blueshift
|
| posted on 20/5/04 at 11:09 PM |
|
|
How best to position tubes at a vertex?
Hello, welding theory question. Rather than try and describe it, which is better out of these two options; red or blue?
tia
|
|
|
|
|
mangogrooveworkshop
|
| posted on 20/5/04 at 11:21 PM |
|
|
IMHO Red for better load spread !
|
|
|
blueshift
|
| posted on 20/5/04 at 11:51 PM |
|
|
That's what I would guess, but lots of welding tidbits have come up on here that are contrary to what I would have guessed as common sense..
hmmm
|
|
|
Alan B
|
| posted on 21/5/04 at 01:35 AM |
|
|
Yes, red...in theory because it allows all the tube centrelines to converge.
In reality it doesn't matter that much.
|
|
|
Cita
|
| posted on 21/5/04 at 05:00 AM |
|
|
I would go for blue because the weld is longer/stronger in one plane.
The centerlines of the tubes imerge when the diagonal is under 45° wich is only the case when a square is braced.
Imagine 3 tubes layed out in a "Z" form and think for yourself what would be the strongest way to connect the 3 tubes,the
"blue" or "red" option.
|
|
|
JB
|
| posted on 21/5/04 at 06:15 AM |
|
|
Tube Junctions
The red one is the better option.
At a mulitiple tube junction all the tube centrelines should converge at the centreline of the main tube.
With the blue option the force from the diagonal (ie the blue member) is all going into the horizontal piece. The weld to the main member is then
loaded in compression and tension. Welds should not be loaded in tension, welds should be loaded in shear.
The red option loads the welds in shear and distributes the load between two members.
JB
|
|
|
Cita
|
| posted on 21/5/04 at 07:01 AM |
|
|
How do you manage to converge all centerlines of the tubes on the main tube without changing the geometry of the rectangle?It seems to me that sooner
or later you gonna end up with the blue option.
No critisism here,just want to learn.
|
|
|
Peteff
|
| posted on 21/5/04 at 08:59 AM |
|
|
If you've already made the outer frame before you put the diagonals in you can't change the geometry of the rectangles. You will have to
make them to fit in the spaces already there. I use the blue option as it's easier.
yours, Pete
I went into the RSPCA office the other day. It was so small you could hardly swing a cat in there.
|
|
|
blueshift
|
| posted on 21/5/04 at 11:36 AM |
|
|
I'm doing my typical thing of worrying about the tiny details before doing anything.. so we haven't actually started on the chassis
yet.
so I can choose to do what I like 
I think the red option wins, unless mark allanson wants to come along and show me I know nothing again.
ta
|
|
|
JB
|
| posted on 21/5/04 at 12:01 PM |
|
|
The main tube is the vertical one. Draw a centre line on this tube, and then draw a line on the horizontal tube.
These 2 cetntre lines will meet (at the centre of the vertical tube).
Any diagonals (which will be added after the previous two tubes) should be position so their centre line meets on the above point.
Another point: When designing a chassis the aim is to load all tubes in tension or compression (NOT BENDING). The blue option will put a bending load
into the horizontal member.
For further reading on this : Prepare to Win by Carroll Smith.
Race Car Chassis Design and Construction by Aird
JB
|
|
|
Cita
|
| posted on 21/5/04 at 12:07 PM |
|
|
Blueshift,if you think the red option is the best solution can you please explain me how you gonna achieve the converging of the centerlines in a
given,or any rectangle?
It sure looks ugly to change from one option to another in the same frame
|
|
|
Cita
|
| posted on 21/5/04 at 12:24 PM |
|
|
i just made a small test drawing in autocad.
A rectangle 1m x .30m in .25m RHS
The red option is....nearly gone up into the...blue option with all centerlines converged!!!!
You almost have to come to a square to get a noticable diff between the red en blue option.
|
|
|
stephen_gusterson
|
| posted on 21/5/04 at 12:51 PM |
|
|
most of the chassis, once panelled - especially if you do the inside as well - is so covered you wouldnt notice style differences!
atb
steve
|
|
|
Cita
|
| posted on 21/5/04 at 02:15 PM |
|
|
If the inside and outside of the frame is covered with sheet than you maby can leave the diagonals out!
Problem solved
|
|
|
blueshift
|
| posted on 21/5/04 at 05:05 PM |
|
|
quote:
Originally posted by Cita
Blueshift,if you think the red option is the best solution can you please explain me how you gonna achieve the converging of the centerlines in a
given,or any rectangle?
