Jump to content


Photo

Spaceframe joints with dissimilar tube sizes


  • Please log in to reply
141 replies to this topic

#101 mariner

mariner
  • Member

  • 1,406 posts
  • Joined: January 07

Posted 07 January 2012 - 12:11

I believe this is one of the more misunderstood and overhyped aspects of hobbyist chassis design. Yes, round is slightly stiffer than square per unit weight. However, square is stiffer for a given tubing size. There's the grand morsel of theory for everyone to beat to death, anyway.

But the real issue is not in theory but in practice. In order to use the chassis frame for anything, you will need to mount suspension, drivetrain, body, fuel tank, etc. and there's the advantage of square/rectangular tubing.



The exact point I was as trying to make - once you give up theoretical stiffness for practicality then apply the practicality logic throughout the chassis. That is why I think switching to round tubes in this part of the chassis might be the best overall solution.



Advertisement

#102 Kelpiecross

Kelpiecross
  • Member

  • 943 posts
  • Joined: October 10

Posted 07 January 2012 - 12:27

[quote name='NeilR' date='Jan 7 2012, 20:19' post='5471536']

Thanks for the pictures - extremely interesting. Its a bit hard to see the chassis tubes in the Saleen (Is a Saleen another name for Mosler?) but they don't seem to worry excessively about perfect triangulation etc. On the Artega the chassis is more visible but again around the rear section especially the triangulation seems a bit lacking. They don't seem to attempt to triangulate the big open areas over the roof either.
Presumably this is acceptable - but there doesn't seem to be the mania for perfect triangulation there once was. I wonder what Costin and Phipps would make of all this?
Areas of box section sheet metal seem to be popular -good idea too - easier and neater than having tubes running everywhere.

#103 rachael

rachael
  • Member

  • 111 posts
  • Joined: October 10

Posted 07 January 2012 - 12:32

I believe this is one of the more misunderstood and overhyped aspects of hobbyist chassis design. Yes, round is slightly stiffer than square per unit weight. However, square is stiffer for a given tubing size. There's the grand morsel of theory for everyone to beat to death, anyway.

But the real issue is not in theory but in practice. In order to use the chassis frame for anything, you will need to mount suspension, drivetrain, body, fuel tank, etc. and there's the advantage of square/rectangular tubing.

Let's be honest here: the real reason enthusiast builders use round instead of square is that in their minds, round looks racy and square looks agricultural. If they backed up a few steps and took a more balanced and objective approach, they'd be more likely to choose square tubing. I like the car pictured here. It represents sound, practical, intelligent design.


100% agree (again)

Major practical disadvantage of square vs round is that you cant bend it which probably means it's crash performance is better.

#104 Kelpiecross

Kelpiecross
  • Member

  • 943 posts
  • Joined: October 10

Posted 07 January 2012 - 12:40

100% agree (again)

Major practical disadvantage of square vs round is that you cant bend it which probably means it's crash performance is better.


Square is a bloody sight easier to attach brackets etc. and screw things to it. And attach sheet panels.

#105 cheapracer

cheapracer
  • Member

  • 10,388 posts
  • Joined: May 07

Posted 07 January 2012 - 13:05

I believe this is one of the more misunderstood and overhyped aspects of hobbyist chassis design......



10/10 - I wouldn't change one word of that excellent post and will bookmark it for future use.



but there doesn't seem to be the mania for perfect triangulation there once was. I wonder what Costin and Phipps would make of all this?


I've noticed the same from some manufacturers and sorry to say it seems to be tiding over to homebuilders who see them as examples.



100% agree (again)

Major practical disadvantage of square vs round is that you cant bend it which probably means it's crash performance is better.


No, round in fact is quite easy to bend, easier than square even.

Edited by cheapracer, 07 January 2012 - 13:10.


#106 rachael

rachael
  • Member

  • 111 posts
  • Joined: October 10

Posted 07 January 2012 - 13:29

No, round in fact is quite easy to bend, easier than square even.


Yep that's what I meant! Didn't think it was possible to bend square tube without cutting and welding?

#107 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 07 January 2012 - 13:37

The exact point I was as trying to make - once you give up theoretical stiffness for practicality then apply the practicality logic throughout the chassis. That is why I think switching to round tubes in this part of the chassis might be the best overall solution.


I wasn't disagreeing so much as expanding upon your remarks.

Oh, except for the bit about American construction. That was a bit harsh. Acre for acre, the English have built as much useless and dangerous junk as anyone, triangulated as it may have been.

