Offset casters for large cabinet

admin

I am adding casters to a large 3'x14' cabinet/assembly table. Small casters don't do well in the shop, so I decided to use 5" casters. But this will make the table too tall. I'm putting 6 wheels under it, two on each end and two in the middle. This matches the structure of the cabinet. The cabinet is heavy, fully loaded in the neighborhood of 500+ lbs.I've decided to bend some 60 degree offsets which will allow the larger wheels without raising the table too much. I'm not concerned about them sticking out as the table top has some overhang. I was thinking I can make the offsets out of some 1.5"x0.375 flat bar (that I have) but now I'm thinking that may be a little on the light weight side. Should I go to 0.5"? Should I up the width to 2"?Does anyone know how to calculate the capacity of an offset bend? I know how to calculate the tons required to bend it in the press brake. How does this correlate to the load bearing capacity? I've put two renderings of the offsets, one in 0.375" and the other 0.50". The wheel flange is 3.5" and the offset flange is 5". The lower short flange extends under the structure to the opposite side.I've noodled on these wheels for a long time but always been unsure if the offset could do it. I have the wheels so I'm building it this week. Attached Images
Reply:Put a gusset between the base & side of the cabinet.  Weld the gusset to the cabinet & backbone of the offset.  Then you can use the lighter material.MarkI haven't always been a nurse........Craftsman 12"x36" LatheEnco G-30B MillHobart Handler 175Lincoln WeldandPower 225 AC/DC G-7 CV/CCAdd a Foot Pedal to a Harbor Freight Chicago Electric 165A DC TIG PapaLion's Gate Build
Reply:Originally Posted by MarkBall2Put a gusset between the base & side of the cabinet.  Weld the gusset to the cabinet & backbone of the offset.  Then you can use the lighter material.
Reply:Use wood for the gussets.  Bolt them to the cabinet, then run a couple screws through the bracket into the gusset.  The gusset will be in compression, so it shouldn't compress.MarkI haven't always been a nurse........Craftsman 12"x36" LatheEnco G-30B MillHobart Handler 175Lincoln WeldandPower 225 AC/DC G-7 CV/CCAdd a Foot Pedal to a Harbor Freight Chicago Electric 165A DC TIG PapaLion's Gate Build
Reply:I recently made some drum dollies for some 55 gallon drums.  I used 3/8" x 2 1/2" flat bar bent to 90 degrees instead of 60. These drums are used for oil changes and are filled with used motor oil and are wheeled around an automotive shop.  I figure the weight of the drums when full are around 450 lbs and they had no problems with them.  I also built some dollies for a couple of 25 gal drums using a design like yours using the same material.  The only way I could test these at home is by standing on them and weighing about 240lbs there was really no noticable give and with you using 6 caster, sharing the weight, then its only hundred pounds per caster.  I say you would be fine if you use 2 1/2" x 3/8" material.
Reply:I built something similar with a partial gusset.  Still very springy (but works great for my application):http://weldingweb.com/vbb/showthread.php?t=40000The Lord has declared, "This is my work and my glory--to bring to pass the immortality and eternal life of man"  Moses 1:39Link: My name is John, and I am a member of The Church of Jesus Christ of Latter-Day Saints.-- ColdCreekWorks.com --
Reply:Alright, this offset question has me bugged. So I downloaded the FEA module for my design software, ripped through the tutorial and pushed my model through. Quick and dirty.I made some assumptions. The bottom flange is fixed and all the force was upwards around the caster mounting hole. I used 250 lbf as my factor because the casters are rated for 250 lbs. Maybe I goofed it up but it look right to me. The first image is the 0.375" material and the second is 0.50" with a width of 1.5". The third image is 2" wide by 0.50". I'm definitely leaning to the half inch now  Attached Images
Reply:Originally Posted by jdh239I built something similar with a partial gusset.  Still very springy (but works great for my application):http://weldingweb.com/vbb/showthread.php?t=40000
Reply:Originally Posted by Jay OI recently made some drum dollies for some 55 gallon drums.  I used 3/8" x 2 1/2" flat bar bent to 90 degrees instead of 60. These drums are used for oil changes and are filled with used motor oil and are wheeled around an automotive shop.  I figure the weight of the drums when full are around 450 lbs and they had no problems with them.  I also built some dollies for a couple of 25 gal drums using a design like yours using the same material.  The only way I could test these at home is by standing on them and weighing about 240lbs there was really no noticable give and with you using 6 caster, sharing the weight, then its only hundred pounds per caster.  I say you would be fine if you use 2 1/2" x 3/8" material.
