important part of the educational purpose of the
competition.
Naturally aspirated engines would keep the same layout (sequence) as currently
required because of concerns that a leaky throttle body downstream could allow
unregulated extra airflow to the engine if it was placed downstream of the restrictor.
Positive displacement boosted engines would continue to use this layout because oil
seal leakage is not an issue and there is no motivation to change the design."
Thanks MCoach.
I don't see how a 'leaky throttle' is a defense though. At FSG the scrutineers will put their hand over the intake to physically plug the engine. If the engine doesn't immediately die, they will ask you to go full throttle as they plug the intake. This should make the engine stop faster... unless there is a 'leak'.
This same process could be performed no matter where the throttle body is located within the system and no matter what kind of induction the engine has equipped.
Why not simply place the restrictor after the throttle body and the turbo?
Oh Jay, think about it :-)
You want 300 horsepower turbo FSAE cars?
On second thoughts, sounds good to me :-) :-)
Pat
The restrictor functions on a difference in pressure differential. Usually it is referenced to atmospheric pressure, which then defines your maximum flow rate compared to the vacuum drawn by your engine. If you move that downstream of the turbo, you then end up in a situation where the restrictor no longer is limited by the vacuum, but by how pressurized you can get the air to force it through that tiny restriction. Not regarding efficiency and eventually melting turbos, power becomes unlimited. Essentially, your mass flow rate is no longer limited, which makes it an easy decision to run a turbo in that configuration over NA.
Turbo-609.9cc here I come. ;)
To circle back to the placement of the throttle, a leaky throttle in an naturally aspirated application allows more air to flow by circumventing the throttle while in a boosted application it would leak down presssure, decreasing mass flow rate, decreasing power. It actually seems pretty practical.
PS...
Please, don't go and prove what the downstream restrictor would practically restrict the engines to. I've had enough lower member swinging contests this week. Also, couldn't pass up a Powerthirst reference.
Could you imagine the top speeds in the straights? They would be hilarious! 5G braking into hairpins, all sorts of oversteer in slaloms. I would bet you could more than double the attendance at a Formula Student event if the cars all had 1hp per pound.Quote:
Originally Posted by Pat Clarke
MCoach - My point about the throttle body placement was more to show that it can be trivial to prove a 'leaky' throttle in either case. Differential pressure sure is exciting, but I don't think the sarcasm came across in my post.
Start here for large differential pressures through restrictors:
http://en.wikipedia.org/wiki/Rocket_engine_nozzle
P.S. Please keep your member to yourself.
Well, now. If I understood there was sarcasm in that post my response would have looked something like this:
downstream, eh? You mean we get to make formula cars with this much power:
http://www.youtube.com/watch?v=qRuNxHqwazs
1hp/pound would be very exciting.
JurrienK,
"... both the electrical cars in the top 5 had problems getting a good time in endurance."
Ok, the results pages didn't mention that. I thought the E-cars might have been conserving their batteries, hence both having very similar (but slowish) times.
~~~~~o0o~~~~~
Mumpitz,
Yep. More below...
~~~~~o0o~~~~~
Troy,
"May I suggest spending more time around the drag strip...
... the quarter mile record is 4.58 seconds in a wheel-drive car as seen here"
Yes indeed, I should get out more often. You are right, and I WAS WRONG!!! (And you have just brought eternal joy to legions of FSAEers! :))
Last time I attended a dragstrip the racing was ONLY 1/4 mile, and they only stickied-up the first 60' or so. I do recall the fastest times in the low 5s and headed for the high 4s, with sustained ~4-ish Gs.
I found a 1/4 mile (402 m) "record" of 4.42 sec which gives average 4.2 G, and your 1,000' (305 m) of 3.73 sec giving ~4.5 G. Given that the cars are travelling at half the speed of sound at the end of the runs, their Gs should be dropping off there, so must be higher earlier in the runs.
[Nevertheless, note to STUPID self: Z, do NOT trust Google! Must try harder... :)]
~~~~~o0o~~~~~
Dylan,
"I would like to see support from your statement that the combustion cars are NOT doing accel "properly". On top of that, support that the electric cars ARE doing accel "properly"."
Right and wrong ways:
C-cars.
======
1. The fact that times have been static for ~30 years strongly suggests that something is very wrong. In open competitions of almost any type, the times always steadily drop.
2. Watching FSAE Acceleration first in 2002, and again last year, ALL the cars smoke their rear tyres at the start-line, keep them smoking for the next 5 to 10 metres, but NEVER make a great deal of forward progress. Having seen lots of other cars accelerating hard, it is blindingly obvious that the FSAE "wrongness" is a lack of rear grip, due to lack of rear weight.
When I question a team about this the response is usually something like:
"Well, this year we turboed the engine, so of course the tyres are going to be spinning more than last year..."
Then another team member adds,
"Yeah, but next year we're going to add Traction Control, so that should fix it..."
To spell out the multi-level-wrongness here.
1. Start with nowhere near enough grip from the driving wheels, for a MEDIUM powered FSAE engine.
2. $pend re$ources to add even more power...
3. $pend even more re$ources to SWITCH OFF the power you added in step 2!
E-Cars.
======
1. They must be doing something right because they are a half-second, and half of the 75 points, faster than the C-cars.
2. On the other hand, this thrashing of the opposition might be quite "wrong". The general rule in motorsport is that as soon as you start winning too easily, all the other teams start whingeing and moaning to the Organisers, who then change the Rules to ban whatever advantage you have. Yes, FSAE is not supposed to be "motorsport", but this nobbling of the E-cars is (part of) the topic of this thread.
I would much rather see both C- and E-cars competing on their merits. Restrict both to the current 85-90 kW for safety reasons. Let the E-cars have their perceived advantage of 4WD, because IMO they have a possible (?) disadvantage of heavy batteries. Certainly, if the Enduro is made much longer, then the E-car battery mass will become a problem.
The two areas in the Rules where the E-cars really do seem to have an "unfair advantage" is Cost, which doesn't seem to reflect their expensive batteries (? corrections welcome), and Fuel, which reads like a fairytale!
Given that the E-car energy is likely coming from a coal-fired steam-engine, with umpteen transmission/conversion losses between there and the car, I honestly can not see how they can be rated as using less than half the CO2 of the C-cars. You can grow ethanol (E85) in the back paddock. Fully sustainable, closed-loop carbon-cycle, etc. But perhaps best to leave that argument for another time. :)
~~~~~o0o~~~~~
I'll give my Acceleration numbers tomorrow, promise...
Z
Meanwhile, here are some pics to go with Mumpitz's.
Typical F-5000 car (late 1960s to early 1980s). This was a low-cost, entry level Formula series, using inexpensive stock-block, but ~500hp, 5 litre V8s. Highish power + heavyish engine = more rear weight, => so also needs bigger rear tyres => so accelerated faster off the start-line and out of slow corners than contemporary F1 cars.
http://www.sportscardigest.com/wp-co...ayEsterer1.jpg
~o0o~
Typical F1 Turbo era car. Renault started it, but this first 1977(?) car wasn't too successful (took them a few more years...), and had nowhere near the 1,000++ hp of the later turbo cars.
http://autoinjected.files.wordpress....-renault77.jpg
~o0o~
Not really relevant, but interesting. F1 Tyrell P-34 "six-wheeler" (1976/7). Design was mainly to reduce aero-drag by tucking front wheels into bodywork. Only moderately successful (1 win, several other podiums...) because Goodyear (?) didn't do enough development of the front tyres. Avon now make tyres for these 10" front-wheels which has led to these cars winning many "Old-F1" races. I think (?) Avon also make these ~40 cm diameter tyres in the sticky hillclimb compound, possibly very suitable for FSAE?
http://media.crash.net/original/AU1091067.jpg
Williams and March also made 4R(tandem):2F-wheeled experimental cars, and Ferrari experimentally fitted 2 x tyres on each rear corner, similar to the rear-engined (~600 hp) Auto Unions of the 1930s.
~o0o~
IMPORTANTLY, all the above cars were designed when the Rules were much more open than now, and the goal was to build cars that go as fast as possible in a straight line, AND AROUND CORNERS. Most modern Formulae have the tyre-sizes, and hence also the weight-distribution, tightly specified by the Rules. Modern racecars do NOT look like they do because "it works better".
Z
If an open class will always have times steadily drop, you are implying that an FSAE car will eventually have a 0.000 second accel run. At some point it does begin to level off. There's a good chance we are already at that point (note, I am not saying we are, otherwise I would use data to back that up).Quote:
Originally Posted by Z
From my experince with accel tuning, it has proven to be advantageous to slip the rear wheels since the "hook up" RPM (for a 600cc motor) is ~mid way through the RPM range. Otherwise the car has a good chance of getting bogged down in a low RPM area that is strongly suffering from lack of power. Of course, this is operating on my team's car that does not have a huge rear static rear weight distribution like you think would be better (of course better for accel, but how will it affect the rest of the events?). For accel, my team does take measures to increase rear weight distribution and increase rear load transfer. Our poor results in accel this year were a cause of timing issues that caused us to lose a faster run (the rerun was then on cold tyres because we were notified after the driver got out) and hold us up in line which forced us to forefit our final two runs (with an improved LC tune). I do not have numbers on our accel setup (weight distribution and CG compared to our autocross setup), but it would be interesting to get those written down to discuss.
As far as Ecars being faster, what if the complaints are actually true? :). And you are right, the general rule in motorsports is if one team is winning, they must have an unfair advantage so they should get some sort of restriction. The issue is, especially for student designed and built cars, what levels the playing field in regards to performance? When you look at some levels of motorsports, it is absolutely ridiculous. Being that some teams may be the only ones running a certain car, if they are doing poorly, regardless of the level of their engineers and drivers, they can convince the rules to decrease their weight restriction or power output to "level" them.
FSAE ACCELERATION - INTO THE LOW 3s.
===================================
Below is a Velocity-Time graph that shows why the first few metres of the Acceleration event, the "Launch phase", is most important for fast times.
BTW, V-T diagrams are good for this sort of thing because the rate of Acceleration is given by the slope of the curve, and Distance travelled is area under the curve. The curve, of course, represents Velocity of the car at any given Time. So most of what you want to know (ie. A, V, D, and T) are all on the same diagram.
Some simplifying assumptions.
* "G" is conveniently rounded off to 10 m/s.s.
* The total mass of car and driver, including equivalent rotating inertias, is 200 kg. Why would anyone want more?
* There is an artificial Vmax "speed-limit" of 30 m/s (= 108 kph), perhaps set by the gearing of the car (or by the Rulemakers?).
* Maximum "drawbar" power of the car is 60 kW (~80 hp). This is the net power available to increase the Kinetic Energy of the car's mass, after driveline-friction (including tyre-slip) and aero-drag are overcome.
* Maximum distance required, from the FSAE Rules, is 75 m.
Four different "runs" are shown.
A - dashed line at left.
=================
This (unrealistic) run minimises the Time required for the 75 m, given the above constraint of Vmax. Perhaps the car got a "flying start", or else you parked it out on the main road, and a large truck travelling at 108.1 kph hit it from behind. Anyway, with the constraint of Vmax, the Time is minimised by squashing the given area of D = 75 m into a RECTANGLE, which gives an Elapsed Time of 75/30 = 2.5 seconds.
C - dot-dashed line at right.
====================
This run has a constant acceleration Ac = 1 G (= 10 m/s.s). Perhaps the car has tyres with Mu = 2 (ie. quite sticky), but because only HALF THE STATIC WEIGHT of the car is on the driven rear-wheels (so that, despite some rear-weight-transfer, the tyres are furiously spinning and losing that extra grip), the car can only accelerate at 1/2 x Mu = 1 G!
At time T = 1 second the car has travelled distance Dc = 5 m (... yaaawwwn ...), and requires power Pc = 20 kW to maintain this rate of acceleration (see graph for all these numbers). Much, much later ("... are we there yet ..."), at T = 3 seconds and Dc = 45 m, the car eventually becomes power-limited, with Pc = 60 kW. It has also just hit the Vmax speed-limiter, so then continues at 30 m/s to the finish line in an even ET = 4 seconds.
C' - dot-dashed extension to C at top-right.
==============================
Team Testosterone want to go faster, and they figure it is all about MORE POWER, ... and SPEEEEED! So they hyper-mega-turbo-boost their engine, and ... ahem, "fix" the speed-limiter. The grip stays the same (probably less, given tyres are now molten puddles), so with the same A = 1 G all the way they hit the finish-line at V = 139 kph, and ET = 3.87 seconds (= sqrt(15), from "distance = half-Aye-Tee-squared").
Despite needing almost 30% more peak-power, and having almost 30% more top-speed, the Elapsed Time drops by ONLY ~3%!
https://lh5.googleusercontent.com/-h...nchVTGraph.jpg
B - solid line around cross-hatched area.
==========================
Meanwhile, Team Toothless concentrate on the "launch" (see below for details). Doing so, they manage a very brief, but also very rewarding, initial acceleration of Ab = 2.5 G. This only lasts for T = 0.4 seconds, and covers a distance Db = 2 m (ie. less than the length of the car!), but it is where they win.
At the end of this 2 metre launch-phase the car is at V = 10 m/s (36 kph), and requires Pb = 50 kW to maintain this rate of acceleration. So the Gs drop off to only Ab = 1.5 G. At T = 1 seconds the car is at V = 20 m/s, has covered Db = 11 metres (I fudged the curve a bit here to keep the calcs simple), and has again hit peak power = 60 kW.
So the driver changes up a gear, and the reduced wheel torque means the car only accelerates at Ab = 1 G, requiring Pb = 40 kW (ie. engine is away from peak power). By T = 2 seconds the engine is again at its peak Pb = 60 kW, Db = 36 metres, and the car has hit the Vmax = 30 m/s speed-limiter. So the car coasts at constant speed to the Db = 75 metre finish-line with an ET = 3.3 seconds.
This time would be a World Record for C-cars and would win most comps. I believe that an E-car once did a T ~3.2 s (memory???). Continuing the above "B" curve above V = 30 m/s, but still keeping the power limit below 60 kW, would reduce the ET to below 3.2 seconds, for an outright WR. More engine power, and a transmission that keeps the engine near that peak power, should allow an ET in the high 2's.
But I stress again that the initial ACCELERATION AT LAUNCH is most important for good times (ie. it squashes the D = 75 m area towards the left). And mega-power is NOT necessary for good points in the other FSAE events (ie. SP, AX, E), nor does it make a big difference here.
~~~~~o0o~~~~~
LAUNCH TECHNIQUE.
===================
So, how do you get those high Gs at the start-line?
Let's start by dismissing the idea of putting only half the car weight on the driving wheels, and then hoping to find some magical Mu = 5 tyres. That is just wishful thinking, so STUPID. :)
The obvious first step is to have the driving wheels carry as much of the gravitational downward Fz force of the car as possible. Assuming RWD, and depending on CG-height and tyre-Mu, this might ONLY require about 60% static-rear-weight. Any forward acceleration causes "weight" to be transferred to the rear, which gives the rear-wheels a greater capability for forward acceleration, so more rear-weight transfer, etc. Generally, the higher your CG, and/or the higher the tyre-road-Mu, then the less R% you need, and vice versa.
(Note that with a 50F:50R Electric-4WD car, with typical CG-height, wheelbase, and tyre-Mu, during acceleration the rear-wheels might carry about 75% of the Fz loads, so require about 75% of power sent to them. Conversely, during hard braking the front-wheels might have to do about 75% of the "regenerative-braking". So the front-motors are 75% for "regen", but only 25% for "acceleration".)
But the BIG STEP to better launches (as all the THs know), is to have the CG of the car LIFT as it comes off the start-line. The only way the CG can move upwards is by INCREASING the road-to-wheel Fz forces. This Fz+ increase, which is above the static gravitational "weight", gives the car a greater forwards tractive Fx force capability.
Here are some ways to increase rear-wheel Fz forces:
1. Pop a "wheelie". Obvious really! Here the rear-wheels necessarily carry ALL the gravitational Fz loads (because fronts in air), and the UPWARD acceleration of the CG adds even more Fz loads. So much greater Fx forces, and more forward Gs. To maintain control you can let the front-wheels droop down so they stay on the road, while the rest of the car lifts. But remember that this launch-phase "wheelie" only has to last about 2 metres. Also, differential choice is quite important (and NOT so obvious...).
2. Set-up the rear-suspension with >100% anti-squat (ie. longitudinal n-lines sloping steeply up-to-front). This forces the rear-wheels down as the CG LIFTS. More grip = more forward Gs.
3. Do both of above. Basically, the higher the CG LIFTS, the more forward G capability. 4WD cars could have lots of anti-squat at both F&R suspensions, but eventually they run out of suspension travel, or all the weight is transferred to the rear anyway. Easier just with RWD.
4. DO NOT have a car with very high Pitch MoI. Dragsters are very long to increase their Yaw MoI, which slows down any Yaw motions. But they have frames that are flexible in bending so that the front of the car does NOT prevent the heavy engine and centre section from LIFTING upward. Fortunately, FSAE cars come standard with lowish Pitch MoI, but try not to let your "wheelie" go the full 180 degrees...
5. On the aero front, have a large undertray that, together with the lifting car, generates UNSTEADY aero forces that suck the car down. Note that your magical "Crayons For Drawing" are unlikely to help you understand these unsteady flows (which, BTW, can generate BIG forces!). Instead, quickly lift a large sheet of plywood up off the floor, and note how it "sucks" down. Also note that by the end of launch (ie. 2 metres!) the car is going fast enough to generate significant aero-downforce.
And a bunch of other stuff...
But as a last suggestion, look at the sprint races at the current Commonwealth Games. The runners squat down low at the start-line, then push DOWN and back as they come up and out of the blocks. Similarly with startled horses or antelopes, who always lift their front legs off the ground when accelerating hard.
Enough for now...
Z
Z,
I look forward to going through your post in more detail.
That being said, there are a couple things I notice:
1. I think your total inertia is quite off: 200 kg means a sub 300 lb car. Including rotational inertia in that figure is even more false, especially considering that your examples require a 4 cyl from what I can tell (>80hp).
2. For a typical CG, wheelbase, and 2.5G of accel, I'm not finding that there's a whole lot of weight transfer to the rears, 5% at best. Maybe I'm doing something wrong here..
But that would mean you'd need a larger % of the weight on the rear wheels statically. Or to raise the CG under accel pretty dramatically. Or have super sticky tires.
So I agree with your physics, but I don't think the realities are there to make this possible. Unfortunately I can't comment on the aero or CG raising techniques, I just don't know much about plausibility there. From what I can see... there's still a good case for 4WD in a straight line.
So I've ran your numbers Z. From the simulation the 130kg(!) car with 70kg driver with a mu of 1.5 (which is according to our tests quite viable) and a 60% CoG at 300mm above the ground. Without any drag but also no aero forces and driving constantly on the tire limit (like only electric cars can, since shifting and power and torque curves and such). With load transfer. No tire load sensitivity. The 2WD car completes the 75 meters in 3.705 seconds (stuttgart did 4.04 at FSUK2014, the fastest combustion).
Now, when maximizing the load on the rear tires without the car flipping over, for the same CoG height, we need 70% static weight distribution. Then the 2WD car completes the 75 meters in 3.545 seconds.
For it to get near the 3 second mark, the mu should be at least 3. And in that case you'd have to shift the CoG down or forward to prevent the car flipping.
I'm sorry if I missed something. But I don't think it's physically possible. The lightest car ever built was (I think, correct me if I'm wrong) the DUT06 at 120kg with a single cylinder (45bhp). I'm not sure what the lightest 4 cylinder (which has more power) was, but I'm guessing it's not less than 175kg.
On topic of the proposed rule changes, I guess this monday they are releasing the draft version which I think is quite exciting. I wonder how they will implement the rules for the rear wheel steering aswell, besides all the aero, power, etc. changes.
There are another couple of things to consider, some of which can be rectified by design, but which are not there today namely:
- It takes time to perform a gear change, during this time there is a period of no acceleration. This also occurs quite early in the run, which has a large effect.
- Secondly even though only modest power is required for the early acceleration it does mean that there needs to be an rpm differential between the rear wheels and the engine for the given gear.
The first and second can be solved by a CVT, the second also by using a clutch controlled launch phase rather than rpm based. I am assuming that most teams have a simple acceleration sim to run these sims.
I'm not sure I have been convinced that we have tyres that can produce a mu of 2.5-3 for a period of 0.4s at low speeds.
The line that is missing here is one for the electric AWD cars. In that case lets assume fairly normal tyres that can produce 1.5g accel throughout the 75m, although I have heard figures of constant 1.7g. The cars have sufficient power throughout the run, and have no need for gear changes this means 2.58s. The only problem with that is that it takes 90kW at the 30m/s barrier. That is not too much more than the mandated limit. So the car accelerates at 1.5g up to a little lower than 30m/s then continues with a lower acceleration. Not in a place to do the calcs, but it ends up with low 3s.
Kev
Just did the calcs ... around 3.1s (obviously close to accel times of the electric cars now). The big point here is that for these 4wd electric cars the tyres only need to provide a mu of 1.5 in all conditions, which is pretty reasonable. The power is controlled through the motor controller, which is the ultimate traction control. No gear changes to sap power. No need to play around with weight balance or COG height.
The power limit comes into effect after the first 1.8s (42% of the run).
This was done for a 200kg car. Change it to a 300kg car the time drops a whole 0.1s to 3.2 seconds (getting closer to where we are now) and the power limit comes into effect after 1.2s (62.5% of the run). The power limit for electric cars is effectively limiting how long you can have the extreme acceleration.
I have heard of 1.7g accel from the electric AWD drive cars, and obviously believable. In this case the run would be improved (with a 300kg car) to 3.1s, with the power limit coming into effect after 0.9s. (71% of the run)
So the AWD electric cars can do the ultimate times that Z suggests, but with a lot less effort, and on the power limit nearly the whole time. All the calcs here (Z's as well) are not including the effect of drag, which obviously slows things down, and will almost completely account between the difference of 3.1s to what the electric cars are currently doing. Also allow for a margin teams will be running to ensure they do not cross the 80kW limit.
Please note that I didn't crack open my proper sim code, and just did this roughly on excel. But I think the results are fairly clear, and close to expectations.
Of concern is that if the AWD electric cars are dropped to only 50kW of power they will do the run in about 3.45s, allow a couple of tenths for drag and we are still at 3.65s and they only need to worry about grip for less than 0.6s. Even with a big drop in EV power we will not be seeing the petrol cars faster than the EV's without doing a lot of work.
Kev
Lightest US car was UC Berkley, last year their car weighed 275 pounds/~125kg with a Swissauto 250 karting engine that put out 35hp.Quote:
Originally Posted by DMuusers
That is (with the current rules) more what you can expect. The lightest GFR/DUT singles were around the 138-145 kg with about 45-50bhp if I remember correctly. I know there are some more teams running the (W/Y)R450 and in the 140-150kg range which gives around 0.30 - 0.36hp/kg in terms of power/weight ratios. The value DMuusers mentioned for the DUT06 was at the time when side impact structures were there just for rules sake and not there to protect the driver. Which is lighter of course. The lightest 4cyl stuttgart cars (both quite light and powerfull) generate about 85hp at 185 kg according to their specs on the website, which amounts to 0.45 hp/kg which is even better. So the 60kW option would be more near the 170kg without driver than the 130s mentioned.Quote:
Originally Posted by tromoly
Or you just do the 75 yard accel like some years back at FSUK. That way you should be able to get in the low 3s on accel.
The 2015 rules are expected to release late August.
From a Facebook post:
A big thing I found surprising in my messing around with acceleration of a 4 cylinder was how little gears and wheelspin really affected times. We seemed to gear taller than most teams at FSAE-A with a 1st gear that ended at 78km/h and 2nd at 105km/h. The majority of our acceleration runs were between 3.95 and 4.00 seconds as we messed around with winding the launch control from being restrictive to barely doing anything at all. The fastest runs we put down were when we were dropping the clutch at 10,000 and holding a 10,000 rev limit until 35km/h. Hardly a well calculated map. We ended up somehow posting a freak run of 3.82 amongst all the 3.95 - 4.00 runs, but I'm not convinced it wasn't a timing error or just pure dumb luck that we parked on the oddly sticky part of the strip.
There were a few people that mentioned our fast gearshift times as helping a lot. We were running a 45ms cut on full throttle upshifts and only shifting once on the accel run. For the heck of it I did 2 runs at comp with no launch control and no gears. I started in 2nd, dropped the clutch at 14,200 and fishtailed my way to the end. We posted a time of 4 seconds flat. At best our "launch control" and use of another gear may have gained us 0.05 but I imagine you could get more by fitting a billet headrest to your car, which if you do the calcs still does jack all compared to static rear weight bias. Also by the time we get to ~70km/h, we're nearly at the end of the run. At that point even 300ms with no torque to the wheels is going to do bugger all to the time after we just spent a second or two clawing at asphalt trying to get our lump of a car moving.
As far as I'm concerned, so long as you have some sweet anti-squat, static rear weight, low driveline inertia and enough power to keep the wheels spinning for the first half of the run you'll win. Every bit of testing I've done with the R25B 13x7 Hoosiers tells me that for the most part more slip is more grip. I'd be inclined to guess at somewhere in the order of 30%-40% slip is where you'll make your max longitudinal.
Also for those that feel you need a light car to win accel, we weighed in at 240kg with a 50/50 weight bias. We simply aim to get the weight as rearward as we can, that's where we ended up with a "conventional" layout SAE car. Also we have a dyno printout that says 59.6hp.
That is an enjoyable read Menisk, and I tend to agree with the spirit of what you are saying. One should understand gear charts and spreadsheets, but just as valuable is to understand the weighting of that info, especially as tyres are spinning as you say.
"More slip is more grip" will ruffle a few feathers on this forums, and I love it!
I believe UTAS did well in acceleration in 2001/2002 with a very heavy car, but a pro had dyno tuned the 4cyl engine several times to get the most from it.
Dyno tuning for acceleration times... Hmmm, I get where your coming from, but, you tune some things on the dyne, the things that get the times are done on the track, not talking about launch/trac control either.Quote:
Originally Posted by Jonny Rochester
Just something to think about:)
I've just been trying to figure out why none of the "Acceleration Specialist" teams have just said the hell with it and gotten some Jr Dragster tires.
Cos they still need to finish Endurance on them? Also do they not need a reasonable amount of heat in them first? (they may not, but if they do....) That's something that will be difficult in series where tyre warmers are banned.Quote:
Originally Posted by AxelRipper
Regarding heat, I found this:
http://www.mickeythompsontires.com/s...?item=ETDragJr
"Requires little or no burnout". Plus they come in a FSAE-friendly 19/8-10 size. Finishing endurance (and takcling corners) is a whole different issue, and definitely a more serious one though...