When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Well, I bought the Penskes, and talked to Anze about getting them reworked. Should come out cheaper than that used FCM setup I was looking at.
Not to sound cliche but I am too math minded to just take STX and stiffen it up a bit. I’ve been developing the numbers I think I want for a while now, bouncing between books, Shaikhs spreadsheets, my own spreadsheets, and the measurements I’ve taken on my own car.
I will I’ll be running class limits of aero and will eventually knock out the rest of things to make it a ‘true’ DSP car. Since this will be a track dedicated vehicle I am of the opinion that as much work should be done as possible with the springs, and leave the sway bars to do the tuning of balance. From the math I’ve been looking at I seems the RX8 is a prime candidate for this approach since you can get the springs stiff enough to control a most of the body movement without going to the extremes of natural frequency. I’m more than willing to document my numbers and their effects along the way for anyone coming along later.
Well, I bought the Penskes, and talked to Anze about getting them reworked. Should come out cheaper than that used FCM setup I was looking at.
Not to sound cliche but I am too math minded to just take STX and stiffen it up a bit. I’ve been developing the numbers I think I want for a while now, bouncing between books, Shaikhs spreadsheets, my own spreadsheets, and the measurements I’ve taken on my own car.
I will I’ll be running class limits of aero and will eventually knock out the rest of things to make it a ‘true’ DSP car. Since this will be a track dedicated vehicle I am of the opinion that as much work should be done as possible with the springs, and leave the sway bars to do the tuning of balance. From the math I’ve been looking at I seems the RX8 is a prime candidate for this approach since you can get the springs stiff enough to control a most of the body movement without going to the extremes of natural frequency. I’m more than willing to document my numbers and their effects along the way for anyone coming along later.
Best of luck. Last year the car started at 550f/350r, pre-aero though but was a good starting point for me. MX5 hotkis front bar (27mm) on middle setting, oem rear bar. The approach was hoping in a pinch can go a bit stiffer fr, or a bit softer rear, and also had the option to go a bit softer rear (remove an endlink).
I have my ideas of a next iteration, but really need the aero on course to have proof of concept.
Best of luck. Last year the car started at 550f/350r, pre-aero though but was a good starting point for me. MX5 hotkis front bar (27mm) on middle setting, oem rear bar. The approach was hoping in a pinch can go a bit stiffer fr, or a bit softer rear, and also had the option to go a bit softer rear (remove an endlink).
I have my ideas of a next iteration, but really need the aero on course to have proof of concept.
That 550/350 split makes for a large spread in frequencies... did you notice any issues with that? That makes sense with what I’ve been seeing for having decent balance with those sways though. My initial plan was to go with 500/500 or 500/475 with the stock sport front bar and no rear bar, should be able to swap to an adjustable but comparable front bar if I need more adjustments.
I searched the thread but didn't see any discussion about motor and diff mounts. Has anyone had success eliminating wheel hop from launching on Hoosiers?
I found the following motor mount options:
- RX8Performance $265 aluminum 70A poly
- Black Halo Racing $395 aluminum ?A poly
- Race-Roots $200 stainless ?A poly
- Megan Racing $307 ?metal 70A rubber
Here are the differential bushings I found:
- Whiteline $36 ?A poly
- Powerflex Purple/Black $77/86 80/95A poly (street/race)
- Superpro $69 70/80A poly (street/race)
I'm sure I've missed some. Is 70A durometer enough to control things in the rear?
I would give up other parts in the budget and spend your money on good shocks. They are the #1 thing we did to my car. Next was having Jeff H help us dial them in
Also, you can run them with the stock front top hats. It's not ideal (some side loading) but we did for awhile. We now have spherical top hats front and rear.
What spherical top hats are you using what brand, and where did you buy?
That 550/350 split makes for a large spread in frequencies... did you notice any issues with that? That makes sense with what I’ve been seeing for having decent balance with those sways though. My initial plan was to go with 500/500 or 500/475 with the stock sport front bar and no rear bar, should be able to swap to an adjustable but comparable front bar if I need more adjustments.
