Group Buy Variable Manual Rack - Feel a little F1 tech in your hands

[RYU]

Dumb question.. what is "VR" again? Variable Ratio?

Also, my AP2 has a 14.1:1 steering ratio IIRC. The AP1 and CR have a 13.8:1. I don' t know what the VR range would be but I have a lot of seat time in both NSX and AP1/AP2.

Before I make further comments... In a quick and easy explanation, for slow speed, parking lot driving I assume the higher ratio of 20:1 is what matters for this conversation, right? If you can keep the slow speed ratio at or near 20:1 then i'd like to discuss more.

 

[v-tecpwr]

7)A new housing will put us way north of $5000+ maybe even $7K+ as more engineering time is needed. This is not an option for us. So in order to stay within the original quote, MVO will provide just the pinion and the rack. P2F/Applied will rebuild your existing unit to keep the cost in control. We still need to check the dealer for parts availability and we won't know 100% if your rack is rebuildable until we disassembly it. [/COLOR]

So it sounds like we will have to have a manual rack of our own for you to put the parts in. Is this correct?

I wonder if the guts are the same on an electric rack. Could we maybe put the new ratio in that rack also?

 

[stuntman]

2)The twitchy-ness(if any) of the car is more or less the driver's behavior not due to the VR so a learning curve of the new rack is a must and suspension change(ie, roll stiffness, springs) may be also.

I disagree that 'twitchyness' is more or less the driver's behavior and due to their inputs. There are numerous mechanical attributes that can make a car 'twitchy'.

8)Questions are welcome but we want to stay away from continuous, never ending what-ifs, subjective questions that we can't answer. Remember, some like rice, some like beef, you have a choice. Our guess is some will love this new ratio and some will hate it. You should go in with an open mind. You can always go back to the oem ratio but there is no refund on custom made parts. Think carefully.

The bottom line is: do you want a quicker rack made by the world's best and are you willing to adjust to its new behavior?


Tell us what you think about the new ratio before we take the next step.[/SIZE]

Are you building and optimizing this rack for your super-light 2,500lb car? How many people who paid to help fund this development have cars that light and are happy with quotes like this:

This rack will be made for quicker response and performance so if your car is heavy, it may not be for you as our car is at no-worry 2500lb. We trust MVO knows what they are doing and steering is all they do. We think its better to wait until our discussion with MVO is finished, then see if you have gain a piece of mind. If not, we are sorry.

While it's probably going to be impossible to know how much more a given ratio is going to increase steering effort. I think it's important to find a good compromise of a quicker rack and steering effort that benefits the most number of buyers, and not just take people's money to fund development for a rack optimized for really light cars - unless all the buyers have very light cars.

0.02

Dumb question.. what is "VR" again? Variable Ratio?

Also, my AP2 has a 14.1:1 steering ratio IIRC. The AP1 and CR have a 13.8:1. I don' t know what the VR range would be but I have a lot of seat time in both NSX and AP1/AP2.

Before I make further comments... In a quick and easy explanation, for slow speed, parking lot driving I assume the higher ratio of 20:1 is what matters for this conversation, right? If you can keep the slow speed ratio at or near 20:1 then i'd like to discuss more.

According to the graph, the VR would be QUICKER (heavier) on-center than stock, and as you turn the wheel more, the steering ratio gradually reduces down to the slower OEM ratio.

What you're proposing is a slower rack on-center (or near-OEM on-center), and quicken it up as you add more steering -which may be the best of both worlds, but is backwards to what is proposed. That idea should also prevent an on-center 'twitchyness' especially at high speeds.

 

[liftnot]

Dumb question.. what is "VR" again? Variable Ratio?
Yes.

