What do sub frame connectors really do...

[SN65]

I have been giving this a lot of thought.

And

Even though a certain amount of twist is applied to the frame rails during acceleration and cornering, I believe that the benefits of sub-frame connectors are not as much torsional (is that a real word?) but in that they extend the lever point on the sub-frames further back in the chassis.

Is this correct?

[astroracer]

I believe that the benefits of sub-frame connectors are not as much torsional (is that a real word?) but in that they extend the lever point on the sub-frames further back in the chassis.

Is this correct?

Yes, but they also improve the torsional (it is a word) resistance a little. A roll cage will improve resistance to twisting more then subframe connectors alone. In combination they work well to stiffen up the "two part" chassis...
Mark

[High Plains Mopars]

Think of it this way, get a shoe box, take the lid off and give it a torsional twist. Now put the lid back on, tape it down, and twist it again.

[silver69camaro]

Think of it this way, get a shoe box, take the lid off and give it a torsional twist. Now put the lid back on, tape it down, and twist it again.

That would be a good convertable/couple analogy...but not sure about subframe connectors. The top is adding another shear plane, which isn't really what subframe connectors do.

How about fastening two rows of popcicle sticks to the bottom?

[Mkelcy]

Think of it this way, get a shoe box, take the lid off and give it a torsional twist. Now put the lid back on, tape it down, and twist it again.

That would be a good convertable/coupe analogy...but not sure about subframe connectors. The top is adding another shear plane, which isn't really what subframe connectors do.

How about fastening two rows of popcicle sticks to the bottom?

Perhaps the correct shoe box analogy is to cut out the center 4" of the bottom of the shoe box from side to side, twist it; then tape on your popsicle sticks (likely any flavor would work) with, say, two inchs of overlap front and rear and try your twisting again.

[silver69camaro]

Cut out the center? Why would that be?

To hell with cars. I'm moving to shoeboxes and popcicle sticks!

[Nine Ball]

Just think of them stiffening the chassis and allowing the suspension to work properly. These old cars weren't meant for high HP in stock form, they have a lot of twist and flex.

Example, when I got my red '69 finished, I took it to the drag strip and ran low 11s with a lot of drama. The car was scary and felt loose, even on drag slicks.

Then I had the 6-point rollbar and Global West subframes welded in, and the car went straight as an arrow down the track with no drama. Without any changes it went 10s and launched perfectly level.

[SN65]

I think people tend to get confused with the "shoebox" annalogy.

But, that is the refference we are stuck with, so I guess I will have to live with it. :-)

Take the shoebox and glue two popsicle sticks to it. 1/3rd glued to the box and 2/3rds hanging off the front. Locate the popsickle sticks about 1/4 of the way from the edge of the box, just like a frame rail. Now take the box in one hand and push up on one popsicle stick and down on the other in a twisting motion.

What happens?

The floor of the cardboard box caves in on one side and bulges out on the other.

Now extend the sticks all the way to the back of the box and repeat. You get the same results but reduced somewhat.

Now, remove the extensions and place a stick across the floor of the box at the end of the frame rails and repeat. You get less deflection in this configuration.

In my mind, by running a brace across the floor at the end of the frame rails, you achieve the same benifits of extending the frame rails, but you also remove a considerable amount of flex from the floor.

Does this make sense?

[astroracer]

Yes it does Bob, but the "crossmember" doesn't do the same job that the connectors will. The connectors tie the car together from front to back. On a hard launch the car will want to "bend" around the weakest point in the floor. This point is at the front of the spring mounts. This bending is not controllable. It will bend differently with every launch depending on conditions, traction, engine tune , whatever. The connectors tie the spring mounts into the forward subframe which makes the floor of the car much stiffer and moves the bending point forward. This increases the effectiveness of the launch. Instead of "bending" the car in the middle at the initial launch the newfound stiffness wants to move the whole car. This reduces the lost motion incurred by the bending, quickens the 60ft times and makes the car more predictable at launch and, as Nine ball stated, much more stable at trap speed.
As an example:
Take your shoe box and cut the "doors" out of both sides. Build your popsiclestick rear frame with the front of the "frame" about where the back of the doors are. Now do your twisting. Add your crossmember and twist it again. NOW add a connector, from the front of your rear frame to the front of the box. How does that work to keep the box from drooping when you tilt it up from the back? it should work much better. The same principle applies to the short subs in the unibody cars.
Mark

[Beige]

That sounds like what was said in the first post.

[bbnova]

Anyone know where I can buy some REALLY BIG popsicle sticks? Are they good for roll cages too?

LOL! Seriously, the shoebox/pop stick analogy is pretty good.

[vanzuuk1]

I have a carbon fiber shoebox with titanium popsicles sticks.

[SN65]

We are currently modifying the unibody structure of a 65 Mustang coupe.

So far, on our test bed, without the addition of a "cage" or "sub-frame connectors" we have reduced chassis "twist" by 67%. We still have a number of mods to make to the chassis. At this point we still have not welded the floor to the rockers, tied the frame rails to the seat pans, installed the export brace, etc...

When complete (without a cage) I am expecting a 75% improvment in chassis stability. Or, if you prefer, a 75% reduction in torsional deflection. Once we add the cage, we are expecting the number to be somewhere in the 80% to 90% range.