#108 cheapracer

cheapracer
  • Member

  • 10,388 posts
  • Joined: May 07

Posted 07 January 2012 - 13:41

Yep that's what I meant! Didn't think it was possible to bend square tube without cutting and welding?


Bent square tube is very common, some will even be sitting on chairs or at tables reading this with bent legs (the chairs and the tables, not their legs!).

Most 3 wheel pickups here have large section square bent chassis, wondrous thing is steel - even Chinese steel ...

http://www.alibaba.c.../showimage.html




#109 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 07 January 2012 - 13:41

It's quite possible to bend square tube but not as neatly in appearance or to as small a radius as round tube.

EDIT: Cheapracer beat me to the reply. What he said.

Edited by Magoo, 07 January 2012 - 13:43.


#110 Engineguy

Engineguy
  • Member

  • 989 posts
  • Joined: February 01

Posted 07 January 2012 - 13:47

Yep that's what I meant! Didn't think it was possible to bend square tube without cutting and welding?


2x3 or 2x4 rectangular tubing... the easy way or the hard way.

AutoWeld Chassis
Posted Image

#111 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 07 January 2012 - 13:58

There's also hydroforming, which can take any round or square tube cross-section and turn it into any conceivable cross-section, from roundy-squarish to trapezoid-y. Or round in one place and square in another. This technology is not available to home constructors yet, but you can see it coming.

#112 Engineguy

Engineguy
  • Member

  • 989 posts
  • Joined: February 01

Posted 07 January 2012 - 14:18

In reply to the discussion of round vs square and wall thickness changes, etc....

I developed a whole set of these charts for various wall thicknesses a couple years ago. They are eye-opening IMHO if you study them for a while. Changed my mind about how to build a frame.

LINK: Torsion Constants of Round and Square Tubing



#113 rachael

rachael
  • Member

  • 111 posts
  • Joined: October 10

Posted 07 January 2012 - 14:44

Bent square tube is very common, some will even be sitting on chairs or at tables reading this with bent legs (the chairs and the tables, not their legs!).

Most 3 wheel pickups here have large section square bent chassis, wondrous thing is steel - even Chinese steel ...

http://www.alibaba.c.../showimage.html


ok what I should have said was 'Didn't think it was possible to bend square tube without cutting and welding in the average home builders garage without buying specialist equipment or ending up with unsightly wrinkles on the inside radius'!!

#114 rachael

rachael
  • Member

  • 111 posts
  • Joined: October 10

Posted 07 January 2012 - 14:52

Not spaceframe but interesting chassis construction method: origami

I believe bentley have evaluated this technique.

#115 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 07 January 2012 - 15:02

ok what I should have said was 'Didn't think it was possible to bend square tube without cutting and welding in the average home builders garage without buying specialist equipment or ending up with unsightly wrinkles on the inside radius'!!


Touche.


#116 Grumbles

Grumbles
  • Member

  • 326 posts
  • Joined: September 09

Posted 07 January 2012 - 22:16

I wasn't disagreeing so much as expanding upon your remarks.

Oh, except for the bit about American construction. That was a bit harsh. Acre for acre, the English have built as much useless and dangerous junk as anyone, triangulated as it may have been.


National stereotypes are great timesavers; they enable one to quickly pass judgement on a car without having to actually study it.

Hence American cars are oversized, overpowered, underbraked and under-chassised.

British cars are small, darkly painted things that leak fluids from the engine and smoke from the electrics.

European cars are brightly coloured, slightly eccentric designs that go very fast when they aren't catching fire.

Japanese cars are fast, well finished and absolutely reliable. Boring, in other words.


#117 NeilR

NeilR
  • Member

  • 478 posts
  • Joined: October 09

Posted 07 January 2012 - 22:54

I believe this is one of the more misunderstood and overhyped aspects of hobbyist chassis design. Yes, round is slightly stiffer than square per unit weight. However, square is stiffer for a given tubing size. There's the grand morsel of theory for everyone to beat to death, anyway.

But the real issue is not in theory but in practice. In order to use the chassis frame for anything, you will need to mount suspension, drivetrain, body, fuel tank, etc. and there's the advantage of square/rectangular tubing.

Let's be honest here: the real reason enthusiast builders use round instead of square is that in their minds, round looks racy and square looks agricultural. If they backed up a few steps and took a more balanced and objective approach, they'd be more likely to choose square tubing. I like the car pictured here. It represents sound, practical, intelligent design.



