# Engineering question about square steel tubing

I've been playing around awhile with calculators on the web and whatnot, but since I don't know what I'm doing, what is the maximum (failure) load
that I can put on the end of a 21 cantilevered piece of 2" standard square structural tubing, 1/4" wall? How about to the point of permanent deflection (unrecoverable bend) What is the safe working load that can be applied? I'm trying to find out what the maximum I can go with for a C press I'm thinking about making. I'm sure I need to offer some more information, but like I said, this is stuff I rarely to never delve into.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 11 Aug 2005 03:38:18 GMT, "carl mciver"

Others will know more about this than I, but here's what I figure using formulae from Machinery's Handbook:
A load of 1302 lb on the end of your 21" cantilever would result in deflection of 0.152 inches and max stress of 30,000 lb/in^2. I used 29 million PSI as E, modulus of elasticity. I think 30,000 lb/in^2 stress is below the yield point for structural steel, and I seem to recall that it's one figure that is used as rule-of-thumb permissible stress (before safety margin) for structural steel.
What is "safe" depends on what you want to use for safety margin.
Load that would permanently bend it would depend some on the particular piece of steel used, might be twice the load stated above.
Max stress will be at the constrained end of the cantilever -- so the joint design there (weld, bolts, clamp, gussets or not, ???) will play an important role.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Without going into alot of math and using the area for square tube with 90 degree ends I came up with a quick ball park figure of around 947 lbs. However, be advised that steel tube has radiused ends so your area will be a little smaller. The other poster is right about max stress being at the fixed point of the beam as a moment and shear diagram would show you.
Also, you need a safety factor of at least 4 to 1 to work safely so your acceptable load now at the end of the beam is 237lbs.
Thanks, Steve
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
In article <KKzKe.4201\$WD.3671
says...

If you're talking about loads of more than a few hundred pounds, you may be asking the wrong question. The design of a C-frame press is almost always driven by deflection, and the member that will contribute most to deflection is the vertical column. The top and bottom of the C will be subject to a moment that increases uniformly from the load point to the attachment point. The vertical member must carry this moment thru its entire length and will be bowed considerably by this moment. The effect of this bow is magnified by the length of the horizontal members (your 21").
Look at an arbor press or open sided punch press (or even a heavy C-clamp) with this in mind and the reason for the deep back column becomes obvious. It also points up the reason that most heavy shop presses are H-frames, where the vertical columns are in pure tension.
Ned Simmons
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

| If you're talking about loads of more than a few hundred | pounds, you may be asking the wrong question. The design of | a C-frame press is almost always driven by deflection, and | the member that will contribute most to deflection is the | vertical column. The top and bottom of the C will be | subject to a moment that increases uniformly from the load | point to the attachment point. The vertical member must | carry this moment thru its entire length and will be bowed | considerably by this moment. The effect of this bow is | magnified by the length of the horizontal members (your | 21"). | | Look at an arbor press or open sided punch press (or even a | heavy C-clamp) with this in mind and the reason for the | deep back column becomes obvious. It also points up the | reason that most heavy shop presses are H-frames, where the | vertical columns are in pure tension. | | Ned Simmons
The vertical column is 3" square tubing, 1/4" wall, so deflection there isn't as much of an issue as the horizontal part there. The connection to the 2" tubing is that the vertical tubing sort of wraps around a horizontal piece of 2 1/2" tubing, plus gussets, and the 2" tubing is pinned into that piece at different spots, and allowed to slide in and out to change the throat. At higher loads the tubing will be retracted and the throat will be shorter, so the answer is really not the maximum I can put on the press, but what is the highest amount that I can work with when at the weakest point. Does that make sense? As the load needed goes up, I reduce the throat which reduces flex. The plan is to ballpark what the load maximums are for each setting, which at the longest is 21" throat and every two inches it can be pinned, to a minimum of about an inch, which is a throat of about 13". I failed to clarify, now that I look at my drawing again (duh!) At a 21" throat, the tubing will be "out" by about 10 inches, so that really changes the numbers. The difference between extension and throat is the depth that the tubing it slides into is part of the throat dimension. I'm not considering the flex in the joint or the female tubing, since it's heavily gusseted and makes for much more complex calculations.
The reason I'm doing it this way is that half the time I've needed a press I couldn't get the part in a traditional H press (why I haven't a clue,) and that made things a real PITA. This is sort of a combination of possibilities, I guess, and of course takes up a little less room than an H press, at least as far as width is concerned. I just wanted to get a better idea of what it was capable of. It's a sort of uprated version of what Grant Erwin's got on his site at http://www.tinyisland.com/PMarborPress.pdf with an adjustable throat.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 11 Aug 2005 19:15:30 GMT, "carl mciver"

