Designing for a Differential |
Post Reply |
Author | ||||||
CLReedy21
Baja Godfather Joined: Nov/30/2008 Location: Marysville, OH Status: Offline Points: 736 |
Post Options
Thanks(0)
Posted: Dec/14/2009 at 10:53am |
|||||
Although it's pretty late in the year we're still in discussion at TTU over whether or not to run out '10 car as a "locked" car that we can open or an "open" car that we can lock up using an electronic locking differential. We can't lock and unlock on the fly so we're debating which position we should design for the switch to be in most of the time.
Near as I can tell the advantages of running locked over open are mainly increased traction over all terrains and less probability of getting stuck somewhere at the cost of increased understeer and the necessity of high roll stiffness in the car in order to combat said understeer. The advantages of an open car being the ability to simply drive through tight corners instead of having to throw a spool car in hard and the addition of low speed maneuverability where a spool car could not unload enough to break traction with the inside tire. This of course comes with the downside of possibly losing traction and becoming a "one wheel wonder". There are advantages and disadvantages to both, but considering the diff we are using is a clutch type limited slip I'm leaning towards an "open" style car with a super low cg but a relatively high roll stiffness to limit body roll and keep the wheels planted. I'm assuming that if we test and decide that that setup is absolutely horrid we could lock it up and add some rear roll control to go back to the old standby 3wheeler setup. We are trying to tailor our car as much as possible to one setup or another in terms of CG location, roll stiffness, and various smaller adjustments in the front suspension geometry. We are also looking at the possibility of cutting brakes for added maneuverability. So, teams with experience with differentials, how do you run your cars? Does it work well? What recommendations do you have for a team new to wheels turning at different speeds? |
||||||
-Chris Reedy
TTU Alumni Fourwheeler Drawer "Quick with the hammer, slow with the brain." |
||||||
adrive7
Baja Godfather Joined: Oct/19/2008 Location: Lancaster, CA Status: Offline Points: 711 |
Post Options
Thanks(0)
|
|||||
My first two years on the team we ran open diffs with mechanical lockers. While it was nice to be able to make tight low speed turns, getting stuck was pretty lame, and we generally ended up leaving the diffs locked out. My last two years we built our own gearboxes and didn't run a diff at all. Much simpler drivetrain, smaller box, and the car never gets stuck. It's just a bitch to push. Plus, you can run a single inboard brake setup in the rear.
|
||||||
-Joe
Chief Organizer, Baja California 2016, 2017, 2019 Ohio State Baja 2005-2009 Co-Host, Science... Sort Of http://www.sciencesortof.com |
||||||
jeiB
Baja Godfather Joined: Jul/17/2009 Location: Montreal Status: Offline Points: 604 |
Post Options
Thanks(0)
|
|||||
i think i will probably depend more on the competition you attend. A track like Montreal 08 would be much easier with a locked diff while illinois 08, oregon 09 would probably benefit more from a diff.
you can make a car handle well without cutting brakes. But cutting could maybe be used in the water maneuvrability to a greater effect....I dont know I would lean towards locked full time and open for dynamic events. |
||||||
Jeremie B.
McGill Baja Racing 2009-2011 Captain minibaja.mcgill.ca |
||||||
dillon_b12
Baja Godfather Joined: Nov/15/2008 Status: Offline Points: 781 |
Post Options
Thanks(0)
|
|||||
The diff. that we ran last year acted as a spool as long as you were traveling straight ahead but allowed slip in turns while always keeping both tires spinning some. Supposedly, it never goes more than 80/20 bias. It was not selectable like yours and it worked great for us IMO.