If your last words are about my car then I thank you for the kind thoughts.
My choice of SHS for most of the chassis is a simple practical one. For an unskilled amateur such as myself I think it is easier to construct a square chassis with minimal jigging from SHS. It is easier to orient tube to tube, lays flat and does not roll about the place, has obvious planes of reference when cutting. I think there is also an advantage when welding the whole lot together and dealing with the stresses - vertical welds first and then others. I'm positive that all of these issues are not there for experienced chassis constructors, but I am not one.
Round is easier with three dimensional joints assuming you can fishmouth the stuff easily

#118 NeilR

NeilR
  • Member

  • 478 posts
  • Joined: October 09

Posted 07 January 2012 - 23:06

another 'solution' (at least in theory) crossed my mind when looking at the mallock and multimatic daytona prototypes. If I use a short length of 2.5mm wall round, capped at both ends. Insert this into the square and have it supported by an inner plate and the end cap of the square. Then protrude it enough (80mm) to have the oblique round fish mouth onto it. Not sure the complication is worth it, but another idea none the less.

#119 cheapracer

cheapracer
  • Member

  • 10,388 posts
  • Joined: May 07

Posted 08 January 2012 - 07:29

ok what I should have said was 'Didn't think it was possible to bend square tube without cutting and welding in the average home builders garage without buying specialist equipment or ending up with unsightly wrinkles on the inside radius'!!


Glad nobody told me.

I make my own square bending dies from scrap steel laying around and use a common hand pumped hydraulic pipe bender the same as I use for the round tube.

The trick is in the sizing of a piece of rod you lay inside the die's radius to push the inside of the box in concave as it bends, get that right and you get nice bends.

Like this guy's effort, very similar to my mine, which is also made from scrap (notice the rod laid on the centre of the radius). His is totally hand lever based but mine come out nicer than his probably because I found early on the dies needed to be bolted together or else you can't remove the box tube from the die so I presume he has too much clearance so he can get his box tube out of the die easily but the finish suffers ....

Posted Image

Because of the concave on the inside radius, these bends are very strong but of course can never look as good as round tube.

Edited by cheapracer, 08 January 2012 - 07:33.


Advertisement

#120 NeilR

NeilR
  • Member

  • 478 posts
  • Joined: October 09

Posted 08 January 2012 - 10:25

Posts back!

Edited by NeilR, 08 January 2012 - 23:11.


#121 rachael

rachael
  • Member

  • 111 posts
  • Joined: October 10

Posted 08 January 2012 - 10:49

Glad nobody told me.

I make my own square bending dies from scrap steel laying around and use a common hand pumped hydraulic pipe bender the same as I use for the round tube.

The trick is in the sizing of a piece of rod you lay inside the die's radius to push the inside of the box in concave as it bends, get that right and you get nice bends.

Like this guy's effort, very similar to my mine, which is also made from scrap (notice the rod laid on the centre of the radius). His is totally hand lever based but mine come out nicer than his probably because I found early on the dies needed to be bolted together or else you can't remove the box tube from the die so I presume he has too much clearance so he can get his box tube out of the die easily but the finish suffers ....

Posted Image

Because of the concave on the inside radius, these bends are very strong but of course can never look as good as round tube.


Sounds like a neat technique - however the concavity must reduce the stiffness not increase the strength?

#122 MatsNorway

MatsNorway
  • Member

  • 2,086 posts
  • Joined: December 09

Posted 08 January 2012 - 10:57

I did not fully understand your method but i still think its worth mentioning that you can use sand to bend pipes too. usually with the aid of heat but with hydraulics im guessing you wil not need to.

having guides on the sides for the square tube will probably prevent it from getting roundish when you bend it.

The big gain with this form of forming is that you don`t get wrinkles in the inner bend, at least not when you use heat. so its probably, structurally stronger.

#123 Tony Matthews

Tony Matthews
  • Member

  • 17,502 posts
  • Joined: September 08

Posted 08 January 2012 - 11:17

I did not fully understand your method but i still think its worth mentioning that you can use sand to bend pipes too. usually with the aid of heat but with hydraulics im guessing you wil not need to.



The big gain with this form of forming is that you don`t get wrinkles in the inner bend, at least not when you use heat. so its probably, structurally stronger.

You can use low-melting point alloys, too. I've known it to be used in bending large-diameter tubing for furniture, where appearance is vital. Fill the tube with molten alloy, let it cool, bend the tube, heat it gently and the alloy falls out - on to your shoes...

The square tubing that I've seen bent has a rather nice smooth concave trough on the inside radius.

Edited as 'allots' doesn't seem to be a word...

Edited again to say I didn't see cheapie's post...

Edited by Tony Matthews, 08 January 2012 - 11:20.


#124 cheapracer

cheapracer
  • Member

  • 10,388 posts
  • Joined: May 07

Posted 08 January 2012 - 13:07

I did not fully understand your method but i still think its worth mentioning that you can use sand to bend pipes too. usually with the aid of heat but with hydraulics im guessing you wil not need to.