I've got a real good idea of what you're doing. In fact I've got a portable press I made along the same lines. However I think you're underestimating what it'll take. I made my press frame from two pieces of 8" channel welding up into a tube. Then my arms have 3" square tube sockets that take 2 1/2" attachments. The arms are very heavily gusseted. I've only got about 13" overhang to the center of the jack when setup. I can get a good 1/2" of flex in the system with a weak jack that I have in it right now. The jacks supposed to be a 20 ton but it's leaking back so badly that I can't lift the front of a diesel truck with it right now so figure I'm not able to get more than 5-10 tons on it.
Wayne Cook Shamrock, TX http://members.dslextreme.com/users/waynecook/index.htm
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Wayne, I was poking around your site, seeing if there was either a direct picture of it, or just it in the background of something, but nothing turned up. Got some pics about!
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 11 Aug 2005 23:24:21 GMT, "carl mciver"

I'll see what I can do. It's in pieces right now stacked near my drill presses. But the pics of the drill presses are old enough that it's not in there. My big problem right now is time. I've got a big stainless fertilizer spreader in the shop that's been giving me a hard time. It's been there way to long but between interruptions and equipment break downs I'm having trouble getting it out.
Wayne Cook Shamrock, TX http://members.dslextreme.com/users/waynecook/index.htm
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
In article <mtNKe.4373\$WD.1860
says...

I think you'll find that even though the 3" tube is approx 4x as stiff as the the 2" tube, it'll contribute more to the deflection than the 2" tubes, unless it's very short. I guess all I'm saying is that effort put into stiffening the vertical member will have a bigger payoff than stiffening the arms.

Don's answer looked good, except I'd reduce the stress in the steel to 22000 psi, which is what the AISC prescribes as design stress for A36 plain vanilla structural steel, so around 800# at 21".
Another thing that may not be obvious is that the deflection of the arms by themselves is a function of the cube of the distance from the verical member to the load. The deflection at the load point caused by the bowing of the vertical member will vary as the square of the distance to the load. This is because the bending moment in the arms increases uniformly as move towrds the attachment point. The bending moment in the vertical member is constant, and equal to the max in the arms, over its entire length.
Ned Simmons
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
| > >>SNIP<<| | | Don's answer looked good, except I'd reduce the stress in | the steel to 22000 psi, which is what the AISC prescribes | as design stress for A36 plain vanilla structural steel, so | around 800# at 21".
Thank you very much for a well thought out response. I was doing some more thinking about load paths and realized that I made another error. The tube I brought up is, for the most part, not in the load path, other than providing a leverage point out at the end. I plan on reinforcing this part on the top and bottom with straps top and bottom on the exposed part. The load path is as follows: At the top of the vertical part of the C, resting on the slip sleeve for the aforementioned tube, will be a hydraulic jack. The jack pushes up on another 2" tube reinforced with a 3/16" or so strap on the bottom and top, and/or with a doubled tube even, and to provide a surface for the jack top to bear on. Near the other end is a pair of holes that a vertical link connects to from it the tube first discussed. At the very end it's pinned to a vertical tube, full of holes to make it adjustable up and down. So the jack pushes the lever up, levering it against the link and pushing the other end down to do the business needed. The leverage factor is from 1:1 up to 7:1 depending on how far you've got the tubes positioned, and which pair holes the link is using. At the ends of the lever there will also be side plates, because the forces in this short area will be pretty high, considering the leverage. Should I just post my drawing and make it that much easier to explain, I feel like I'm making everyone try hard to imagine it as I describe it?
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
In article <KmWKe.4294\$Wi6.1087
says...

A picture would be helpful. I've formed a picture in my head of your press that may not look much like what you're imagining.
Ned Simmons
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

| A picture would be helpful. I've formed a picture in my | head of your press that may not look much like what you're | imagining. | | Ned Simmons |
It was about time I did that! They are
http://metalworking.com/DropBox/PressDsn.jpg and http://metalworking.com/DropBox/PressDsn.txt The text didn't come out as well as I would have liked, but it's the idea that's most important.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
In article <yrfLe.5231\$WD.2303
says...

