I would like to ask a unibody engineer some questions

Shocks, Springs, Brakes, Frame, Body Work, etc

Moderator: Team

j-c-c
Guru
Guru
Posts: 1321
Joined: Sat Sep 03, 2016 9:03 pm

Re: I would like to ask a unibody engineer some questions

Post by j-c-c » Fri Jan 19, 2018 8:53 pm

mk e wrote:
Thu Jan 18, 2018 9:48 am
Brian P wrote:
Sat Mar 21, 2015 10:26 am
The beam loading is only part of it; torsion is a biggie because a one-wheel bump tries to twist the bodyshell, first up front, then in back.
I did the chassis design for an fsae car years ago and dug up some reasearch from the 60s that said its a waste of time to worry about beam strength becausev any design that had adiquate torsional strength also has plenty of beam strength. The was true for all the design iterations i ran through fea and i remember referencing the work in the report.

Today I think you more hit the nail when you mensioned crash testing and safety. I'm pretty sure that any design that will pass the crash test standards will have both adequate beam and torsional stiffness, at least for street use, racing applications change things a bit.

So the question I have for the original poster is what exactly is the concern with beam strength as it's normally only important in things like truck load capacity and mot very interesting if you're talking about racing?
Would be nice to review that 60's beam strength understanding mentioned above. And i take from that, most typical sub-frame connector end user designs that focus on vertical beam strength vertical rectangle tubes are misdirected?

mk e
Guru
Guru
Posts: 3813
Joined: Mon Sep 17, 2012 3:19 pm
Location: Elverson, PA

Re: I would like to ask a unibody engineer some questions

Post by mk e » Sat Jan 20, 2018 7:31 am

j-c-c wrote:
Fri Jan 19, 2018 8:53 pm
Would be nice to review that 60's beam strength understanding mentioned above. And i take from that, most typical sub-frame connector end user designs that focus on vertical beam strength vertical rectangle tubes are misdirected?
I remember is was all based on ladder fame stuff, which was the standard of the day...and it makes a lot of sense if you thing about it.
A stamped channel is pretty strong up and down, as long as id doesn't twist which is why they add the cross members. If you make it big enough it works just fine, but big enough means it gets heavy and heavy costs money and reduces performance which is way the factories were looking for more efficient designs.

I was working mainly with space frames but modeled a basic ladder frame and the previous design I was replacing was a monocoque so I also modeled that specific design for comparison but made no attempt to improve it.

Subframe connectors serve a couple purposes depending on the car. If the subframes are rubber mounted they are moving relative to each other in every which direction which is very unhelpful to repeatable setups. Adding subframe connectors, the bigger the better, basically converts it to a ladder fame, but then the whole body above that was designed to add stiffness handles the torsional loads and it works better.

On something that was stamped and welded subframes, like a foxbody mustang the connectors work a little different. Again the bigger the better applies but this time you really have a monocoque structure and you are reinforcing it. stock the tunnel and rockers form the structure and the roof adds the torsional strength, but now you are adding 2 additional beams which can dramatically stiffen the floor and the whole chassis along with it.

Adding metal pretty much always helps stiffen things up, but getting the stiffness /lb as high as possible is the goal. Monocoque structures are about the best at stiffness/lb, really had to beat.....but cars need doors and engine bays and such which mean cutting big wholes in the skin and that then requires frame like shapes be added to support those areas. When you look at something like an F1 chassis they call it a monocoque but really its a ladder frame with HUGE side rails and sear panels added here and there.

Space frames were the subject of my research and work very nearly as well as monocoque structure when they are designed properly for torsion, but are great for cars because they give you "hard point" to mount drivetrain and suspension too. The problem is most examples I found were not designed very well. To work right a space frame needs to be made from triangles, but stupid engine bays and door opening bugger that up so a lot of structure needs to be add just like on the monocoque but most examples I found simply removed the diagonal to create the opening without beefing up the structure which absolutely destroys the torsional stiffness of the whole chassis. I did a formula car and started by designing a structure all triangles with no opening for a cockpit of anything else and did the fea. Then I removed only the 1 diagonal across the cockpit area and the chassis torsional stiffness dropped by 75%! A lot of care is required around openings. The final design in steel was on par with the late 80s- early 90s F1 carbonfiber stuff stiffness/lb wise so it can work very well.