Interesting, your numbers don't really jive with what I'd use. I don't use the FCM sheet either, so I guess maybe that's why?
My guess is with camber appropriate for Hoosiers (+3 deg front, ~2.5 deg rear) the suspension will probably exceed the 2 deg/g rule of thumb on asphalt.
If you're going for a DSP build that you'll take to Lincoln, you'll want enough margin in your setup to increase roll stiffness for Lincoln concrete. This is the approach I took with my setup.
For reference, I've seen a nationally-competitive STX RX-8 run a big 32 mm Progress FSB at full stiff (!) when paired with KW v3 coilovers, which has (at least in the front) about the same spring rate as you're proposing.
I don't think the differential brace is legal in SP.
Originally Posted by comebackqid
What spherical top hats are you using what brand, and where did you buy?
Our rears we custom made. Happy to share more details via pm if someone needs it. The fronts we bought from Guy Ankeny. Anze can also make them but Guy had a set ready to go when we needed them.
That 550/350 split makes for a large spread in frequencies... did you notice any issues with that? That makes sense with what I’ve been seeing for having decent balance with those sways though. My initial plan was to go with 500/500 or 500/475 with the stock sport front bar and no rear bar, should be able to swap to an adjustable but comparable front bar if I need more adjustments.
I had zero issues, not just zero issues, car behaved perfectly. Admittedly I didn't drive it on the street at all last year, and have bumpy lots with some jumps locally.
I came to this starting point based on Ken Tsang and Jeff's STX conversation about liking something like a 65% front wheel rate bias, and honestly scaling back a touch. I'm likely to try 650/400 next.
I got to a .604 motion ratio for the FSB, .509 for the RSB, .752 for front spring, and .855 for the rear spring. My friend and I were not careful though, but the car was acting the way I like so i didn't really second guess if i had the MRs right
That 550/350 split makes for a large spread in frequencies... did you notice any issues with that? That makes sense with what I’ve been seeing for having decent balance with those sways though. My initial plan was to go with 500/500 or 500/475 with the stock sport front bar and no rear bar, should be able to swap to an adjustable but comparable front bar if I need more adjustments.
I think you will want a much larger than stock front bar to gain roll stiffness without giving up bump compliance. Said another way, too stiff of springs to control the body roll results in loss of bump compliance, which doesn't happen with a big bar. That said, I think your spring rates are on the soft side for Hoosiers on concrete even if you were going to run a large bar.
We ran lower rear frequencies relative to front most of the season, but tested a slightly higher rear vs front near the end and we liked it. We have aero though.
Let me just say, I am by no means an expert on any of this, I just enjoy the math behind it and spend way too much time reading/analyzing things. I could be (and most likely am) completely wrong about 75% of what I say at this point, but I am going to say it anyways since it is the best of my knowledge and I'd love to spark discussion as well as maybe some trial and error that my pocketbook isn't directly funding...
Originally Posted by gigglehurtz
Interesting, your numbers don't really jive with what I'd use. I don't use the FCM sheet either, so I guess maybe that's why?
My guess is with camber appropriate for Hoosiers (+3 deg front, ~2.5 deg rear) the suspension will probably exceed the 2 deg/g rule of thumb on asphalt.
If you're going for a DSP build that you'll take to Lincoln, you'll want enough margin in your setup to increase roll stiffness for Lincoln concrete. This is the approach I took with my setup.
For reference, I've seen a nationally-competitive STX RX-8 run a big 32 mm Progress FSB at full stiff (!) when paired with KW v3 coilovers, which has (at least in the front) about the same spring rate as you're proposing.
Luckily, I have a site in my local region that is identical to Lincoln. I run Gulf Coast Region down in Florida, we use a helicopter refueling pad in Dothan AL along with Wiregrass Region. Every year we have a ton of SE people come in to run on the surface in prep for Lincoln, it's pretty awesome!