Will the diff. that you are using allow one wheel to be completely stopped? If that is the case, I would make spool my primary drive mode. I don't think you'll need cutting brakes. We did them in 2008 with our cam and pawl diff. and no one really liked them. We did, however, outfit an older car with a hydraulic e-brake and it works awesome for doing super tight slides/turns. Our 2009 car can power-slide better and more predictably than either our 2008 car with cutters or our old car with e-brake though. I suppose cutters would be nice for water maneuverability but I don't think you all have had a problem with that in the past so you can probably do without them. A lot of it goes back to driver preference. Some might prefer to be able to take those tight turns without having to throw the car in hard. Most of us are comfortable enough with our car to know if you throw it, it will stick and you'll be fine. So, even though we have the ability to baby it around those slow turns and let the diff. work, we go with the faster method of throwing it in hard and hanging on. One advantage to a spool is that if you snap a drive shaft, you can either limp in or even race as a one-wheeler. Edited by dillon_b12 - Dec/14/2009 at 12:29pm |
||||||
adrive7
Baja Godfather Joined: Oct/19/2008 Location: Lancaster, CA Status: Offline Points: 711 |
Post Options
Thanks(0)
|
|||||
Thats a good point. We finished Wisconsin on one drive shaft. |
||||||
-Joe
Chief Organizer, Baja California 2016, 2017, 2019 Ohio State Baja 2005-2009 Co-Host, Science... Sort Of http://www.sciencesortof.com |
||||||
OctoberSky
Welding Master Joined: May/20/2009 Location: Los Angeles Status: Offline Points: 123 |
Post Options
Thanks(0)
|
|||||
If you are going to use a diff, you have to design for it to be open and suffer the consequences when it's closed...not the other way around. In 2008 we ran an open, nonlocking diff with a suspension set up to keep the rear planted on both sides. I wasn't around at competition, but I have driven that car a bit, and manueverability is great, while traction on rough terrain is not (no news there). What is interesting is that last year we ran a spool, but our suspension guy decided for whatever reason to more or less copy the 2008 car's suspension....creating a situation similar to what is going to happen to you when you lock up the diff. The car handled well enough, but we added a swaybar to help with the understeer (13th in manuv. In Oregon).
What I am getting at is that the advantages of having a diff and a suspension designed for it open will outweigh the disadvantages when it's locked. Throw in a detachable swaybar or cutting brake on the front tires for the locked situations, and you should be set. IMO though, those would just be added complexities to help locked diff handling, when in reality the diff won't be locked for situations requiring good manueverability (maybe withthe exception of rock crawl, in which case you wouldn't want a swaybar anyway). Also, if you set up for a locked diff, when you unlock it for manuv, you're going to 3-wheel even slower than you would with the spool...even the best ATB diffs lose considerable efficiency when unloaded. Hope that helps |
||||||
UCLA Racing
2010-2012 Captain "we'll never win mini baja" -Cheddar |
||||||
paasch
Welding Master Joined: Feb/17/2009 Location: Corvallis, OR Status: Offline Points: 234 |
Post Options
Thanks(0)
|
|||||
I would recommend the default be unlocked. More on why below. Chris, I'm sure you've talked to our guys about how our car works, but I'll add what I can. Oregon State has been running an open diff since 2001, and a driver selectable locker since 2004, so we have some experience with these set ups. Keep in mind we run a heavy rear weight bias, similar to your past TTU cars. Also keep in mind that OS designs primarily for "west" competitions, so doing well in the rock crawl is an important requirement that drives our design. The rock crawl typically requires articulation, traction, and the ability to lock and unlock while on the course. You will have to maneuver on three wheels.
I disagree slightly with the "over all terrains" part. An open diff, and especially a limited slip may actually have more traction through a corner than a locked diff. To get the spool to turn through power-on oversteer you have to break traction with the rear tires. The open diff can help keep the tires planted, especially on corner exit. The other consideration is that most spool car designers push their cg forward to get the car to turn, so they're giving up some of their traction advantage by pulling weight off the rear wheels. You TTU guys with your past high cg cars have been an exception. Like you, we really like the traction you get from substantial rear weight bias.
With an open diff, OS has generally been pretty competitive in maneuverability. The team that consistently beats us is USF. I've posted this video before: http://www.youtube.com/watch?v=Jpr3RGAZlAw&feature=related showing the USF and OS maneuverability runs at Illinois 08. Watch the rear tire on both cars. Two very different ways to get around a tight course. Obviously, you'll lock up for hillclimbs, acceleration and sled/chain pulls. We never engage the locker during maneuverability and virtually never during endurance. The high speed cornering attributes of a diff in endurance are at least as advantageous as the tight turn attributes in maneuverability.