It funny because I just had basically this same discussion with my bother last weekend but talking about pizza ovens. He builds masonry wood fired ovens and has people who want them on a trailer from time to time. When he did the first one like 10 years ago we designed a a ladder fame using 4x4x1/4 steel tubing and that oven is going to events all over the state and happy as can be. Masonry cracks so the frame had to be very stiff, beam strength wise if could carry 20 ovens but we just wanted 1 oven with no cracks. That frame was friking heavy and the welding shop talked him into 3x3x3/16 on the next one which was 1/2 the weight and still "plenty" strong...but that oven now has odd cracks in it. So the talk was about can he do something like cars do where all the storage space is structural and helps replace the heavy frame......about an hour later explaining the design concerns the frame will be 4x4x1/4 because its a frikin masonry oven and nobody but the welding shop cares how heavy the frame is and anything you do to lighten it for them adds WAY more cutting and welding which drives the price way up. Sometimes simple is good.
Mark
Mechanical Engineer

pdq67
Guru
Guru
Posts: 7233
Joined: Thu Mar 04, 2010 8:05 pm

Re: I would like to ask a unibody engineer some questions

Post by pdq67 » Sat Jan 20, 2018 7:58 am

Mark,

Don't know if it is possible, but can you use ceramic fiber or block insulation in the arch of your traveling pizza ovens?

Light as a feather, but I don't know about any health issues due to ceramic fiber, "dusting"? Make the hearth/floor out of brick because they naturally will have cracks due to their joints! A strong lightweight insulating firebrick might be used instead of dense brick to make the ovens lighter?

Might install a suspended corrugated stainless steel, "shroud", in front of the ceramic fiber to catch any dusting?

"Corrugations", just to handle the constant heat cycles is all here.

Back when I was with AP Green Refractories, I designed a lot of furnaces using all of these types of material and construction techniques.

pdq67

mk e
Guru
Guru
Posts: 3813
Joined: Mon Sep 17, 2012 3:19 pm
Location: Elverson, PA

Re: I would like to ask a unibody engineer some questions

Post by mk e » Sat Jan 20, 2018 10:04 am

pdq67 wrote:
Sat Jan 20, 2018 7:58 am
Mark,

Don't know if it is possible, but can you use ceramic fiber or block insulation in the arch of your traveling pizza ovens?

Light as a feather, but I don't know about any health issues due to ceramic fiber, "dusting"? Make the hearth/floor out of brick because they naturally will have cracks due to their joints! A strong lightweight insulating firebrick might be used instead of dense brick to make the ovens lighter?

Might install a suspended corrugated stainless steel, "shroud", in front of the ceramic fiber to catch any dusting?

"Corrugations", just to handle the constant heat cycles is all here.

Back when I was with AP Green Refractories, I designed a lot of furnaces using all of these types of material and construction techniques.

pdq67
I don't know much about ovens, that's my bother's thing. I know we talks about thermal mass being important so he's generally not interested in reducing the weight of the oven itself, but he have talked about materials that have higher thermal storage/lb. I think the whole mobile setup weighs just under 6k lbs with the heavy 4x4x1/4 frame.
Mark
Mechanical Engineer

PackardV8
Guru
Guru
Posts: 5294
Joined: Sun Jul 30, 2006 2:03 pm

Re: I would like to ask a unibody engineer some questions

Post by PackardV8 » Sat Jan 20, 2018 11:48 am

exhaustgases wrote:
Sat Mar 21, 2015 2:54 am
1st question has to do with choice of jack points especially the use of the knife edge of a pinch weld, ie very small surface area and very little side stability.
Many vehicles these days are no longer equipped with either a spare tire or a jack, so no worries about where to jack. And if a road service guy changes a tire, he's going to put the jack where it's most convenient for him. Doesn't matter whether it's a factory dimple in the pinch weld or the solid keyed jack point, him crawling around on his belly in the mud and snow looking for whichever ain't gonna happen anyway.
All cars should have solid keyed jack points and not a flimsy little thin strip to jack on. By keyed I mean a pad with a hole in it for a post on the jack to fit in or something on those lines. And have the top of the jack with a plastic pad incorporated on it, am I the only one that cares about protecting stuff?
I asked one engineering team why the above weren't universally used. The answers:
1. Only an insignificantly small minority of owners ever jack up their own car, so their convenience is not cost-effective.
2. The design goal is to reduce pieces, reduce welds, reduce assembly time.
3. Once a unibody is designed and tested, the team is then asked to go back over it and find ounces which can be removed from every area without compromising strength.
4. Our cars don't have spare tires or jacks.
Jack Vines
Studebaker-Packard V8 Limited
Obsolete Engineering

User avatar
MadBill
Guru
Guru
Posts: 13301
Joined: Tue Nov 15, 2005 10:41 am
Location: The Great White North

Re: I would like to ask a unibody engineer some questions

Post by MadBill » Sat Jan 20, 2018 10:42 pm

Back in the mid-seventies, my pal had his new GMC van recalled because they discovered the jacks wouldn't fit under the trucks if the tire was flat.. #-o
Felix, qui potuit rerum cognscere causas.