I don't see why it should be a problem: work out the angles involved or trial fit and mark up, cut from the centre line of the tube out to the
edges, two diagonals, file to fit.. it won't be perfect but it would be better than blue.
|
|
|
Cita
|
| posted on 21/5/04 at 06:01 PM |
|
|
The problem is that when you want those centerlines converge in one center you will end up most likely with one small side and one large side and the
diff between those two sides will be so great that you...just as well could be gone for the blue option in first place
|
|
|
Peteff
|
| posted on 21/5/04 at 08:00 PM |
|
|
This would have been even easier but how would it perform compared to the others?
 
Rescued attachment barjoint.jpg
yours, Pete
I went into the RSPCA office the other day. It was so small you could hardly swing a cat in there.
|
|
|
Mark Allanson
|
| posted on 21/5/04 at 09:22 PM |
|
|
I'm with JB on this one, the idea of triangulation is to focus all the loads at specific nodes. If the nodes are not fully bisected, you get
leverage, and then stress fatiguing
OK the chassis would be minutely less rigid, probably not measurable for a locust, but in something a bit more highly stressed, it could be a factor.
I made an overhead crane for the CEGB at Hinkley Point, they were fussy!
If you can keep you head, whilst all others around you are losing theirs, you are not fully aware of the situation
|
|
|
blueshift
|
| posted on 21/5/04 at 10:12 PM |
|
|
It will please me to know that our chassis is put together in the best way possible, even if it means fiddlier fitting and filing. Thanks for comments
all.
|
|
|
crbrlfrost
|
| posted on 22/5/04 at 06:03 PM |
|
|
Good thread, I like the points brought up. I too would have to agree that the red tube will be better in loading, so would shoot for it on most my
bracing (or did as the case may be) but I often used the blue style on lightly loaded or non-structural sections. Cheers!
|
|
|
Cita
|
| posted on 22/5/04 at 07:11 PM |
|
|
IMHO for a locost it's a waste of time and energy to go for the red option.The loads in the locost frame are so light that the gain would be
inperceptable.
The pick up points for the suspension are "floating" so why go through the hastle of difficult, time consuming bracing of the frame.
The most important thing however is that the builder feels good about what he has built wether it be the red or blue option.
|
|
|
Spyderman
|
| posted on 23/5/04 at 01:03 PM |
|
|
quote:
Originally posted by Cita
IMHO for a locost it's a waste of time and energy to go for the red option.The loads in the locost frame are so light that the gain would be
inperceptable.
The pick up points for the suspension are "floating" so why go through the hastle of difficult, time consuming bracing of the frame.
The most important thing however is that the builder feels good about what he has built wether it be the red or blue option.
If the loads are so light then why bother with the cross bracing?
It is only in extreme measures that you realise how good your structure really was, or others ponder over how such an unsafe vehicle was allowed onto
the roads!
Not everything is always immediately apparent.
Build it as safe and as strong as you can or you may not live to regret it!
Something built in the correct way will always be safer even if the welds are substandard as the welds themselves are taking less strain.
I understand what you are saying about the strengths etc, but it is always better to encourage correct methods rather than easy ones!
Feeling good about what you have built has nothing to do with which is correct or not!
All IMHO of coarse!
Terry
Spyderman
|
|
|
Cita
|
| posted on 23/5/04 at 01:21 PM |
|
|
Thanks for the reply.it is not my intention to encourage unsafe or dangerous methods if that appears to be the case than i'm sorry for this
'cause it would be wrong to do so.
All i'm saying is that, first the drawings in this thread give a somewhat wrong impression in that all the diagonals are set at an angle of 45°
witch only occurs within a square.How many true squares are there to brace in a locost?
Everybody who's jumping on my neck(no offence taken)doesn't seem to have taken the time to actually draw an example and see how strange
things can turn out and even if it look like the blue option was used, it actually was the red option.
|
|
|
britishtrident
|
| posted on 23/5/04 at 02:27 PM |
|
|
When you think about it very carefully for a chassis made of RHS there are grounds for suspecting the eaiier blue option of actually being
stronger.
|
|
|
NS Dev
|
| posted on 23/5/04 at 02:50 PM |
|
|
There are always grounds for thinking in all sorts of ways about this stuff, as Mark Allanson says, you can be fussy or not, the choice is yours. If
you have bought a locost chassis, odds are this is irrelevant because this kind of analytical thinking will most probably not have gone into it,
that's why you can buy it for £450 or so! To take it on a stage, with either option, if you MIG weld it, unless you are an exceptionally good
mig user (and Mark Allanson, you may well be if maiking cranes for the CEGB, as you say they are fussy!) you will not notice any difference simple
because a MIG is not controllable enough.
If on the other hand you use a TIG for pretty much everything like JB does, you may see the difference (i.e. there may be a difference)
|
|
|