Most every FormulaSAE team, including ours, uses a limited slip. We also had a "manual" locker on our FSAE car last year for the brake check and for acceleration. We don't have limited slip on our Baja car. We have considered it in the past, but haven't found it necessary. We haven't tried a rear sway bar either, we haven't been able to get good kinematics and packaging over 12" of travel.
With a diff you can keep your cg low and back. We really like our rear cutting brakes. Especially effective with rear weight bias. We use it constantly in maneuverability and endurance, and sometimes it's the only way to get the car to turn on the rocks.
We've found it to work well. I really admire you guys for trying something completely different. Maybe we'll have to design a floater sometime. One that also will work in the rocks. |
||||||
Bob Paasch
Faculty Advisor Oregon State SAE |
||||||
m-neal
Bolt Sorter Joined: Oct/23/2009 Location: Ada, OH Status: Offline Points: 1 |
Post Options
Thanks(0)
|
|||||
To me cutting brakes seem counter active. When you lock up one tire you lose inertia. Since were power limited it seems more logical to keep the tire spinning and use the maximum available inertia to help power out of the corner!
|
||||||
-Mike Neal
Ohio Northern University |
||||||
blue2kss
Welding Master Joined: Dec/23/2008 Location: USF Status: Offline Points: 169 |
Post Options
Thanks(0)
|
|||||
As for my take,
I would love to try a differential someday. As everyone here knows, we have always had a locked rear, whether it was through the swingarm with the solid axle or the spool that we had last year. Like Dr. Paasch already mentioned, with the locked rear you must break traction with the rear to have power on oversteer. Take it with a grain of salt, but our motto was we would rather be traction limited rather than power limited (even though in all reality we still are). Breaking traction in off-road racing is not nearly as critical as it would be a paved road going projectile. We were able to rely on breaking the traction with the locked diff and our past Land Maneuverability runs and Susp. and Traction runs have shown that it works well for us. Granted, Rock Crawl is one of our worst events, except for that Illinois pebble crawl that we managed to get 3rd in. A diff with a locking option but normally open would be the best solution to the problem. I will admit that we have never designed the car with the rock crawl really in mind. We have always been willing to take our points hit on the event and focus on our favorite dynamic events and endurance that appears at every race. As for the CG discussion that Dr. Paasch brought up, our CG was (acccording to the model and measured by our LongAcre corner scales) to be 2 inches behind the drivers back. It was as centered as we could get it (as well as low) and we had no issues breaking traction. Pushing the CG forward no doubt would help to break the wheels loose either, but in our kind of competition with our terrain, vehicle weight, and power level we found it completely unnecessary to push it anywhere but as smack dab center as we could get it. We had the pneumatic ARB system last year for the adjustability and the fact that it was my first time digging into suspension design that far. It was nice to be able to tune it for any potential misgivings that my design might have had as well as each event and it worked well for us except the last race when we sprung a leak and ran the whole endurance race without it. Traction was still broken but not to the tune that Kevin was used to. I will admit, when it worked it worked very well. But trying to design the system with the fact that the gas is compressible was a bitch and we only got in the ballpark (hence why the adjustability was so important). We have relied on our drivers to be able to handle the oversteer that the car gives us, and we have been blessed to have had excellent drivers since 1999. I will be the first to say that relying on oversteer to get yourself around a corner is def. not always a good solution, but we like the results that we have had in the past with our typical amount. Oregon States approach (and everyone else's that differs from me) is a very valid one and I would have loved to been able to take their car around the track at any point during my tenure with our club to see how their differential felt. Maybe someday if I am up in that neck of the woods for training... |
||||||
Dustin Bride
University of South Florida SAE Alumni/Consultant Mechanical Engineer - Naval Surface Warfare Center, Marine Corps. Counter IED Development |
||||||
Waffles
Welding Master Joined: Jun/02/2009 Status: Offline Points: 110 |
Post Options
Thanks(0)
|
|||||
There are many cars out there which do not have reverse. Is there a benefit to having a one-way locking differential. Differential is the wrong term, this would be a system where you have a locking system which lets a wheel overrun the diff, but not let the diff overrun the wheel - thus it is one-directional. In this way, the wheels are not fighting each other for traction in a turn. Obviously if you try to do power induced oversteer, you will probably exceed the traction at the inner tire, slip, and then lock the outer tire in (which would be no different than a spool), but you would have a choice in the matter. A differential would do a better job at splitting up the torque so that it doesn't slip the inside wheel, but then you must have a brake or selection system to transfer the torque so that none of it goes to spinning a wheel.