Before I make further comments... In a quick and easy explanation, for slow speed, parking lot driving I assume the higher ratio of 20:1 is what matters for this conversation, right? If you can keep the slow speed ratio at or near 20:1 then i'd like to discuss more.
the oem slow speed ratio is at 20. To make the rack just slightly quicker defeats the purpose of this project.
its 14.46 on center(harder to turn @ zero speed) and gradually towards 20:1 at lock like oem.
front engine AP1/AP2/CR is at 13.8 and 14:1 and you know how it feels. (is that with P/S or not?)
FWIW, Miata is 15.0
Elise is 15.8
Alfa 4C is 15.7
Caterham is 8:1
We read it from reliable sources but don't kill us if its not. This is for reference only and we don't know if they are fixed ratio or VR.
We have driven the Miata and a Caterham so we know what that feels like.

So it sounds like we will have to have a manual rack of our own for you to put the parts in. Is this correct?
Yes. otherwise the time and money needed for engineering is out of sight. Much higher than we anticipated. If you have any ideas, please let us know. MVO also suggested retro-fit the oem rack is the most economical.

I wonder if the guts are the same on an electric rack. Could we maybe put the new ratio in that rack also?
we never had one apart to compare. If someone with both service manuals, please post it so we know. we would guess its "probably not" transferable but we could be wrong.

I disagree that 'twitchyness' is more or less the driver's behavior and due to their inputs. There are numerous mechanical attributes that can make a car 'twitchy'.
we asked that specific question and that is their best reply. We did not argue at length, because it does depend on several factors of a given car (wheel width, tire type, track width, wheel spacers(some people), knuckle arm length, weight transfer, wt. distribution, suspension set up, etc.etc), They like to engineer the whole front but we can't go that route. we can stay in the "safe" zone. There is only so much we can do when we reverse engineer a complex part which will change the behavior of this car w/o messing it up totally and have it done right the 1st and only time is a tall order.

Are you building and optimizing this rack for your super-light 2,500lb car? How many people who paid to help fund this development have cars that light and are happy with quotes like this:
Yes on the 2.5k lbs and No on the other. this new rack is for the oem stock wt. range. we did not give them a specific weight. That would not be prudent.
Clarification: it just happen our car is 2500lb. it is not the criteria we provide to MVO for the rack design. It is for the common oem 3000lbs. range.

While it's probably going to be impossible to know how much more a given ratio is going to increase steering effort. I think it's important to find a good compromise of a quicker rack and steering effort that benefits the most number of buyers, and not just take people's money to fund development for a rack optimized for really light cars - unless all the buyers have very light cars.
Indeed, the ratio is not final yet and the balance of quick ratio and steering effort is tricky to satisfy all so if we can't achieve economy of scale, we just have to let it go.

According to the graph, the VR would be QUICKER (heavier) on-center than stock, and as you turn the wheel more, the steering ratio gradually reduces down to the slower OEM ratio.
Yes. The graph is up side down and we will ask MVO to make it right so its inline with the RE article to avoid confusion.
What you're proposing is a slower rack on-center (or near-OEM on-center), and quicken it up as you add more steering -which may be the best of both worlds, but is backwards to what is proposed. That idea should also prevent an on-center 'twitchyness' especially at high speeds.
This is what we are after. the problem is no body knows it 100% until its in the car and then its a either love or hate relationship with P2F. We are putting our neck out there so if folks think we are "taking your money", then just stay with oem.

Here is an article from RE from another Primer, it may help to explain it better than we can.

 

[BBVNSX]

7)A new housing will put us way north of $5000+ maybe even $7K+ as more engineering time is needed. This is not an option for us. So in order to stay within the original quote, MVO will provide just the pinion and the rack. P2F/Applied will rebuild your existing unit to keep the cost in control. We still need to check the dealer for parts availability and we won't know 100% if your rack is rebuildable until we disassembly it.

So it sounds like we will have to have a manual rack of our own for you to put the parts in. Is this correct?

I wonder if the guts are the same on an electric rack. Could we maybe put the new ratio in that rack also?

Well i have to make the same question i did in the past:

if we are talking about just the pinion and the rack why can't this be retrofitted on a EPS car? and RHD shouldn't be a major problem too...