I have tried sand a dozen times in my life including last year and it has always failed miserably.

And yes, before someone posts it, dry sand, packed in carefully and last time I even made a bolt press/packer on the end (welded on nut and fine thread bolt to really pack the sand down), all the "tricks" - still bad results

I have very good success with my pipe bender as long as the walls are not too thin, 1.8mm+, and as mostly it's for roll bars it's fine.


Sounds like a neat technique - however the concavity must reduce the stiffness not increase the strength?


They are very strong, surprisingly so.

.

Edited by cheapracer, 08 January 2012 - 13:14.


#125 Kelpiecross

Kelpiecross
  • Member

  • 943 posts
  • Joined: October 10

Posted 08 January 2012 - 13:20

You can use low-melting point alloys, too. I've known it to be used in bending large-diameter tubing for furniture, where appearance is vital. Fill the tube with molten alloy, let it cool, bend the tube, heat it gently and the alloy falls out - on to your shoes...

The square tubing that I've seen bent has a rather nice smooth concave trough on the inside radius.

Edited as 'allots' doesn't seem to be a word...

Edited again to say I didn't see cheapie's post...


I think "allots" is a proper word.

#126 kikiturbo2

kikiturbo2
  • Member

  • 668 posts
  • Joined: December 04

Posted 08 January 2012 - 13:43

You can use low-melting point alloys, too.



saw a "how it's made" episode on discovery about the manufacture of trumpets or some similar instrument... they filled the brass tubes with water, cooled it so that it would transform into ice, and then bend them.. :)

#127 cheapracer

cheapracer
  • Member

  • 10,388 posts
  • Joined: May 07

Posted 08 January 2012 - 13:54

saw a "how it's made" episode on discovery about the manufacture of trumpets or some similar instrument... they filled the brass tubes with water, cooled it so that it would transform into ice, and then bend them.. :)


I have tried water but never considered ice which expands of course - mind you I don't have a 2 or so meter long freezer either ...


I think "allots" is a proper word.


I had some in my oup with dinner tonight.


#128 kikiturbo2

kikiturbo2
  • Member

  • 668 posts
  • Joined: December 04

Posted 08 January 2012 - 14:08

well, I am currently giving headache to people with tube benders, having only one tube of 40x1.5 mm 4130 left, and I need a roll hoop bent.. :)

#129 Tony Matthews

Tony Matthews
  • Member

  • 17,502 posts
  • Joined: September 08

Posted 08 January 2012 - 15:04

I think "allots" is a proper word.

You are correct, squire, I saw it out of context and didn't recognise it.

#130 cheapracer

cheapracer
  • Member

  • 10,388 posts
  • Joined: May 07

Posted 08 January 2012 - 15:30

well, I am currently giving headache to people with tube benders, having only one tube of 40x1.5 mm 4130 left, and I need a roll hoop bent.. :)


Possibly slide it up the guts of some 45mm PVC pipe?

Why not 38mm x 2.5 mild DOM just for the rollbar?

#131 mariner

mariner
  • Member

  • 1,406 posts
  • Joined: January 07

Posted 08 January 2012 - 16:09

This is a hazy memory but I think plumbers used to have long coil springs they wound down into round central heating tubes.
Then they bent the tube , either by hand if it was just copper , or in a pipe bender. the bend was a really nice quality and afterwards they unwound the coil spring to get it out of the curve.

I guess the snag is you can only reach so far before the friction between the spring and the inner tube wall overcomes the tendency of the coil spring to lose diameter as you wind it down the tube.

I do know they used it in preference to the sand method.

#132 mariner

mariner
  • Member

  • 1,406 posts
  • Joined: January 07

Posted 08 January 2012 - 16:16

Somewhat off topic as this guy did carbon fibre tubes but quite an acheivement I think ( and good testing as well!)

http://sheldonbrown....arbon_fiber.htm

#133 kikiturbo2

kikiturbo2
  • Member

  • 668 posts
  • Joined: December 04

Posted 08 January 2012 - 16:25

Possibly slide it up the guts of some 45mm PVC pipe?

Why not 38mm x 2.5 mild DOM just for the rollbar?


not a bad idea.. :)

#134 Tony Matthews

Tony Matthews
  • Member

  • 17,502 posts
  • Joined: September 08

Posted 08 January 2012 - 18:48

This is a hazy memory but I think plumbers used to have long coil springs they wound down into round central heating tubes.
Then they bent the tube , either by hand if it was just copper , or in a pipe bender. the bend was a really nice quality and afterwards they unwound the coil spring to get it out of the curve.