Happy is he who can discover the cause of things.

User avatar
panic
Guru
Guru
Posts: 1957
Joined: Sat Dec 11, 2004 12:04 pm
Location: Ecbatana
Contact:

Re: I would like to ask a unibody engineer some questions

Post by panic » Sat Jan 27, 2018 9:49 am

They went on the develop the Hubble telescope...

Nefario
Member
Member
Posts: 60
Joined: Thu Jul 21, 2011 5:32 pm
Location: Kentucky

Re: I would like to ask a unibody engineer some questions

Post by Nefario » Thu Feb 01, 2018 5:19 pm

mk e wrote:
Sat Jan 20, 2018 7:31 am
It funny because I just had basically this same discussion with my bother last weekend but talking about pizza ovens. He builds masonry wood fired ovens and has people who want them on a trailer from time to time. .... Masonry cracks so the frame had to be very stiff, beam strength wise if could carry 20 ovens but we just wanted 1 oven with no cracks.
Interesting. I've always had an interest in truss and structural design and won several bridge building contests in college. I've also noted how BAD many competitive old racecar structures are when you consider triangulation. The "Petty Bar" would seem to be an obvious thing to do and alone, it's not that great.

Regarding concrete pizza ovens! :D
Are you aware of basalt fiber and basalt rebar? You can buy basalt cloth which could have some very interesting automotive applications involving extreme heat resistance with nearly the strength of carbon steel. Basalt roving could be incorporated into the dome of a pizza oven. In masonry, basalt rebar:
- Is a fraction of the weight of steel.
- Can't rust - it's a rock
- Nonmagnetic, nonconductive
- AND, it's organic! (like high-end pizza!)

https://www.basalt.guru/use-basalt-reba ... -concrete/

User avatar
MadBill
Guru
Guru
Posts: 13301
Joined: Tue Nov 15, 2005 10:41 am
Location: The Great White North

Re: I would like to ask a unibody engineer some questions

Post by MadBill » Thu Feb 01, 2018 5:28 pm

mk e wrote:
Sat Jan 20, 2018 7:31 am
..
It funny because I just had basically this same discussion with my bother last weekend but...
Is it a Freudian slip mk e that in two posts in a row you call him your bother? (I've got a pair of those myself...) :lol:
Felix, qui potuit rerum cognscere causas.

Happy is he who can discover the cause of things.

mk e
Guru
Guru
Posts: 3813
Joined: Mon Sep 17, 2012 3:19 pm
Location: Elverson, PA

Re: I would like to ask a unibody engineer some questions

Post by mk e » Fri Feb 02, 2018 5:57 pm

MadBill wrote:
Thu Feb 01, 2018 5:28 pm
mk e wrote:
Sat Jan 20, 2018 7:31 am
..
It funny because I just had basically this same discussion with my bother last weekend but...
Is it a Freudian slip mk e that in two posts in a row you call him your bother? (I've got a pair of those myself...) :lol:
If that's the worst typo in 2 posts that pretty good for me :lol:
Mark
Mechanical Engineer

Ratu
Expert
Expert
Posts: 624
Joined: Sun Oct 02, 2011 7:58 pm

Re: I would like to ask a unibody engineer some questions

Post by Ratu » Fri Feb 02, 2018 7:36 pm

mk e

You write, " I did a formula car and started by designing a structure all triangles with no opening for a cockpit of anything else and did the fea. Then I removed only the 1 diagonal across the cockpit area and the chassis torsional stiffness dropped by 75%! A lot of care is required around openings. The final design in steel was on par with the late 80s- early 90s F1 carbonfiber stuff stiffness/lb wise so it can work very well."

That is fascinating. Can you post some pictures/drawings of it?

mk e
Guru
Guru
Posts: 3813
Joined: Mon Sep 17, 2012 3:19 pm
Location: Elverson, PA

Re: I would like to ask a unibody engineer some questions

Post by mk e » Sat Feb 03, 2018 9:24 pm

Ratu wrote:
Fri Feb 02, 2018 7:36 pm
mk e

You write,.....

That is fascinating. Can you post some pictures/drawings of it?
Couple.....it was many years ago. I did a couple designs for a mid 911 engine speedster I was contemplating, 1 with working doors and one that was just silly rigid but no doors.