The R&D is on the pinion and the rack developement, the remaining would be just to adapt this to the existing parts of the OEM steering box... am i just simplifying too much?

For EPS owners like me finding a non EPS rack is not that simple... i think the OEM part is discontinued...

Please see my original post here:

It can be a dumb question... or just simplifying too much... but i prefer to ask and be educated than to remain uninformed...

When changing the steering rack ratio we are talking about changing the two parts circled in red bellow correct?

So why can't the rack be modified and the rest of the steering shaft be replicated from the OEM part so that if we replicated the OEM steering shafts and redesigned the rack this could also be fitted on the EPS unit and allow the change in ratio but keeping the EPS on EPS cars?

Developing a new EPS unit would cost a fortune... but i think it is not about a new EPS unit, aren't we are talking about new steering shafts and rack to retrofit on the OEM unit?... would it involve a much more dificult task than producing this parts for the non EPS cars?

I would prefer not to loose EPS.

 

[liftnot]

^if we are talking about just the pinion and the rack why can't this be retrofitted on a EPS car? and RHD shouldn't be a major problem too...
we don't have an EPS on hand to look at. Its quite a bit larger. the internals are "probably" not the same?
MVO want to be sure the RHD is indeed a mirror image of the LHD. There were cars in their past ran into this problem. They want to be sure and so do we rather than be sorry.

^The R&D is on the pinion and the rack developement, the remaining would be just to adapt this to the existing parts of the OEM steering box... am i just simplifying too much?
Correct. on the pinion and rack only. They can certainly machine it so it will replace just the oem rack and pinion portion of the rack to keep the cost real. To make a completely new rack is really cost prohibitive. We asked and they said you don't want to know:redface: especially at such a low qty.

^For EPS owners like me finding a non EPS rack is not that simple... i think the OEM part is discontinued...
Correct. its hard to track down crashed NSX and hope the rack is straight. You do have the option of letting Matt to rebuild your EPS. He is very good at it. In order for MVO to do a feasibility study on retro fit the EPS rack will be a separate, new project and a different focus.

 

[stuntman]

Here is an article from RE from another Primer, it may help to explain it better than we can.

So to be clear:

You are designing and optimizing this rack for your 2,500lb car and not for the weight of most people's NSXs (2,800-3,200lbs)?

If you chipped in financially to help pay for this development, do you have a 2,500lb car and/or are okay with this?

-
-
-

You do realize the Racecar Engineering article supports RYU's recommendation of a slower (near stock) ratio on-center, and quickening it as you add steering lock -which is the INVERSE of what you have planned?



From that article:

"Broadly speaking, this is the effect you want, but unfortunately it doesn't happen fast enough or with the right profile to be useful. Ideally, you want the profile shown by the red line, with a high ratio, slow response dead ahead rapidly falling and becoming faster as steering lock is applied.

(****This is the OPPOSITE of what is proposed by Pole2Flag****)

Why? We approached Nevo Developments to find out. The company introduced its VR steering to Formula 1 three years ago and currently supplies several F1 teams. It believes there are four main reasons why teams are looking to VR steering technology.

Firstly, you ideally want to be able to tune the torque needed at the steering wheel for different conditions. With high-downforce racecars the vertical aerodynamic load on the front suspension is felt through the steering geometry as heavier steering. But this only occurs at high speed, either in a straight line or on fast bends where steering inputs are small. Under these conditions a driver needs high ratio, or slow, steering to lower the effort needed at the wheel.

Secondly, fast steering at small angles of lock can be hazardous for yaw response, making the car nervous at high speeds. Conversely, in tight, slow bends where steering inputs are large yet downforce is low a driver needs a low-ratio steering with a fast response. Not only does it reduce the amount of wheel angle needed to negotiate a chicane but it also speeds up the driver's response to extreme yaw situations, improving the chances of catching a spin.

Thirdly, it can reduce the number of turns from lock to lock. This is appreciated by the mechanics trying to guide the car into the pit garage and can also help with some design issues. Finally, in this age of power steering on racecars, careful management of the torque requirements of the
steering can help with the size and power consumption of the power steering system needed to get the job done. In some cases, it can eliminate the need for power steering altogether by reducing the highest loads the driver will experience.

