saxyman990

BACKGROUND:

Quite a few years back I decided that I would like to use a restraint harness to help keep me from sliding around in my seat during auto-x and HPDE’s. So I decided to purchase a Rotary Extreme harness bar. From the day it arrived at my house, I absolutely HATED the design. I contemplated ditching the RE bar and purchasing an M2/Raceshop bar instead. This bar is much beefier, and thus stronger, than the RE one. But it still carries many of the same severe design flaws. I will elaborate on this later.

There are many people that would argue a racing harness should never be used without a full roll cage. There are some truths and myths to that theory, but I will not be discussing it here. Note that I never use a racing harness on the street, and when I used one on the track, it was always in combination with the stock retractable seatbelt. Regardless, I found it nerve-racking (to say the least) to use a device intended to be a connection point for a safety restraint, that was clearly WAY under designed, and not up for the task.

I’m a degreed Mechanical Engineer, and I finally decided to find a little free-time to create a design of my own that corrects these issues that I dislike.



OLD DESIGNS:

Some examples of the inherent drawbacks of current harness-bars (RE, M2, raceshop) are listed below. Please note: this is not intended to be a bash-fest on these products. I am only trying to relay some of my reasons behind designing a different bar.

Improper eye-bolt use:
First off, a properly used eye-bolt is plenty strong enough for use in harness connections (as evidenced by their approved use in FIA). Most of the high-strength eye-bolts used for racing harness connections have a load capacity of around 3000 lb. What many people don’t realize is that this rating is for loading that is directly inline with the axis of the eye-bolt. When a load is applied at an angle, it significantly reduces the load capacity. In all of the above bars, the load is applied at 90 degrees to the eye-bolt axis. At this angle, maximum loading capacity is reduced to around 15%. This would mean that a 3000 lb eye-bolt in this configuration would fracture at a mere 450 lb!!! It doesn’t take a very large impact to create this amount of force (for example: given the right circumstances, this could be accomplished by hitting a wall at 20mph). Restraint device requirements laid out by DOT are far above what this setup can handle. As a side note, I’ve also seen quite a few roll cages that utilize incorrectly oriented eye-bolts.

Inefficient loading paths and improper design:
If you use an eye-bolt, the load from the shoulder strap is applied more than 1 inch above the main axis of the harness bar. Some individuals choose to bolt the harness directly to the bar, rather than use an eye bolt. This configuration eliminates the above problem of an improperly used eye-bolt, but it still places the load ˝ inch above centerline of the bar. In either case, the offset load causes a twisting-moment about the bar’s central axis. Thus, the bar must deal with 2 loads: a forward load from the shoulder straps, plus the twisting load due to the offset connection point. Due to the inherent design in these bars (and their anchoring methods), all of these forces become transmitted to the mounting bracket. This results in a severe and catastrophic combination of direct loads and multiple bending (twisting) loads being applied to the mounting bracket.

Sub-par mounting brackets:
The RE bar uses the stock Mazda mounting brackets, which quite obviously were never designed for these loads. The M2/Raceshop bar uses a much beefier mounting bracket, but because of the severe combined loads that I just discussed, it will still fail at a sub-standard threshold. The noted exception to this, with regards to the M2/raceshop bar, is in situations where individuals have the entire roll bar/cage. In this situation, the mounting plates are strengthened slightly with regards to racing-harness forces, and can more closely tolerate the required loads (although, not with a factor of safety that I would personally like to see).

Weakly designed bar (applies to RE only):
With regards to the bar itself, the RE version has poorly engineered ends resulting in a design that cannot handle the loads required by even DOT standards (let alone any standards set forth by any sanctioned racing body). Of course, this really isn’t an issue because, as was already mentioned, the stock mounting brackets are extremely weak and will fail first. The M2/Raceshop bar is a much stronger design. By itself, it can handle the necessary direct/twisting loads. But as mentioned before, failure can occur as the loads transfer to the mounting brackets (assuming that the eye-bolt doesn’t fail first, that is).