I guess the snag is you can only reach so far before the friction between the spring and the inner tube wall overcomes the tendency of the coil spring to lose diameter as you wind it down the tube.

I do know they used it in preference to the sand method.

Posted Image
Internal pipe-bending spring.

Posted Image
External pipe-bending spring!


Edited by Tony Matthews, 08 January 2012 - 18:50.


#135 24gerrard

24gerrard
  • Member

  • 2,008 posts
  • Joined: March 11

Posted 08 January 2012 - 22:30

God Tony dont remind me.
I spent two Months helping a friend with the condense pipes in his winery bending what must have been thousands of bends using that method. Sand is much better if you are doing a lot, the springs can be a devil to get out.

#136 Lee Nicolle

Lee Nicolle
  • Member

  • 6,249 posts
  • Joined: July 08

Posted 08 January 2012 - 23:38

Glad nobody told me.

I make my own square bending dies from scrap steel laying around and use a common hand pumped hydraulic pipe bender the same as I use for the round tube.

The trick is in the sizing of a piece of rod you lay inside the die's radius to push the inside of the box in concave as it bends, get that right and you get nice bends.

Like this guy's effort, very similar to my mine, which is also made from scrap (notice the rod laid on the centre of the radius). His is totally hand lever based but mine come out nicer than his probably because I found early on the dies needed to be bolted together or else you can't remove the box tube from the die so I presume he has too much clearance so he can get his box tube out of the die easily but the finish suffers ....

Posted Image

Because of the concave on the inside radius, these bends are very strong but of course can never look as good as round tube.

for light guage squ tube bending I have done it around a Vee belt pulley. Both 1/2 and 3/4. Hold the pulley in a decent vice and just wrap it around. Time consuming and hard work with 3/4 but I have never lost any material yet.
This is only good for making speedway wing frames, and the several light brackets and mounts [dash, panelsetc] on my cars.
I have bent similar size round but it does not want to bend as neatly surprisingly.

#137 NeilR

NeilR
  • Member

  • 478 posts
  • Joined: October 09

Posted 11 January 2012 - 05:51

Someone posted "The short - hand advice in their pamphlet "Design of SHS Welded Joints" is that provided that the thickness of the plate is 1.5 times thicker than the tube wall thickness, then the joint will be 100% efficient." and I copied it to have a tink about, but we saeem to have lost some detail in the hiccup. I assume this comment relates to the 'knee' joint in the Chorus paper?

Kelpie the Artega is a production road car with an alloy tub style centre section and a rather odd rear end. It does seem to lack some tubes. The Saleen is difficult to see. It is clothed in alloy honeycomb panels, much as I plan to use on parts of my car. The Mosler has tubes going everywhere! I like how they have gone to considerable effort to get the rear shocks upright.



#138 ozdude

ozdude
  • Member

  • 30 posts
  • Joined: August 09

Posted 11 January 2012 - 06:53

We used to bend 3/4 and 1" aluminium square section tube for wheelchair frames. The fixed part of the tube was clamped similar to round tube. The section to be bent was held between plates that moved with the bending mechanism to stop the tube from spreading. Fixed to the mechanism, on the inside curve of the tubing was a thin steel 'tongue' with a rounded edge that pressed a neat rolled groove into the inner curve.

The bending lever had a simple roller to bear against the outside of the tube. It worked well, and could bend beyond 90 degrees if needed. I think this was similar to the way furniture tubing was bent.

Cheers, ozdude

#139 NeilR

NeilR
  • Member

  • 478 posts
  • Joined: October 09

Posted 14 January 2012 - 11:05

engineer I use bends SHS all the time. I will call down there soon, so will take a pic of the bending setup.

Advertisement

#140 Kelpiecross

Kelpiecross
  • Member

  • 943 posts
  • Joined: October 10

Posted 14 January 2012 - 13:00

gone to considerable effort to get the rear shocks upright.


You mean - not lying down with bellcranks etc.? Which photo are you referring to?

#141 NeilR

NeilR
  • Member

  • 478 posts
  • Joined: October 09

Posted 14 January 2012 - 22:07

the black tubular chassis, rear view. Rear shocks are mounted on outlying tetrahedrons.

#142 Kelpiecross

Kelpiecross
  • Member

  • 943 posts
  • Joined: October 10

Posted 15 January 2012 - 04:02

the black tubular chassis, rear view. Rear shocks are mounted on outlying tetrahedrons.


I see now what you mean. Reminds me a little of the original photo in Cheapy's "Clever" thread.