I was wondering at first why you weren't just using the jack's ram directly, but I can see a few advantages to your design, the most obvious being flexibility in mounting tooling.
The first thing I would suggest is to replace the single tube with gusset that you're using as the column with a pair of tubes spaced apart under the horizontal sleeve. This'll make a huge difference in the stiffness and strength of the column. (I hesitate to call it a column because columns are normally thought of as carrying compressive loads, while this member, even though vertical, is subject to bending loads that more resemble a beam.) To really stiffen it up you could add a web (or triangulation) between the two tubes, at which point you've got something that resembles an I-beam with tubular flanges (or a bar truss, in the case of triangulation.)
Another thing that occurs to me is the possibility of a lever that you can actuate manually in place of the jack. Sometimes the feel of a manual press is preferably to having lots of force available.
Ned Simmons
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
| > It was about time I did that! | > They are
http://metalworking.com/DropBox/PressDsn.jpg and | > http://metalworking.com/DropBox/PressDsn.txt | > The text didn't come out as well as I would have liked, but it's the | > idea that's most important. | > | | I was wondering at first why you weren't just using the | jack's ram directly, but I can see a few advantages to your | design, the most obvious being flexibility in mounting | tooling.
That's the plan. This can do the job of a much smaller press, even an arbor press, as well as something much heavier duty, all with the same tool. I like the versatility and real estate saved in my garage. Being able to change the press tonnage from nothing to incredible easily appeals to me as well. I can use a one ton jack just as well as I can a twenty ton, and they just sit in same place since they're all roughly the same height.
| The first thing I would suggest is to replace the single | tube with gusset that you're using as the column with a | pair of tubes spaced apart under the horizontal sleeve. | This'll make a huge difference in the stiffness and | strength of the column. (I hesitate to call it a column | because columns are normally thought of as carrying | compressive loads, while this member, even though vertical, | is subject to bending loads that more resemble a beam.) To | really stiffen it up you could add a web (or triangulation) | between the two tubes, at which point you've got something | that resembles an I-beam with tubular flanges (or a bar | truss, in the case of triangulation.)
Ah, excellent idea. Would the same be gained by just welding two tubes to each other? The inner tube would be in tension under load, while the outer tube would be in compression. I just had the notion that I could fill the outside, "unused" tube with concrete, which resists compresssion (and buckling the walls of the tube,) but I'll have to cap the ends off, of course. Then again, one or two small I beams or channel between them sounds pretty strong too, with the outer tube full of concrete would be good. Resists twisting really well, too. Whaddya think? Gusseting the horizontal tube would get interesting, of course, but most of that load just passes around it, rather than the whole thing being a cantilever beam. There's a hard to read note about the column "wrapping around" that tube, which passes the load around it nicely.
When I first starting thinking about this, before I was aware of the PM design, I had a much simpler design using an I beam as the column, but I was concerned about it twisting, so was kinda stuck. You've helped me work that out.
| Another thing that occurs to me is the possibility of a | lever that you can actuate manually in place of the jack. | Sometimes the feel of a manual press is preferably to | having lots of force available.
That's one of those details that was so easy to implement, I just noted it in text, and the text pretty much washed out in the scan. With a double tube lever, I'd just weld on a couple pieces of tubes to the side of the top one and slip in a piece of pipe to suit. This way it's just as easy to use the press in manual mode or when necessary, just reach over and start pumping the jack, since the jack isn't attached to the upper lever. There's many times I suspect that the jack's job is over, and the rest of the operation requires less pressure, so all you have to do is just reach up and pull.
| | Ned Simmons
Thanks for sticking with me this far through it! Your input has been invaluable.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
In article <_1ALe.5598\$WD.5081
says...

I figure this would be about 4.5X as stiff as the single tube. If you were to separate the two tubes with a web of 1/2x2 flat bar it would be 12.6X as stiff.

The farther you separate the tubes, while restraining them from shifting relative to one another, the closer you'll get to this ideal. In an idealized I-beam one flange is in tension, the other in compression, while the web is in shear as it resists the tendency for the flanges to slip relative to one another.

Unless you prestress the tubes as the concrete cures, I don't think you'll see any benefit. And even if you got it to work, I doubt it would make enough difference to make it worth the trouble.
Then again, one or two small I beams or channel between them sounds

I'm afraid I'm not following this. Are you saying that the horizontal tube is not carrying the force being applied by the press ram? The situation is not as simple as if the jack were at the end of the arm and acting directly, but the horizontal tube is still carrying a considerable load.

My pleasure. It's an interesting idea.
Ned Simmons
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

| I'm afraid I'm not following this. Are you saying that the | horizontal tube is not carrying the force being applied by | the press ram? The situation is not as simple as if the | jack were at the end of the arm and acting directly, but | the horizontal tube is still carrying a considerable load. |
Okay, I had to draw it out in a line diagram to see it. The link will be pulling up on the horizontal tube, so the force will be a cantilever up. The vertical ram passes through the sleeve at the end, so it can't flex too much or it will bind. Maybe it would be worth it to shorten that sleeve a lot. I had to remove the connection between the horizontal tube and the vertical ram to get that out of my head, then the large up forces show up as a simple lever. I can up the sizes of the tubing to make the horizontal stronger, since it's a slip joint. If I were to weld in straps on the upper and lower sides of the tube using plug welds, it wouldn't affect for the slip joint, but will stiffen things up in the vertical direction. I had thought about doing that with an entire second piece of tubing telescoped and welded inside, so maybe that's an option. Or both, as long as I can get a pin in it. Any good ideas in that respect?
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

it's the

even an

same tool.

able to

to me as

and they

two tubes

the
could fill

(and
of
them sounds

good.
but most

being a

"wrapping
the PM

but I was

work that

just noted

double
the top

to use

start
There's
the
reach up and

been
One problem I see is the jack might have to be on pivots if the ram is going to move any appreciable distance. The pivoting arm isn't going to stay parallel with the pad the jack is sitting on, and the contact point along the pivoting arm will change too. I suspect the ram may tend to bind in the socket, too.
I'd be tempted to make a wood or cardboard model of the linkage first
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

but
ton,
while
compresssion
off,
course,
of
a
of
easy
lever.
has
is
contact
first
On second thought, having the jack on pivots would introduce a force along the axis if the pivoting arm once the arm was not parallel. A roller on the top of the jack would reduce that..
The linkage reminds me of that found on a pitcher pump....
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

## Site Timeline

• Share To

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.