The KW V3s I believe have a 450lb/in front spring on them, so not quite as stiff. I have been looking at a bunch of people's math and they all come pretty darn close to the FCM spreadsheets out there, although I personally use one I've put together over the last year or so. I'm not going for 100% perfect the first time out, but after playing around with a few different cars (friend has an ASP FD RX7 that typically FTDs out events) I have really become someone who believes in "set the car up right (mostly by the numbers) and learn to drive it".
The fact they could get by with that stiff of a front bar definitely comes down to what they were running on the rear (I don't see how they couldn't have been running an equally as stiff bar back there with the V3 rear spring rate being so low) and driving style. I'm quite excited to start feeling this out and making improvements!
Originally Posted by mrazny
I had zero issues, not just zero issues, car behaved perfectly. Admittedly I didn't drive it on the street at all last year, and have bumpy lots with some jumps locally.
I came to this starting point based on Ken Tsang and Jeff's STX conversation about liking something like a 65% front wheel rate bias, and honestly scaling back a touch. I'm likely to try 650/400 next.
I got to a .604 motion ratio for the FSB, .509 for the RSB, .752 for front spring, and .855 for the rear spring. My friend and I were not careful though, but the car was acting the way I like so i didn't really second guess if i had the MRs right
Those MR's seem spot on for the swaybars I know there is some contention on what exactly the motion ratios for this car are, but I think that comes down to some people taking into account the shock mounting angle and all too (plus it changes as the suspension compresses, etc)
What swaybars do you run? I assume you run a decently stiff rear bar as well? Just observing what all is happening out there currently it seems people get kind of out of whack on spring rates then make up for it with sway bars to get the car back into balance. There are 100 ways to skin this cat, I'm just trying to do it the most efficiently!
I'm assuming you mean 65% front wheel rate bias as in Front Roll Couple percentage? That makes sense with what would make a pretty balanced (but understeers when overdriven) setup. Makes me a little more confident in my numbers, I would just like to get there mostly via spring, instead of relying on stiffer sways to make up the difference in balance.
Originally Posted by Tamra
I think you will want a much larger than stock front bar to gain roll stiffness without giving up bump compliance. Said another way, too stiff of springs to control the body roll results in loss of bump compliance, which doesn't happen with a big bar. That said, I think your spring rates are on the soft side for Hoosiers on concrete even if you were going to run a large bar.
We ran lower rear frequencies relative to front most of the season, but tested a slightly higher rear vs front near the end and we liked it. We have aero though.
Let me preface this with the fact that this car will be 100% autocross 100% of the time, so bump compliance is only an issue as far as keeping the wheel in contact with the ground and the tire loading happy, my butts happiness is of zero concern.
I'm not sure where you guys are on roll stiffness currently but I should end up in the 2000-2200 lb-ft/deg range with the setup I am currently considering... which roughly puts the roll at 1G as about 1.5deg. This should not be anything excessive by any means, and I am also of the opinion that body roll isn't a bad thing.
What has made sense to me as far as roll stiffness goes, as long as you can keep the body controlled enough you aren't getting excessive suspension movement out of its 'happy place' as far as camber curves go you are good. Transferring weight from the inside to the outside via a sway bar will ALWAYS reduce the overall available grip (all other things equal) since you gain mechanical grip slower than you lose it as far as pressing the tire into the ground is concerned. By this logic, allowing the body to roll to the extent that it needs slows that transfer and maximizes your time with closer to equal loading and closer to max grip. The downfall of that idea is when you have enough roll that you end up turning the car the opposite way before the body has settled at 'full roll' for that given turn, which can definitely be the case in autocross. This is why natural frequencies in the 2.2+Hz range are needed in most track situations, even higher with autocross depending on the slalom you're looking at.
I'm really excited to finally get to play with all of this in a real world setting. Hopefully the Penskes will provide enough adjustment I don't screw myself over and have to revalve them multiple times before coming to an acceptable solution.
So based on your math, what natural frequencies are you aiming for?