All this means the ideal ratio profile for a motorsport steering rack is shown in figure 3. Compared to the constant ratio shown in blue, the VR profile in red has low steering gain in the straight ahead position, ramping up quickly to its maximum over the first 90 degrees of steer angle and to its maximum gain for the remainder of the travel."



I think you have the wrong approach to be making this to serve your specific car and not what will benefit the masses of people who will buy it, and especially those who helped you fund this, but I may be wrong and look forward to hearing from those who have.

What MVO's thought process on the inverse of what most people do with VR racks in motorsports? I think what you are proposing will only be good in a low speed autocross arena and will not be ideal for a car on a road course. I think you can kill 2 birds with 1 stone by following RE's explanation (and RYU's request) of a near stock rack on-center and quickening it up as you add steering. That's what most motorsport setups do and I think it makes the most sense.


0.02

 

[liftnot]

I see why it can be interpreted as "upside down". We'll ask them to graph like the RE Fig 3 to avoid confusion.
and No, we are not doing this just for our car. Its for the mass and the mass is very small.

 

[jwmelvin]

I see why it can be interpreted as "upside down".

Can it be interpreted another way?

 

[solidol]

My understanding is that for this proposed VR rack design, rack gain is function of {pinion-deg/rack-teeth-profile}
Rack gain say around +-90deg (or as per posted graph +-300deg) is a sequence G like G={g(-90)...g(-2), g(-1),g(0), g(1), g(2)...g(90)} which is projection of rack teeth design.

At low speed rolling resistance of the tire will be higher then say at 60mph. For simplicity let's not factor lateral forces as well as aero effects at different speeds. So straight line movement at constant speed (no braking, no acceleration, constant load due to particular aero setup). This VR rack is vehicle speed agnostic (opposite to say oem power assisted implementation) and will produce the same sequence G at 5mph as well as at 100mph.

If above is correct, can somebody explain how can this VR rack achieve different on center sensitivity/gain at different speeds?

 

[liftnot]

We will ask MVO to plot: Steering ratio vs steering wheel angle & rack gain vs steering angle like the RE articles and we have deleted the Gain vs Pinion rotation curve above to avoid confusion. Thx for you input.

 

[stuntman]

Both the RE article and the proposed rack curve from MVO use the same Steering Angle (degrees) for the X-axis and Gain (mm/rev) for the Y-axis...

Apples:apples and your MVO curve is inverse of the RE article, which calls for a slower ratio on-center, which is what you want.

 

[liftnot]

^ yes. we want the same as the RE articles. New graph will follow to make it clear.

 

[greenberet]

(****This is the OPPOSITE of what is proposed by Pole2Flag****)

[...]

0.02

I'm not sure whether it's just me but what Pole2Flag has been proposing has been very clear to me from the first post. The guy at MVO who made the chart was probably deciding between a "Steering Ratio" and a "Rack Gain" chart and simply slipped up. I look forward to the corrected chart.

If anyone is unclear what Pole2Flag is proposing, I suggest reading the RE article attached above. That's it.

Personally, I'm glad Pole2Flag is putting in the time and money so that we can maybe fit variable geometry racks and pinions into our NSXs.

 

[stuntman]

I'm glad they are doing this as well, but MVOs first graph is the opposite of what I'm interested in, what RYU is interested in, and what is stated in the RE article.

 

[liftnot]

We'll be back. Waiting for our turn at MVO. They can certainly reverse the curve so its 20:1 at dead center and 14:46 at EOT. They said the RE graph is hydraulic assisted. We are not sure why that would that be any different than a manual curve as there is no force plotted. Anyway, we need and will stay on course with their guidance so they may not be able to answer ALL your questions because we are on the clock and we don't want to pepper them continuously with tons of questions but not moving the project forward.

ie. When we remove 50lbs from the front, the steering effort at oem 20:1 is much less......seem very intuitive to us but we are still fuzzy MOV said otherwise. Perhaps there is more to learn about steering. The RE article on the last page showed some drivers perceive it differently. Its good to read that article a few times so you understand what you are getting into. We believe the concept is great and it will make our car more "up-to-date".

We are proposing making one proto for testing purpose at the ratio they recommend so we can invite Primers to have a test drive and feel if we love it or hate it. That is the most sensible and the only way to be sure and we need the qtywink: your help) to make this project come true. We'll see what they say.

Thank you greenberet for the RE article.

 

[RYU]

First of all i'm glad Tim is going thru all of this and spending a ton of his personal time to deliver a new product to the community. I'm also glad Billy and other folks are here to help bridge the technical discussion with real life examples. Thanks for the explanation @stuntman. It is crystal clear to me now.

In simple terms, I agree that the on-center feel of the NSX is appropriate for the ~3,000 lb car. The on-center feel for the S2000 is a little twitchy but it fits the personality of the chassis. Many of us who would be interested in this new NSX VR rack have likely pursued other lightweight mods that can yield ~2,800 or less in weight. However, with that usually comes wider front tires which adds for a "heavy" steering feel especially in 0-5mph conditions. Keep in mind the car was designed with light 15" diameter, 185 section width tires IIRC.

I would like to put forth 2 driving examples in 2 cars. Also, correct me if i'm wrong but it seems important to make the distinction between on-center ratio vs. off-center ratio. I understand that it's a graph.

Angeles Crest Highway - Medium to Long sweeping 45-90mph corners (calm down prime police)
NSX - The stock manual rack shines here. The car feels relaxing, engaging, and confidence inspiring for the most part. It could benefit by a slightly quicker rack off-center, but on-center it feels appropriate. The longer wheelbase helps here I think.
S2000 - The S2000 feels a little bit more "scary" at high speeds. It feels twitchy w/o downforce in the rear. Still fun but not engaging like the NSX where in these conditions the NSX begs for more more more.. the S2000 reminds you right away how twitchy it can be which makes me feels slightly nervous. I have to actively think about the unwinding of the steering wheel coming out of the apex. Let's not talk about alignment issues to keep things simple.
PREFERRED: NSX

Malibu - Short/Tight to Medium turns 20-45mph corners
NSX - The NSX gets tiring in the tight turns. It is not unheard of where i'd have to go hand over hand wrestling with the steering wheel. The turns are tight and the steering ratio is slow. With that said, the unaided manual rack also creates some significant whiplash if you're not careful. Requires significant muscle even with an efficient treat pattern in the PSS 225 18" front tires.
S2000 - The S2000 is a hoot to drive up here. The light chassis and the tight, power-assisted steering rack is entirely appropriate for the personality of the shorth wheelbase chassis.
PREFERRED: S2000.

For tracks like Streets of Willow which is probably the tightest track in these parts... even the NSX could benefit (IMHO) from a slightly tighter ratio off-center. However, due to the unassisted nature I really cannot imagine a situation where an S2000'ish off-center steering ratio could benefit the NSX chassis.. I would really like to hear Billy's opinion on this. Maybe i'm a minority here but I really like the NSX on-center feel. The rate at which the steering wheel tightens the steering ratio could benefit from being tighter though. Is that a 14:0? I don't know.
@liftnot - I want to see you succeed and I want to buy a rack from you. I hope my opinion helps with your decision to pursue this. However, i'm not an expert by any means. It's just my $0.02.

 

[stuntman]

Well said Regan. You don't want too much on-center response for that exact reason, high speed instability. I too agree that the NSX's on-center feel is quite good, but where the NSX requires 180* of steering input on most tracks (requiring shuffle-steering), most modern cars rarely go past 90-110*.

If we can keep the NSX's ratio from 0-30* or 45* then increase the ratio from 45*-on, it would be possible to improve the off-center steering response, reduce the need for shuffle steering, and reduce the requires steering input from 180* in hairpins down closer to most modern cars (~100-110*). I think this would be a win-win and follows the reasoning the RE article mentions and why they do that on racing racks.