On the formula car I started the work with a pretty simple rectangle chassis, that was the version I did the with/without functional cockpit test thinking the engine bay which is also not properly triangulated many times is easy enough to solve with bolt-in bracing but the cockpit needs to be open. After confirming it was a major problem that became the focus of the work and the solution I settled on moved the front roll hoop forward about 6 inches and I added a couple members that go above the drivers shoulders and form a "v" in the cockpit opening. Very stiff and strong and worked very well. The speedster design is a pair of the formula chassis siamesed together.
Attachments
20180203_210034.jpg
20180203_205720.jpg
20180203_205520-resized-1024.jpg
Mark
Mechanical Engineer

mk e
Guru
Guru
Posts: 3813
Joined: Mon Sep 17, 2012 3:19 pm
Location: Elverson, PA

Re: I would like to ask a unibody engineer some questions

Post by mk e » Sun Feb 04, 2018 7:00 am

I'll add this. Way back when I was doing the analysis on the formula chassis I noticed that a quick and very accurate way to tell how well optimized the design is was to look at what happen when the tube diameter (or wall thickness) is changed. If the design is perfect all the loads will be tension and compression so the change in stiffness will be directly proportional to the change in tubing cross sectional area. If the design is not perfect there will be be bending loads so stiffness will also be a function of I (moment of inertia) of the tubes. IIRC the finished formula chassis was 90%+ proportional to the tubing CSA,

Playing with the thought of adding an upstairs to my new garage and wishing I'd been smart enough to order the trusses with a room up there I've been running a bunch of numbers on modifying the trusses. The trusses as-is are triangles and I get within 5% the same answer if I call the joints rigid or flex which means there are basically no bending loads. Restructuring to have a room means remove the center triangulation, which means bending loads and a 40% difference if I treat the corners as joint vs fixed, so the same result I saw with the formula chassis design just showing up a different way.

I found this free on-line program, its only 2D but you'll see pretty quickly the difference between bending and properly triangulated

click "Deflections"
original truss fixed joint.
http://en.sopromat.org/2009/?id=45833

original flex joint
http://en.sopromat.org/2009/?id=45844

modified flex joint (where applicable)
http://en.sopromat.org/2009/?id=45846
Mark
Mechanical Engineer

pdq67
Guru
Guru
Posts: 7233
Joined: Thu Mar 04, 2010 8:05 pm

Re: I would like to ask a unibody engineer some questions

Post by pdq67 » Sun Feb 04, 2018 8:17 pm

pdq67 wrote:
Thu Feb 04, 2016 6:49 pm
exhaustgases wrote:No matter what, its still dumbness to design a jack point like that. Just have someone inexperienced like most people driving on the roads are, doing the jacking, and once the jack area gets folded over all strength in it is lost and it will just keep folding over after its been straightened, been there done that. All cars should have solid keyed jack points and not a flimsy little thin strip to jack on. By keyed I mean a pad with a hole in it for a post on the jack to fit in or something on those lines. And have the top of the jack with a plastic pad incorporated on it, am I the only one that cares about protecting stuff?
As soon as paint is removed with a metal on paint edge a nice place is made for rust to start its job, especially in the snow belts where salt is over used.
Good come-back about the dedicated jack points and plastic jack body protector!!

I forget what car it came from, but the jack laid on the ground and raised the car up at a dedicated point in the undercarriage. Kinda like a sheet metal mechanical floor jack.....

Maybe it was from one of the little BOP compacts back in the early '60's?

Oh, my '67 Camaro has a hole at each "corner" of both bumpers just inside the bumper bolts to insert the bumper jack stud into..

pdq67
I was just re-reading this again for whatever and remembered that my '67 Camaro had a recall letter sent out because the spare tire hold down hook wasn't strong enough. it would bend and come loose from its mount thus allowing the spare tire to bounce around in a bad wreck. I think they sent(?)/gave(?) me a hook that was like a 1/2" in dia. vs 3/8" or so, I forget? Might have been 7/16" in dia., I would have to go and look...

pdq67

Dan Timberlake
Guru
Guru
Posts: 1373
Joined: Tue Jan 15, 2013 8:10 pm

Re: I would like to ask a unibody engineer some questions

Post by Dan Timberlake » Thu Feb 08, 2018 11:47 am

Packard V8 said - "body-on-frame car with a bumper jack and saw doors which wouldn't open, doors which flew open, windshields or rear windows cracked by body twist, bumpers which twisted into the body panel".

Some 60s BoF cars claimed to have fancy body mount biscuits, in strategic locations to isolate the frame from the body for NVH improvements.
http://oldcarbrochures.org/index.php/NA ... ll-Size-24
"node point body mounting"

Certainly there are times when controlled flexibility is required for a great design

I always silently ponder the reverence with which some folks say "and it's got a full frame ."

Post Reply