I think you could do this with an uber agressive limited slip and still be one way. Your typical lsd adds to an open diff by putting friction plates behind the spider gears. Since under power the gears want to move apart, this activates the friction plates, but you are limited to the angle of the gears in terms of how quickly the gears move apart, and it is two way (forwards and backwards). If instead of gears you had ramps (very similar to the ramp on a CVT secondary - the saw tooth kind, not the hill and vally kind), the ramp angle would determine how quickly or aggressively the friction plates would engage. It wouldn't be a direct drive like a diff thou, for this to work, the center pin would engage the ramp, the ramp would apply force to a friction plate, and the friction plate would have to drive the wheel. But by doing it this way, you can make an uber agressive limited slip in the forwards direction, and in reverse (relative to the diff), it would sit against the backside of the ramp, and apply no pressure to the friction plate, allowing it to overrun the diff. Just a thought... Oh, and pushing the CG back helps with oversteer - remember you are saturating the sideforce of the rear tire. In a pickup truck, if you turn tighter and tighter, you will brake the front end loose (understeer) before the back end, since it has more mass; only if you apply power can you slide the rear end first. |
||||||
paasch
Welding Master Joined: Feb/17/2009 Location: Corvallis, OR Status: Offline Points: 234 |
Post Options
Thanks(0)
|
|||||
You don't have to lock the wheel. Our drivers have more control than that. They may lock a wheel in a really tight maneuverability corner with the front wheels at full lock. In endurance, they typically just slow the inside wheel down a little. Tractive force is what accelerates you out of a corner, not inertia. A spinning tire isn't generating the traction that a planted tire can.
|
||||||
Bob Paasch
Faculty Advisor Oregon State SAE |
||||||
CLReedy21
Baja Godfather Joined: Nov/30/2008 Location: Marysville, OH Status: Offline Points: 736 |
Post Options
Thanks(0)
|
|||||
You just described, more or less, a Detroit Gearless Locker. Looks like we may be heading down that path so that we can have our cake, eat it too, and not have to decide which one we want when. Your oversteer theory doesn't make sense. You might be adding to the side loading, but you're also increasing the normal force on your tires which tends to increase the friction force... We did the whole super rearward CG last year. We still got it to oversteer, but that car has mega rear traction and requires it's beefy ARB to really slide well even on friggin wet grass. Your pickup hypothesis has more to do IMO with trucks handling like, well, trucks. Tall sidewalls, heavy duty suspensions, general lack of performance vehicle dynamics. I on the other hand always have trouble with the opposite problem, I can't seem to keep the rear of my Z71 planted, but maybe you have experience with Fords or something :) |
||||||
-Chris Reedy
TTU Alumni Fourwheeler Drawer "Quick with the hammer, slow with the brain." |
||||||
paasch
Welding Master Joined: Feb/17/2009 Location: Corvallis, OR Status: Offline Points: 234 |
Post Options
Thanks(0)
|
|||||
Nice post Justin. As jeiB mentions above, it really comes down to your design goals, especially what competition(s) you're designing for. Oregon State designs for the "west", and we've found a formula that works for that, and occasionally works at midwest style events as well. You guys at USF go to all three events, and so you design a car that will do well everywhere. TTU has traditionally designed a car for "east" events, and they have done very very well at those events over the last decade.
