Harness bar

[Series8217]

I recently finished building and installing a harness bar in my '88 Fiero. I designed it so no parts need to be removed from the car except the B-pillars, the stock seatbelt can still be used, and no holes need to be drilled in the chassis.

Note that I am not a professional safety equipment builder, and I do NOT recommend building your harness bar the same way I built mine! Use a professional race shop and install a roll bar if you intend to use a race harness. Race harnesses harness bars and roll bars are NOT to be used on the street.

Two holes midway up the B-pillar are uncovered by peeling back a rubber sheet:


I designed a bracket to go over these holes. I mock up all my brackets in cardboard, then trace them onto the material to be cut on a bandsaw, and bend it using my vice and a hammer or adjustable wrench:


This is a mockup of the inner/backing bracket that goes inside the B-pillar:


There is a hole at the bottom of the B-pillar that's big enough to fit my arm through to feed the inner backing plate into place. I had two M12 nuts welded to the backing plate so I don't have to fumble with nuts on the inside.

I bent the harness bar from 1.5" OD, 0.120" thickness 4130 "chrome moly" steel using a JD Squared Model 32 manual bender:


The "hump" clears the center console. Without the center console in place, the bar can be straight.



The brackets (inner and outer) were cut from 3/16" 4130 steel plate and bent in my vice:


I may add another M8 or M12 bolt in the 3rd hole, but I don't think it's really necessary.

If anti-rotation bars are desired, the main harness bar can be tied into plate bolted to the OE shoulder belt mounting points. Here are mockups of some brackets for this purpose:


The bracket shown mocked up in cardboard would be made from 3/16" 4130 steel plate and would be tied into the main harness bar with 1.5" OD 0.120" wall thickness tubes. This can help prevent the main harness bar from rotating forward when the shoulder harnesses pull on it in a collision.

When I have enough parts ready to fill a large batch, I'll probably have the harness bar powdercoated. Some race teams don't powdercoat or paint their roll cages or harness bars, so that it's easier to inspect for cracks. However, some surface rust will develop over time.

[crzyone]

I like the idea of a harness bar but not digging the bends. Do you use your center glove box often? I'd run the pipe right through it for safety sake. Or you could run some diagonals from lower on the b pillar and make it a bit stronger?

Since you road race this car, a full hoop and harness bar wouldn't be a bad idea.

I went full safety build with my GTR. Only downside is noisy interior with no carpet or sound deadner and you can't look cool getting in and out of a caged car lol.

[Series8217]

I like the idea of a harness bar but not digging the bends. Do you use your center glove box often? I'd run the pipe right through it for safety sake. Or you could run some diagonals from lower on the b pillar and make it a bit stronger?

Yes, I use the center glove box frequently. That's where I store my paperwork, GPS, keys, etc. There is no other storage compartment in this car except the trunk. I thought about just making a new center console, perhaps one with a horizontal storage compartment, but I didn't have time to do that before the last track day. It's something I may still do.

Since you bring up the issue of the bends making the bar weaker (it could be more likely to "fold" at the first bend after the shoulder harnesses), I'll go ahead and model it in Solidworks to see how strong it is. Running diagonals to the b-pillar would not make any difference. Look at the direction the force is applied in by the shoulder harnesses.

Since you road race this car, a full hoop and harness bar wouldn't be a bad idea.

I went full safety build with my GTR. Only downside is noisy interior with no carpet or sound deadner and you can't look cool getting in and out of a caged car lol.

Since I don't have a tow vehicle and have no place to put a tow vehicle or trailer (one of the many joys of living in urban Los Angeles county), I have to drive this car to and from track days and autocrosses. That's typically a 1-hour to 2-hour drive. It will not be getting a cage anytime soon, since that would just make it far more dangerous on the street. BTW I don't road race; just HPDEs and autocrosses.. If I was racing against other cars or trying to push this car 10/10ths of course it would receive a full cage.