NEW DESIGN:

So after realizing the above points (along with many other smaller issues with these bars that simply annoyed me as an engineer), I thought it would be a fun project to redesign and build a stronger and better harness bar.

Pictured below is the 5th variation of my design. It is not officially complete yet, as there are still a few details that need to be adjusted. The components were all 3D modeled using Autodesk Inventor Pro 11. Initial stress and loading calculations were done by hand. These calculations were validated/verified by performing Finite Element Analysis with an extremely advanced program called Abaqus. The end result is a completely new system that can completely withstand and exceed all loading requirements established by DOT and FIA.


Modeled components being worked on with Inventor:



Quickly rendered image of completed assembly:



Couple of quick examples of some FE analysis being performed on the bar and mounting brackets (note: some details and information hidden for proprietary reasons):






Some features of my newly designed harness bar:

- CNC manufacturing. Computer modeled components allow for every part to be produced using CNC equipment, assuring high precision and quality.
- Completely aluminum construction. As mentioned before, this new design can handle far greater loads than anything that is currently on the market. The aluminum construction helps to keep weight to a minimum while keeping the strength at necessary levels. Total weight with all necessary brackets and fasteners should barely exceed 6 lbs. Aluminum also allows for various aesthetic options (polished, anodized, painted, powdered, etc).
- Central harness attachment points. The shoulder harnesses are attached to the central axis of the bar (notice the slots along the center in picture 2) which eliminates any torsional load on the bar. Unfortunately, this does require the plastic interior divider to be removed when the harness is installed. Personally, this is an okay compromise, as I usually remove this piece at the track anyway.
- Elimination of eye-bolts. Harness can quickly be installed and removed using high-strength quick release pins. This provides for a stronger attachment technique than even a properly loaded eye-bolt design, and is just as quick to attach/detatch. No unsightly eye-bolts sticking up when the harness is not installed, either.
- New method for anchoring bar to mounting brackets. This new design in attaching the bar features precisely designed mating surfaces and a large vertical bolt placed in double shear. Coupled with the central harness attachment points, this anchoring method transmits virtually no bending loads to the mounting brackets. Thus, all loads on the mounting bracket are direct (normal) loads, providing for a much stronger bracket.
- No interior alterations required. Interior component do not need to be modified for this bar to be installed. Even includes custom designed bracketry to attach the cargo cover.



I am currently working on getting a prototype constructed, which should be completed within the next 30 days or so (if everything stays on schedule). Stay tuned, as I’ll post more pictures and information as things progress.

I’d be more than happy to elaborate on the above info or to explain any concept/details to those who may be interested.


-Rob Bailey

BuffDaddy915

WOW...you can more then deff. count me in when this thing is done. Great to have some good knowledge now. THANKS!

gracer7-rx7

iTrader: (16)

That looks great!

One question - What are the rules regarding securing the shoulder portion of the belts to the harness bars? It seems like many sanctioning bodies require you to wrap the belts around these bars not use eyelets. Are pins acceptable?

turBRO240

iTrader: (1)

What forces are these designed for? (lbs.).... the DOT Standards for eyebolts are 5000lb.

Are the tests for load done on static or dynamic forces?... I dont think Abaqus calculates dynamic forces. Its true that the main point of the harness is a mounting point for the harness and the load is purpose built for one way force (the pulling of the harness) but if the harness bar is going to be mounted where the stock one is, the stock one is also a roll bar, therfore calculations for dynamic forces should be made because of the many directional forces that a roll bar is under as a roll bar.

Great job though... im really hoping you continue with this... im looking forward to it. But before you create a prototype and everything... i really think you should take into consideration what i said in the second paragraph (static vs. dynamic forces)

Take into concideration that i am only 16 and im just giving out suggestions that i think would make a big differnce.

- Artin

scotty305

iTrader: (2)

1. It looks pretty nice, and it's great to see that you're doing some analysis and calculations rather than just fabricating something and hoping it will be strong enough.


2. This bar is obviously beefier than the OEM 'strut brace' that it replaces. I'm assuming that you've taken lateral loads into account, but here's one more thing to consider: there might be slight differences in each car's mounting points, depending on how much use/abuse they've seen. I hope you've got a little bit of adjustability built in to your mounts.