I am still debating if we should add any type of sub frame connector. If the rest of our design is correct, I feel that we really should not need them. Also, if we did not add a cage, you would not be able to tell that we made any mods at all.

We are adding a full cage to the car, along with some other goodies, to make a run at the Maxtion Mile and possibly the Silver State Classic Challenge. Our goal is to bring the car in at under 3,000 lbs. Once we know the final weight, we can determine our HP requirements.

Well, it is time to get back to work.

OBTW, we will be showing the completed car at the Detroit Autorama. Maybe this year they will put us right up front with the big dogs. :-)

[SN65]

Bolting in the export brace brings deflection down to .319 The original deflection measurement was 1.062". That currently brings us to a 70% improvement over stock.

[High Plains Mopars]

Wow, never thought the shoe box would have sparked so much conversation.

In any case, yeah, I guess adding the top to the box is more of a convertible/coupe comparison, like silver69 pointed out. So in that case, I'd say taping the top down to the box is like adding subframe connectors. It reduces deflection. Eventhough an untapped lid will reduce movement, there is still some there. Tapping is down reduces it further.

This is why; a lot of the cars we love are built out of sheetmetal that is spot welded together. Although is may be thick sheet, it maybe strong sheet, and there are lots of spot welds in it, it is still a collection of seams and spots. These seams allow deflection. Adding subframe connectors are a quick and easy way to shore up some of the flex that is inherent with a unibody design. It goes without saying that these conenctors are most effective if they tie into the floor pan rather than just tieing front and rear sub frames together.

Now, if you are so inclined, you could fire up the welder and stitch every spot welded seam on your car. This will produce the same results in rigidity because it eliminates flex points. Although SN65 didn't mention what mods they performed to their latest project to reduce deflection, welding seams would likely produce improvments like those mentioned in the last post. But, it takes a long time and some skill to do this much work. Most guys I meet would rather slap in a set of frame connectors and go.

For another analogy, if you try to rip a new, button on shirt off your bod, does it tear down the middle of the material, or does it pop the buttons off? Those spots welds are just buttons holding your car together. ..and in most cases they are 25-40 year old buttons that have been left outside most their lives.

[David Pozzi]

I'm interested in handling performance which requires torsional rigidity. I'd like to see how much subframe connectors increase it. I think there needs to be a diagonal component to the subframe connectors to significantly help torsional rigidity. Typical subframe connectors are a good thing, but I think there is much more to be gained if a better design were developed for us open track guys. Look at how GM designed the convertable, they added a plate and structure underneath to tie both sides together. Something that could do this but farther back tying in the front spring perches would be nice, but it wouldn't leave much room for exhaust.

[High Plains Mopars]

Xing the subframes is definetly a big help. I can't provide you with numbers thoug. I can say that I have always added X braces underneath the floor pan of my track cars.

As you point out, exhaust tubing clearence is a big issue which is why I haven't tried it on any of my street cars. However, it certainly would be possible. My X braces have always been round DOM tubing. By altering the size and location where they mount into the sub-frames, or how they track along the floor pan, you can adjust how much room the bracing takes up. For example, in my Dodge Aspen, the dissimilar levels of the floor pan means any bracing starts under the floor at the front of the car and ends up on top of the floor at the rear of the car. This requires cutting into the floor pan to allow tubing to pass through. This does provide an advantage in creating a more solid struture, but makes loads more work. It also makes carpeting a bit more of a hassle and costs some rear seat leg room.

Similarly, you could create pass through ports for exhaust tubing. Look at Morrisons G-Max chassis' for an example of this. These days we also have the luxery of oval exhaust tubing. It is a bit expensive, but could be utilized to reduce exhaust height by half without compromising flow. See Spin Techs web site for oval exhaust tubing, although I'm sure you can find it other places.

[chicane67]

It also depends on material.... round tube vs. rectangular.... wall thickness and diameter.... and how its attached all have huge impacts in the mechanics.

What is more tortionally ridgid, rectangular or round ?? Start there and apply common sense.

A shoe box and pop sickle sticks may give you an elementry introduction to what is going on here, but has very little to do with the end result.

[Damn True]

I'm interested in handling performance which requires torsional rigidity. I'd like to see how much subframe connectors increase it. I think there needs to be a diagonal component to the subframe connectors to significantly help torsional rigidity. Typical subframe connectors are a good thing, but I think there is much more to be gained if a better design were developed for us open track guys. Look at how GM designed the convertable, they added a plate and structure underneath to tie both sides together. Something that could do this but farther back tying in the front spring perches would be nice, but it wouldn't leave much room for exhaust.

I think the cage, if designed properly, satsfies that no?

I have a feeling (no tech to back it up) that adding some structure between the front ends of the 1st Gen sub and something good and sturdy used at the trans x-member would diminish the lever arm effect of a stiff spring way out on the end of a subframe member.

Instead of front hoops on a cage? No. But in addition too.

[David Pozzi]

True,
A cage is the usual way road racers have beefed up unbody cars. I think Mark Donohue was one of the first guys to do it, and he stated the results were great. Cages were not done much before that due to the extra weight, only roll bars. SCCA actually restricted roll cage attachment, it had to connect to the unibody and I think only the front braces could connect directly to the sub.

The roll bar we put in my wife's 73 Camaro made it feel a bit more solid.

Front braces firewall to sub on a Camaro would help tie it in, the fenders do quite a bit too.