Also, my bump compliance comment was referring to race only use actually. Lincoln has steps in the concrete pads that can result in a too-stiff car skating over the surface due to lack of bump compliance... but we're talking 3+hz frequencies before you're to that point. I think your mentioned rates of 4-500ish will be far far below that point. Personally, I would find an RX8 with those spring rates, on Hoosiers, to be too soft and squishy and not able to transition quick enough for my liking... but then again I came from CSP Miatas so I'm used to cars that slalom insanely well.
Running a 550/500 setup would put me in the 2.1F/2.25R range for frequencies, which is probably a bit low for Hoosiers in an autocross setting, but there will always be room for stiffening it up a bit, although going upwards of 550 gets somewhat difficult with spring selection (big jumps). I would really like to see frequencies in the 2.3-2.5Hz range, but I feel it will be easier to have a drivable car out of the gate starting off a bit soft and working up than the other way around.
I get what you mean about wanting the car to transition quickly... I did a few events on BFG R1S's with my slapped together STX setup and it was insanely not stiff enough, I managed to *BANG* bumpstops in a slalom, decided to put the slicks up until I could get some real spring under the car.
Those MR's seem spot on for the swaybars I know there is some contention on what exactly the motion ratios for this car are, but I think that comes down to some people taking into account the shock mounting angle and all too (plus it changes as the suspension compresses, etc)
What swaybars do you run? I assume you run a decently stiff rear bar as well? Just observing what all is happening out there currently it seems people get kind of out of whack on spring rates then make up for it with sway bars to get the car back into balance. There are 100 ways to skin this cat, I'm just trying to do it the most efficiently!
I'm assuming you mean 65% front wheel rate bias as in Front Roll Couple percentage? That makes sense with what would make a pretty balanced (but understeers when overdriven) setup. Makes me a little more confident in my numbers, I would just like to get there mostly via spring, instead of relying on stiffer sways to make up the difference in balance.
I'm on the stock rear bar actually. And yes, Front Roll Couple %.
The fact they could get by with that stiff of a front bar definitely comes down to what they were running on the rear (I don't see how they couldn't have been running an equally as stiff bar back there with the V3 rear spring rate being so low) and driving style. I'm quite excited to start feeling this out and making improvements!
<snip>
I'm really excited to finally get to play with all of this in a real world setting. Hopefully the Penskes will provide enough adjustment I don't screw myself over and have to revalve them multiple times before coming to an acceptable solution.
The setup I referenced used a Progress rear bar, which was either stiffened or disconnected depending on the situation. Usually, it's at full soft. In reference to KW's having similar spring rates to your proposed spring rates, I consider a 50 lbs/in change to be "small" (about 0.1 Hz). Good for balance adjustment, not gross alterations to the balance of the car.
If Anze valves your shocks like he did mine, you'll be able to make pretty radical changes to spring rate and still have room to adjust, but at the same time you'll be able to dial in changes in spring rate as low as 50 lbs/in. They're pretty trick like that... I only have a small gripe with the valving on the front shocks, but it's only my preference, not anything bad.
The FRC% I have been looking at have been in the 61-63% range, which seems to be a decent place to start.
Any comments on roll stiffness? My CoG height numbers are ballpark at best, but even with a decent margin of error it seems I would be hardpressed to break 3 degrees at anything under 2G of lateral grip. That seems like an acceptably small number but I've never had any real world numbers to compare it to...
(The fronts are on the other workbench, one of the upper spring perches doesn't want to come off the shaft)
Got the Penskes in the mail and have started stripping them down to send off. They're in fair condition, going to have to replace mounting hardware and the spherical bearings though. The front tophats are machined aluminum with a spherical, I am pretty confident they are the Mazda Motorsports ones, I will have to call them and see if I can get sphericals through through them since they are machined to key onto the upper spring perches.
It came with NC tophats on the rear, with the standard pillowball upper mount. I know the Delrin adapter for the 2.5" springs will fit on the RX8 tophats as well, but since these were the NC length rear shocks am I going to have to stick with the NC tophats? They appear to be shorter then the RX8 ones. If using the NC ones, is there an issue with not having the upper tophat mount? (the one on the side of the housing)
I'm going to be giving Anze another call this week to work out what exactly they can do and discuss numbers, I'd assume they have or can get parts to replace a lot of the little stuff while they are getting worked on. Sadly it looks like I may need a reservoir hose or two. This set was supposedly off of an MX-5 cup car, and they've definitely been raced hard.
If the front top hats are from an NC, they most likely will not fit the RX8 (speaking from experience).
On the rear tophats, the NC top hats were too short when we tried them (would have bottomed out our shocks). They are 2" shorter than the OEM top hats. Ultimately we had 2" of unused compression travel, so when we converted to sphericals in the rear we shortened the OEM hats by 1", subsequently gaining 1" of droop travel while leaving 1" of unused compression travel as a safety margin. We DIY'd the rears (need a welder and some skills), but I'm sure Anze could work something up for you too.
Are you going to have Anze revalve the shocks too?
Lots of interesting discussion in here since I last checked in. I knew that I had posted about my roll couple percentage somewhere, but apparently it wasn't in this thread. It looks like we talked about it before, roflcopter, but I'll repost it here just to keep all the data in one place.
...what I've found is that my car is just on the pushy side of neutral at about 66.4% FRC, and going stiffer on the rear bar to 64.7% makes it pretty loose and softer on the rear bar to 67.7% makes it pretty pushy. To give a sense of how small this window is though, when I ran the original Rivals on the car, I'd run it at 66.4% FRC for my first run and my co-driver's first run, but once the tires were warmed up it would become loose, so I would soften the rear bar to the 67.7% FRC setting and the car would behave correctly again.
And for the record, I ran my STX car on concrete with R compounds on it a few times, and thought that aside from the OS Giken diff misbehaving pretty badly, the handling was pretty great. I think it could stand to be stiffer, but I wasn't mad at it at all.
And on top hats, I also made my own. For some reason I don't have a shot of a rear, but mine are very similar to Tamra's in that I welded a bearing cup onto the top of the OE mount, except that mine have a bolt-on retainer to hold the bearing in.
On my fronts though, I actually need to rework mine a little so that I can fit more tire inboard because I'm getting to the point where I'll rub on the top hat with the offsets that I'm running. This shot is a repeat from earlier in the thread, but you can see where there is a bit of rub just outside of the top hat, so with even higher offset, I'll have to cut as much thickness out of the top hat as possible. Or maybe I need to raise the car a teensy bit.
If the front top hats are from an NC, they most likely will not fit the RX8 (speaking from experience).
On the rear tophats, the NC top hats were too short when we tried them (would have bottomed out our shocks). They are 2" shorter than the OEM top hats. Ultimately we had 2" of unused compression travel, so when we converted to sphericals in the rear we shortened the OEM hats by 1", subsequently gaining 1" of droop travel while leaving 1" of unused compression travel as a safety margin. We DIY'd the rears (need a welder and some skills), but I'm sure Anze could work something up for you too.
Are you going to have Anze revalve the shocks too?
What didn't fit for the NC front tophats? Are the bolt patterns different?
I like the idea of converting the rears to spherical, and that should be pretty easy to do something like what you guys did (plenty of fab skillz available, for steel at least). Do you know what length rear shocks you are running? When talking to FCM in the past they recommended finding rear shocks that were sized for the NC as it allowed more usable travel on a lowered car, but I do not know if that was meant when used with the NC or RX8 tophats.
Yes I had planned on Anze doing the revalve work.
Originally Posted by Kennetht638
Lots of interesting discussion in here since I last checked in. I knew that I had posted about my roll couple percentage somewhere, but apparently it wasn't in this thread. It looks like we talked about it before, roflcopter, but I'll repost it here just to keep all the data in one place.
...
I remember that roll couple discussion previously, I still think it looks funky compared to other numbers I have seen for this car (and other cars). I feel like I'm more knowledgable now than I was when that discussion happened, and I'm going to take a shot at explaining that... With the varying quoted motion ratios for the front end of this car and no consistent statements of whether or not shock/spring angle has been taken into account when calculating frequencies, there could easily be a ~3% discrepancy in FRC numbers. The front end has been noted as a .82 MR in some places, .752 in others... I don't remember where I found those numbers but they are both added in the notes section of my spreadsheet from my past self. I'm willing to bet if I put my numbers in your spreadsheet and yours in mine, we'd end up talking on the same terms.
On that note, I plan on getting the car up in the air here very soon and taking all of my own measurements, I'm curious to see what numbers I get.
Continuing that subject, has anyone gone through the effort of measuring the CoG height on their cars? I did it roughly a while back for mine.
As far as fitment goes, I think I have plenty of clearance at this point with the wheel and tire. Are most of you running offset bushings for front camber? I had yet to modify anything with that. I can get about 2.3 degrees with the stock adjusters on my centered bushings currently IIRC. I'm running 18x10 +38 RPF1, probably will switch to 275 tires for added shoulder clearance. I think being able to make the alignment happier and giving up a little rubber will provide for a better performing car, hopefully.
With the varying quoted motion ratios for the front end of this car and no consistent statements of whether or not shock/spring angle has been taken into account when calculating frequencies, there could easily be a ~3% discrepancy in FRC numbers. The front end has been noted as a .82 MR in some places, .752 in others... I don't remember where I found those numbers but they are both added in the notes section of my spreadsheet from my past self. I'm willing to bet if I put my numbers in your spreadsheet and yours in mine, we'd end up talking on the same terms.
Interesting, I've never heard of the front MR quoted as 0.82. The shock angle is included in the motion ratio, if it's included. The formula is:
MR_controlarm * sin(shock angle) = MR_true
Where MR_controlarm = length_shockmount/length_controlarm (as a rough first estimate) and shock angle is the angle from the control arm to the shock.
For giggles, I plugged 0.82 and 0.752 in for MR_controlarm and MR_true, respectively, and got ~66 deg for the shock angle. That sounds about right for an RX-8, but I don't have my car handy to confirm this.
Obviously, wheel offset affects the motion ratio as well... that factor would be included in the length_controlarm factor.
As for the rear shocks, I could believe that the NC rear shocks are shorter than ones used for an RX-8. Depending on the length, this could be a benefit, so you wouldn't have to run a helper or tender spring to keep the spring converter in place. You'd need to take some measurements and run the numbers to see if that'll work for you.
The male post that needed to poke through the body shock tower mount didn't fit. Basically the meat around the spherical is a different size. A hole saw could have fixed it but that's not legal in DSP.
I'm not sure what our rear shock length is and they're back on the car, so kind of hard to measure. Too long to use NC top hats though, I know that!
I like having as much droop travel available as possible, so using helper springs is beneficial IMO. I want as much travel as is possible considering the lowered height of the car.
Had our first outing this weekend. Felt great to be back in my RX8! It was a NASA club so no relevant SCCA competition to bench mark against, but had a great time nonetheless.
The course was fun, long, and fast. Bumpy in typical New Jersey fashion (I think I caught air once). Car was a tiny bit loose but in a fun way.
While driving home from Atlanta and pondering life last weekend, I had a realization... and POOF! Just like that, all this suspension stuff became clear.
I won't go into the justification for it at the moment, I want to experimentally determine a few things (If anyone has used race scales for sale hit me up!) before I go too far off the deep end. BUT! I think this realization explains why the FRC and natural frequency numbers that are commonly quoted for the RX8 don't line up with most of the rest of the automotive world. Those values, after all, should be common among cars in general when talking about handling characteristics and such.
The first truth:
1) The rear motion ratio of the RX8 is 1.0.
And how I arrived at truth number 1:
2) Wheel offset does not affect the effective motion ratio at all.
Feel free to tell me why I am completely and utterly wrong, I'm going to find some scales and a free afternoon and figure out how far from the truth those two things REALLY are.
The shock bolts to the hub, so I think you're correct. It looks like the shock has a small angle of inclination such that it will remain roughly tangent throughout the arc of the hub.
03-13-2018, 07:40 PM