3. Which harnesses will these fit? I had a 4-point harness in my Subaru, and it was great. I haven't installed a harness in my RX-7 yet, but I think I'll want to use 3" belts. Will I need to wait and buy custom-length harnesses for use with this bar?

-s-

Jack

iTrader: (2)

Would love to purchase one!

saxyman990

Thanks. Rules vary slightly depending on the sanctioning body. Generally they require individual attachment points with anchoring positions that are capable of handling 3000 lb loads. Grade 5 or higher pins are acceptable according to FIA, SCCA, NASA, SAE, and most other sanctioning bodies (as long as the system is designed properly). Of course your local tech inspector may be a jerk and give you a hard time. Many tech inspectors like to see wrapped harnesses because otherwise people could bolt them on improperly (as I alluded to in my initial post). A harness can still be wrapped around this bar if need be.

I am by no means guaranteeing that this design will be approved for all competition racing classes. What I am saying is that it's designed to meet (and exceed) all general required standards. The local tech inspectors will have the final say whether it is authorized for track use. It really should not be a problem for auto-x and HPDEs.

-Rob

7racer

can you make some apex seals????

bar looks great Rob!

saxyman990

Atrin,

First, thanks for the positive feedback and suggestions. I do this type of work for a living, and have been for quite some time. Hopefully, I can answer some of your questions, and shed some more light on some of your comments.

Pertaining to this comment: there actually are no DOT standards for eye-bolts. What DOT (in combination with SAE) does require is a minimum of 3000 lb load capability on each mounting point. I designed this bar so that it can withstand even greater loads than that at all 4 mounting points simultaneously.

Quite to the contrary, Abaqus is exceptionally good at doing dynamic and even non-linear analysis. We even use it at work to model harmonic resonance in jet engine compressor blades. The short answer to your question is that analysis and calculations were performed on both static and dynamic loading conditions (including, but not limited to, impact and deformation).

I guess you'll just have to trust me when I say that all of that was taken into account when this was designed. The purpose of what I stated in the first post was to help explain some of the design differences in a relatively simplified method.

-Rob

saxyman990

1. Thanks for the compliment!

2. Yes, I've taken lateral loads into account in the design. A small amount of adjustability and tolerance has been integrated into the design. This should allow the bar to adapt to any FD (assuming that the frame's not wacked from a previous crash, that is).

3. This should fit any 3" harness with standard bolt-in mounting ends. You would not need any custom length or special harness.

-Rob

saxyman990

Yes, but where's the fun in that

Thanks!

turBRO240

iTrader: (1)

Just out of curiousity... what is your "job title" and what degree do you hold?


and also.. let us know when you have the prototype

scotty305

iTrader: (2)

-s-

turBRO240

iTrader: (1)

-___-

XxMerlinxX

iTrader: (2)

This is his Kung Fu... and it is strong.

FDeez

I like it.

dclin

iTrader: (4)

Since I'm the only one here that has a budget, I'll be the one to ask 'what is the targeted price point'?

Any possibility of doing a matching multi-point front bar, ala Autoexe?

http://www.autoexe.co.jp/products/to...owerbrace.html

Your bar is gong to look terrific in a brushed finish. I'd get two, one to use, one to frame on the wall.

BobfisH

Looks very nice saxyman. I have harnesses in my RX7 that are currently bolted to the rear seatbelt mounting points. Not ideal but im sure its strong enough.

A piece like yours would look lovely im sure. Willl you be selling in when its done? I have your solenoid system in my car and im very happy with that.

BobfisH

Also, just the play devils advocate - have you looked at the strength of the suspension turret tops?

I was told to not even think about bolting my harnesses to the rear suspension strut tops, because they wouldnt be strong enough. I know the angles involved would be completely different and all, but have you done any calculations on this?

saxyman990

I was originally debating between polished and black anodized. But now you have me seriously considering a brushed finished..... hmmmm.

If nothing else, I hope to design a front bar that at least matches the rear one cosmetically. I didn't really plan on doing a multi-point for the front, but that is something that I may consider. I'll look into it.



As far as a targeted price-point:

At this exact moment, I am not sure if this will ever see full production. I suppose it depends on testing results for the prototype, and if there is enough interest to warrant small production runs (READ: if you would be interested in this bar then LET ME KNOW. The more seriously interested people that I know are out there, the more likely I'll be to actually manufacture).

If the bar does make it to production, then I am hoping for a selling point at or right around $300. Don't quote me on that though, as actual production and material cost may cause that figure to fluctuate wildly. But that is what I would be shooting for.

-Rob

saxyman990

Bob, thanks for the kind words.

The shock towers are actually much stronger than people give them credit for. It's extremely difficult (and time consuming) to do extensive analysis here without some of Mazda's data and specifics in hand. I did, however, accomplish some brief analysis on the shock towers. Enough to convince me that my design was more than adequate.

Keep in mind that even large auto manufacturers cannot accurately predict results from EVERY possible impact scenario. So there's no way you can place any kind of guarantee when it comes to this.

Obviously, anyone that is involved in competition track racing still needs a roll cage, which will further help solidify these mounting points (regardless of whether you use this particular bar, or install the harness directly to the roll cage). Needless to say, the use of just a harness bar, without the aid of a roll cage, should be limited to auto-x and occasional HPDE's. I will always give this same advice to anyone interested in purchasing this bar if it becomes available.

-Rob

saxyman990

I realized that I made an erroneous statement in my initial post. Since I can't go back an edit that post, I'll correct it here:

In the underlined section titled "Sub-par mounting brackets," I made a comment that the addition of a raceshop roll cage strengthens the mounting brackets. This is actually UNTRUE.

I have since learned that the example I was basing this information on had been slightly modified with welded and reinforced plates. Apparently, the standard raceshop roll cage does NOT include this, and therefore does not strengthen the mounting brackets with regards to shoulder-harness loads.

I apologize for the error

-Rob

Sgtblue

iTrader: (16)

I've had a WTB for an M2 bar up for sometime for AX use. And you had my undivided attention after "....better bar" for a couple of reasons. It seems like a great idea and if it works out, my paypal is ready.
One quick question.... I'm just an aging civil servant, and have no engineering background at all, so the data you posted is hard for me to make much out of. If this is a way off stupid, I apologize in advance. But just wondering about loads to the bar caused not by the harness but by a sudden shock from a side collision. And if CNC aluminum is the right choice for the mounting brackets for those reasons, since they tend to be more brittle than steel.

ferragame

iTrader: (1)

I'm extremely interested as well. I've been searcing for a harness bar as well but they are few and far between especially since M2 and RE are gone. I'd take one in a heartbeat at $300.

Terry


As far as a targeted price-point:

At this exact moment, I am not sure if this will ever see full production. I suppose it depends on testing results for the prototype, and if there is enough interest to warrant small production runs (READ: if you would be interested in this bar then LET ME KNOW. The more seriously interested people that I know are out there, the more likely I'll be to actually manufacture).

If the bar does make it to production, then I am hoping for a selling point at or right around $300. Don't quote me on that though, as actual production and material cost may cause that figure to fluctuate wildly. But that is what I would be shooting for.

-Rob[/QUOTE]

saxyman990

Don't worry, what you posted is definitely not "way off stupid."

I'll try to keep this as simple as possible: This bar is much stronger than anything currently available, including side-loading conditions (yep, that includes side impacts). Its designed to transfer most of these loads as effectively and efficiently as possible. Due to its inherent design, it's actually much stronger in a side-loaded condition than in a frontal impact. As mentioned before, there's absolutely no way to guarantee for EVERY possible scenario, but this bar accounts for much more of those conditions than previous harness bars.

Excellent comment about aluminum as a material choice.

In general, you are right, aluminum is more brittle than steel (for most forms of Al, that is). But the point at which it fails is not just a property of the material, its also a property of the product design. You can design an aluminum part so that it can handle a large amount of loading, which is what I have done here. If you want me to, I can elaborate on this further?

-Rob