I was unclear. When talking about "pushing the cg forward" I meant relative to our (OSU and TTU) cars. We run 35/65 f/r mass distribution. I believe USF is closer more like 45/55, as are most of the other "power-on-oversteer" cars like Auburn, ETS, RIT, Sherbrooke, etc. TTU has run 35/65 in the past, and in fact from the Alabama data, TTU had the least % weight on their front wheels of any team at that competition. With that much weight on the rear it's hard to induce power-on oversteer. TTU has accomplished that by a higher cg and relatively narrow rear track. We don't try for power-on oversteer, we turn with a diff.
For the record, I'd like to acknowledge USF, the BajaSAE maneuverability champions: 1st 2006 Midwest 1st 2006 West 1st 2007 South Dakota 3rd 2007 Rochester (I include this as the only time OSU beat your time. I suspect a scoring error) 2nd 2008 Tennessee 1st 2008 Illinois 1st 2008 Montreal 2nd 2009 Alabama 1st 2009 Wisconsin
And I have no argument with teams that design for a spool, if we ever go to all three events we might go to a spool too. A big downside to our approach is the hit we take in the cost report for that transaxle. It also takes a lot of development to get it right, we're now on our 8th transaxle iteration. We're also extremely lucky to have the support of a great company, Linn Gear, 20 miles from OSU. Dustin, if you're ever in the Northwest let us know and we'll set a special drive day just for you.
Edited by paasch - Dec/16/2009 at 1:40pm |
||||||
Bob Paasch
Faculty Advisor Oregon State SAE |
||||||
paasch
Welding Master Joined: Feb/17/2009 Location: Corvallis, OR Status: Offline Points: 234 |
Post Options
Thanks(0)
|
|||||
Chris beat me to it, as he says you increase the normal force as well, making it harder to break traction in the rear. We have the ability to run both ways. With our locker engaged, our car turns like a pig unless you have a very low traction surface. It is very difficult to break traction with that much weight on the rear and the miniscule power we have. We also have trouble with the light front end, not enough weight up there to generate a good yaw moment.
|
||||||
Bob Paasch
Faculty Advisor Oregon State SAE |
||||||
dillon_b12
Baja Godfather Joined: Nov/15/2008 Status: Offline Points: 781 |
Post Options
Thanks(0)
|
|||||
On our car, the ARB is really nice for tuning for each event and, I think ultimately having some adjustability is important.
We took 3rd in maneuverability at Alabama with our ARB on the second softest setting. In Wisconsin, which was a much slicker maneuverability course, our first run was on the second setting and the rear of the car was loose the whole run. We dropped it to the softest setting and picked up 2 seconds. I suspect if we had unhooked it completely we would have gotten faster still. Even so, we were able to get 11th. With driver, we were 45/55 and I would classify us as a "power-on oversteer" car. |
||||||
CLReedy21
Baja Godfather Joined: Nov/30/2008 Location: Marysville, OH Status: Offline Points: 736 |
Post Options
Thanks(0)
|
|||||
So, not to stifle the thread but rather encourage it I figure I'll give a little update.
We're chosen to slam everything to the deck and run a multilink rear setup similar to OS and Sherbrooke, but modified to mount the shock onto the rear lower member. We've tentatively selected the Detroit Gearless as our weapon of choice unless something better gets invented between now and sometime next week. Drivetrain gets smaller and lighter and no longer has to incorporate outside controls so I'm stoked. I'll have to make a quick build thread at some point, but it's shaping up to be a real tight package that does everything we want. I'm currently in favor of at least keeping a sway bar in mind as we flesh out the final details in the rear so we have room for one if it becomes advantageous, but it looks like TTU is going with wheels that turn at different speeds in addition to moving up and down separately for the first time that anybody can remember. Hooray progress! |
||||||
-Chris Reedy
TTU Alumni Fourwheeler Drawer "Quick with the hammer, slow with the brain." |
||||||
Waffles
Welding Master Joined: Jun/02/2009 Status: Offline Points: 110 |
Post Options
Thanks(0)
|
|||||
If lateral force was linear to normal force, it wouldn't matter where the CG was, both ends would slide at the same time, given a constant speed and radius turn. However since tires are inherently nonlinear (especially in lateral force), as you load up a tire there reaches a point where you saturate the tire and it can no longer provide more lateral traction, even with an equivalent increase normal force. As such, the side with more mass will reach this point sooner, meaning for a constant turn, the end with more mass will slide first. This is all thrown out the window of course if the front and rear setups are different, or power is applied. If the front tires have much more camber or camber gain + roll than the rear, the rear will saturate sooner (given a 50/50 bias). Much of this is car setup, but concerning tire saturation, I disagree. CLReedy, looks like you guys are gonna have a very interesting TTU car! Edited by Waffles - Dec/16/2009 at 4:47pm |
||||||
jeiB
Baja Godfather Joined: Jul/17/2009 Location: Montreal Status: Offline Points: 604 |
Post Options
Thanks(0)
|
|||||
TTU always as an interesting car.