I just needed a way to put harnesses in the car for autocross and track days, since there's no way to lock the stock seatbelts on a Fiero, and I kept falling out of my seat beyond 0.8 Gs (I could actually see the limit on my datalogger!). I think the harnesses and bar make the car safer than the stock seatbelts, but it's in no way equivalent to a roll cage.

I don't see the utility in adding a hoop without forward braces, and the forward braces would make it unsafe for the street.

[fieroguru]

That harness bar is pretty cool! I really like the simplicity of the installation.

[Series8217]

I like the idea of a harness bar but not digging the bends. Do you use your center glove box often? I'd run the pipe right through it for safety sake. Or you could run some diagonals from lower on the b pillar and make it a bit stronger?

Since you bring up the issue of the bends making the bar weaker (it could be more likely to "fold" at the first bend after the shoulder harnesses), I'll go ahead and model it in Solidworks to see how strong it is. Running diagonals to the b-pillar would not make any difference. Look at the direction the force is applied in by the shoulder harnesses.

Did a quick FEA with the geometry of my bar (bends in the middle and at the ends), and a bar that is straight in the middle, with 22000N applied to the shoulder harness area on each side (so driver and passenger in a 45G crash).. no significant difference in performance. In fact, neither the straight bar nor my bent bar is strong enough for such a severe collision. The factor of safety is between 1.0 and 2.0 for the span, but on the ends is below 0.25. It looks like 1.50" 4130 steel is simply not strong enough as a harness bar span in a 45G crash without a diagonal support, whether or not its straight. That jives with the NASA CCR which requires that the harness bar intersect the main diagonal, but allows it to bend aft-ward for seat room.

I'll play around with the model some more; make the applied forces more accurate in positioning, and see if additional bars on the ends, or gussets will reinforce it enough. Not that a 45G crash is otherwise survivable in a Fiero...

[The Dark Side of Will]

How'd you arrive at 45 G's?

Is that the "generally accepted" limit for human survivability under ideal conditions?

[Series8217]

How'd you arrive at 45 G's?

Is that the "generally accepted" limit for human survivability under ideal conditions?

I can't find the source, but I think it was either an FIA or SFI standard of some kind. According to this article, 100G accelerations have been survived by IndyCar drivers without serious injuries.

The Schroth harness installation guide cites 30 g as a test standard, but I don't think that's the same as a design standard.

Emphasis added by me:

What happens during a frontal impact:
This data is based on an optimised installation with an upright seating position during dynamic testing. It simulates a 90° head on collision, utilising a 75 kg [165 lb] mass dummy, an impact velocity of 50 kph [31 mph] and a stopping distance of approximately 400 mm [16”] with a maximum deceleration of 30 g [FIA Standard 8854/98 dynamic test requirements]:
· The pelvic load, expected to surpass 14 kN [3,100 lb] at each side, will elongate the lap belt and compress the tissue on the pelvis. The pelvis will slide forward by 80 to 100 mm [3”-4”]
· The upper torso load is expected to surpass 7 kN [1,550 lb] on each strap, will elongate the shoulder belts, the upper body will roll in, the adjusters will move up the chest by approximately 200 mm [8”] and forward head trajectory will be up to 400 mm [16”]
· The pelvic movement combined with the shoulder belt forces will load the anti-sub straps to more than 6 kN [1,320 lb] each in a 5- or 6-point racing harness and can be intentionally higher in a F‑type model where the anti-sub straps are routed rearwards.
In more reclined seating positions, the pelvic load will be reduced to approx. 9 kN [2,000 lb] since the seat pan, designed as a ramp, will take some of the load. Therefore it is essential the seat or chassis manufacturer ensure the seat pan is strong enough not to bend or even collapse under extreme loads.
In any case, seats are a significant component of the safety system!

Actually I just found this later in the Schroth harness guide:

Each shoulder and lap belt anchor point must withstand a minimum load of 14,7 kN (3.240 lbs).

So my test load should be 29.4 kN (14.7 per shoulder strap) for each person in the vehicle. That's equivalent to the force from a ~63 g collision with a 165 lb person. I need to increase my simulated loads a bit.