0.02

 

[RYU]

Yes! I agree. Looking at the diagram below 0-30* looks like the sweet spot in the NSX that should be maintained. Let's ramp up the ratio after that.

 

[RYU]

It would really really really (x100 more) be nice to not have to take my hands off 9 and 3 o'clock for most spirited driving conditions in the NSX. Whatever that ratio is.. I don't know. I think it's around 15:1 or maybe slightly less.

 

[stuntman]

It would depend on when you want to start ramping it up. I am leaning towards making the distance the rack travels at 110* degrees to be ~the same as the OEM at 180*. That way you can keep your hands at 9&3 the entire time. Hopefully accomplish that won't increase steering effort too much.

Who knows, it's already physically difficult to catch slides. Maybe doing anything to the rack will make it physically impossible and any slide will result in a loss of control since it's too hard to catch it. I sure hope not and am keeping my fingers crossed that our goals don't increase the effort too much. Time (& testing) will tell. But I think we're on the same page RYU.

 

[RYU]

Thank goodness I'm not the only one who things that whiplash effect is hard to catch! once it goes it might even break my wrist if i'm not careful.

 

[stuntman]

Means you have to be an even better driver to 'recover' (steer the car back to where you want it to go once you caught the slide) sooner and more precisely to try to reduce the amount of steering effort needed. But even so, it's still physically difficult.

Hopefully this rack/ratio does not make this too much worse, or make it undriveable.

 

[liftnot]

Lets wait and see what MVO comes back with. We all want the same thing but the trouble is we all have different levels of expectations. All of us have many legit personal preferences, strong feelings/likes/dislike and obvious high expectations for this soon-to-be a very expensive part with only 65/10k+ owners who are interested but we know for sure when the price is final....that is the hard reality doing this kinda of project on a club level. FYI, We are the 1st "club" requesting this type of work that have gone as far as we have. Most turned away after the initial quote due to price but we see there is alot of value in the VR rack. This is not their first dance but it is ours.

Frankly we know very little about this area nor do we have the proper vocabulary to describe VR to MVO what we want other than KISS, we want a quicker rack base on the Bishop technology. Greenberet saved the day by providing the RE article so we have proper language for MVO to move on. It seems simple enough from the outside but there is a lot more to this and we are finding many things are proprietary. True or not, we don't want to know.

AFAIK, non of us have any VR steering rack design or manufacturing experience so we have to depend and trust the source. The more we learn made us realize how less we know about this area of vehicle dynamics. We should not TELL MVO what to do without a drawing, as that is typically how they work. we can only ask questions and trust them understand what we are saying to stay on the same page and we are willing to adapt to the outcome. Personally we think we will be pleasantly surprised with the outcome and willing to adapt and learn driving skills as the car will behave differently than before.

Reverse engineering of such a high precision part is very tricky as it affect the behavior of the whole car. most companies provide their own design or ask MVO to design for them base on vehicle engineering drawings that we do have. And the fact is, if we don't get it right the 1st time, you're going to feel it every time you drive and hate us even more than you hate Dali for taking your money even though we've "asked" we want what the RE article saz. IMO this is a big gamble from the beginning but this will transform the car.

As for now, some manual guys and the EPS guys are feeling blue in the left field due to the cost of the new housing design. If you feel we have robbed your money in any way, please do not allow us to have that privilege of doing it again as we move on to the 2nd phase.

 

[stuntman]

I hope no one takes my posts as pessimistic or negative. I am very interested in this project, am a potential customer, and I have been involved with the development of a lot of production cars and professional race cars. This type of development is what I do on a daily basis, and I also have the opportunity of working with some of the greatest minds in the industry, so this is right up my alley.

I'm obviously fond of the NSX, want to improve the NSX's steering properly, and just want to make sure that this project does not get steered off into left field and I would like to help in any way I can.