|
||||||
Jeremie B.
McGill Baja Racing 2009-2011 Captain minibaja.mcgill.ca |
||||||
Old Greg
Bolt Sorter Joined: Jun/24/2009 Location: Tampa, FL Status: Offline Points: 34 |
Post Options
Thanks(0)
|
|||||
Sure it can. More normal force always produces more lateral force, albeit in a diminishing returns sort of way. The effect I believe you're thinking of is a function of slip angle. |
||||||
USF SAE
|
||||||
blue2kss
Welding Master Joined: Dec/23/2008 Location: USF Status: Offline Points: 169 |
Post Options
Thanks(0)
|
|||||
And Greg nailed it on the head. I left my copy of RCVD at work otherwise I could tell you the page its stated on. Our CG was pretty much right in the ball park at 45/55 like you mentioned once the driver was in the car. The cost aspect was a large one because I must say that our cost report scores were helping the teams placement immensely. You can talk to any of our guys, but the cost report is seriously brought up daily during the design brainstorms and squabbles that we had and we make a serious effort to keep it in mind throughout the year. Hell, we even had our first top ten placement in an FSAE event with our cost report (9th). Dr. Paasch, at the RIT race we fell ill to the same wheel bearing debacle that 15-20 other teams did. But no doubt, your car held together and you guys performed very well. Regardless, we train every once in a never up there in Oregon and even if the driving doesn't happen I would just love to come see your facilities and the campus some day. Edited by blue2kss - Dec/16/2009 at 11:23pm |
||||||
Dustin Bride
University of South Florida SAE Alumni/Consultant Mechanical Engineer - Naval Surface Warfare Center, Marine Corps. Counter IED Development |
||||||
mauropinhel
Double Secret Probation Joined: May/11/2015 Location: Brasil Status: Offline Points: 2 |
Post Options
Thanks(0)
|
|||||
Hello guys, I am new in this forum. My name is Mauro and I am from a brazilian Baja team, I am wondering if is any possibility of not having a gearbox, and use the differential as a reductor. I dont know if it is possible, but I really liked the ideia of throwing away the gearbox to save space and weight and rely just on the diff to get our reduction.
Thanks
|
||||||
Pedro UFPBaja
Baja Godfather Joined: Mar/03/2010 Location: Brazil Status: Offline Points: 886 |
Post Options
Thanks(0)
|
|||||
Differential itself only allows the traction wheels to rotate at different speeds. It's the ring'n'pinon gear set attached to it what reduces. In the example below it's bevel gears, but can be any kind Good exemple of baja's single speed gearbox with diff: If the differential is switched to a spool, the reduction will be still the same.
|
||||||
Former Team Captain
|
||||||
easterracing
Bolt Sorter Joined: Jan/19/2014 Location: United States Status: Offline Points: 36 |
Post Options
Thanks(0)
|
|||||
NKU Has run an open locker for the last three years, and we absolutely abuse our cars beyond what any endurance track could ever provide. We almost always run open, and all drivers are trained to go locked if they get stuck. As long as you are moving slowly, or going in a perfectly straight line, we were able to lock and obviously unlock on the fly. Therefore, if you get stuck once (without traffic obviously, getting stuck because the person in front of you was stuck is a different story) go through it locked from then on. We can travel dirt creek beds and up rocky, muddy hills open, usually without a problem. Then again, we have a lot of normal force on our side...
|
||||||
Northern Kentucky University
2011-2015 |
||||||
Post Reply | |
Tweet
|
Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |