In retrospect, I probably should not have offered the initial reference. The letter is overly detailed and makes further reference to documents not presented. Plus, it sounds a bit testy, probably because the author (me) was disappointed at what should have been a first draft of the Spec. Let me try again, beginning with an historical perspective...
We first discussed this subject with Arnie Kuhns of SFI in 2002 in Indy at the PRI show, which immediately followed the SAE’s Motorsports Engineering Conference where several good papers were presented on the subject of head and neck restraints. Arnie is a good guy, and he commented to Brian Butler and me that there was a lot of interest from sanctioning bodies for a standard for H&N restraints post Earnhardt, but that interest had dropped off. We speculated that this was probably because there was not an established measure of performance.
I mentioned that several thoughtful folks were drifting toward the concept of head load reduction (HLR), i.e. the percentage of head load that is reduced by a particular product. A simple concept, it ranges from 0-100% for a given crash protocol (direct frontal, offset frontal, lateral, etc.) and the higher the number the better. Using it drivers don’t need to bother deciphering ads containing copy such as “Our product reduced Fz loads to 3850 Newtons on the 50G, 35mph delta V sled at XYZ labs utilizing a male Hybrid III anthropometric test device! Buy now and get free shipping!!!” Who needs that nonsense? Drivers deserve to know how well something works—preferably in 25 words or less.
We completed testing in August of 2002, so when we were invited by NASCAR to test on the Delphi sled four months later we declined. It was the same setup with a slightly higher impact. The results were predictable (2,000N-2,500N upper neck Fz), so why bother? In hindsight, we probably should have tested there as the SFI Spec calls for testing on that sled, which generates about 70Gs.
To summarize SFI Spec 38.1 sections regarding crash testing, you need to conduct three tests at a nominal 68G peak load: two frontal yielding 3,200N or less load and one 30 degree offset at 4,000N or less load. These are easy numbers for an Isaac system. Honestly, after seeing 15+ years of data in which researchers have hit dummies at offset frontal/frontal impacts, this bores me to tears. Doing this stuff again is a waste of time. Dr. John Melvin, a big HANS proponent and all around good guy, has roughed up equations-—similar to ours-—that predict head loads in impacts beyond 100+Gs. So why do we ignore all the data from the 50G sled at Wayne State and go with a recent ~70G sled on which few devices have been tested? And what about the big honkers, the new 100G sleds? If we test the Isaac on a 100G sled and kick butt do we
not receive SFI certification because the Isaac is too good? Let’s get real.
But I digress. Our problem is not with the crash-test sections of the Spec-—that’s easy--it is with other sections.
So, which products meet the
performance requirements called for in 38.1? The Isaac system, the HANS device, the White device and the R3. The Hutchens device, apparently, does not. (I’m taking bets on this, BTW. I’m good for a beer.)
The interest thing to watch is the recently "SFI certified" R3 device developed by Trevor Ashline (another all around good guy), who developed the Hutchens device. If NASCAR says the Hutchens is out because it failed to pass the SFI test, must the R3 be in because it passed? According to NASAR, no. They want to think about it.
------------------
Gregg Baker, P.E.
Isaac, LLC
http://www.isaacdirect.com