Reply:Forhire,A couple thoughts after looking at the FEA models you posted screen shots from.The screen shots are showing you the stress in the part.  See if you can get a look at the strain or distortion instead.  Your software should be able to give you this.Looking at the stress distributions, it seems to me that you're model is assuming that the load is a point load, all the way at the right-most tip of the bracket.(This is where the highest stress levels appear) I think a more accurate model would use the load is really distributed over teh entire 'foot' of the bracket, with the highest stresses showing up at the first bend.  The point where the bracket comes out from under the cabinet.Your models have revealed a basic design principle.  Generally speaking, shape or form factor has a much greater influence on stiffness than material.You might want to approach this design question from a different point of view.  Instead of looking at the stresses involved, decide how much strain or distortion is tolerable.  Say no more than +/- 0.1" vertical movement at the left most end of the bracket, where the caster sits.Yield strength for A36 steel is 36KSI minimum.  Use one half this(or 18KSI) as the maximum allowable stress, since you don't want the brackets to yield at all and x2 is a reasonable factor of safety.  Remember you're modeling static load, and that dynamic loads are going to happen.  Say you're pushing this cabinet, and a single wheel strikes a rock.  Now all the kinetic energy that the cabinet has is applied to that one bracket.  The factor of safety gives you some cushion for these type of events.  Engineers will debate how much of a factor of safety is needed.  x2 is just my guess, YMMV.With a thesefactors, you can find the ideal width and thickness for the steel strip used in the bracket.  You'll have some assurance that the brackets won't fold up if you crash the cabinet into something while moving it.  Someone else suggested that you bolt a wooden block to the angled portion of the bracket.  This would definitely brace the wheel, and cut down on any fatigue in the bends.   Think of it as an up side down wedge between the bracket and the wall of the cabinet..Benson's Mobile Welding - Dayton, OH metro area - AWS Certified Welding Inspector
Reply:Originally Posted by forhireThanks for the link. I saw this the other day and that's what got me thinking about this design again.For fun I modeled your design. I assumed 3" wide by 3/16" thick. I didn't model the gusset. I used 75 lbf for this model. I assumed the compressor weighs about 250 lbs and I rounded up a bit for vibration. I had assumed that the flange end would be the issue but it appears from the model that the bends actually stiffen the part. I bet if you added a piece down the length from the center it would remove some of the spring.This is making my head spin
Reply:Originally Posted by forhireI bet if you added a piece down the length from the center it would remove some of the spring.
Reply:At two inches wide and made of 0.5 inches thick steel. Each one will deform or bend at about 10,000 pounds. They will flex well before that. I know from bending them in a "V" break. What the hydraulic pressure and surface area of the piston creates upon the metal to bend them. It takes about a constant 13,000 pounds to keep them constantly bending. But at about 10,000 pounds there is permanent damage. What most people do not see. Is the 500 pound loads velocity. As it is moving along at four miles an hour. And hits a half inch steel plate on the floor. Trying to send as much or more then 500 pounds that suddenly is moved to that single jammed wheel. The velocity created is almost straight up, over a quarter inch of travel. It can make that load 10 times what it was or much more, because of the speed at which the pressure is applied. It becomes a hammer or shock load. Very hard to know what you need.        Sincerely,             William McCormick
Reply:A_DAB_will_do - Good catch on my mistake with the fixed flange. Now the stress is in the correct location, typical learning curve. The deflection on the earlier models is about 5/16" which may make it feel springy. jdh239 - I'm using Alibre Design which has a stripped down version of AutoDesk Algor for the FEA modeling. http://www.alibre.com/products/fea.aspWilliam - I'm hoping to build in enough margin that it can tolerate a little junk on the floor Attached are two more models. I've gone up to 0.5" x 3" bar and decreased the load to the expected 100 lbs per caster. For me this is a design exercise, I'm sure I could build it using gussets and such. I on occasion have a need to use offset wheels and knowing the limitations will help me build better. Looking at it, had I used a 3" caster the offset would have been much smaller and easier to hit my target capacity First image is stress and the second is displacement. Enjoy. Attached Images
Reply:A piece two inches wide and a quarter inch thick, will require a constant 3,600 pound force to keep it bending very, very slowy. But will bend and be ruined at around 1,600 pounds. It will flex at 500 pounds.       Sincerely,             William McCormick
Reply:Both joints are going to flex about half the total flex with the long connecting piece. Taking up the total pressure put on the bracket. Both joints will have the same pressure on them. But they each will flex half of the total flex. If you do not drop the table from a distance, you will not see the flex at 100 pounds on the 2" by 1/2" bracket. Even with 500 pounds you will not see the flex. Unless you do wheelie's or jump busses. If you see flex more then likely it is in the mount, not the bracket. What could you possibly mount this to? That will not flex more?If you race the table around and hit things you might flex it. Or if you load up the table with a lot of stuff to take across the shop and hit something, or drop it into a hole. Lose a wheel rubber. You might bend them, but even at two inches that is probably overkill, if the table will only see a maximum of 500 pounds. Just look at the flanges on a standard or wide flange beam. And what the trolley weight they can carry is. I think in this case the program is sending you way out in the other direction at only a total 500 pound load.  The thing that will bend it, is the hammer or rawhide shot. If you take a piece of steel a quarter inch by 2", in the size you are proposing, I can put it in the vise and bend it up for you with little shots from a medium to small handheld rawhide mallet. Not even work up a sweat. I know it takes almost a ton of slow pressure to bend. Yet Just small shots do it. If you might load a bracket to 5,000 pounds, and then drive it over bumpy terrain I would go with the 3" bracket otherwise the two inch one should do it well. The pressures when you go from quarter inch steel to half inch steel, to bend it, is almost 4 times greater.        Sincerely,             William McCormick
Reply:Originally Posted by William McCormick JrA piece two inches wide and a quarter inch thick, will require a constant 3,600 pound force to keep it bending very, very slowy. But will bend and be ruined at around 1,600 pounds. It will flex at 500 pounds.
Reply:Originally Posted by William McCormick JrIf you do not drop the table from a distance, you will not see the flex at 100 pounds on the 2" by 1/2" bracket. Even with 500 pounds you will not see the flex. Unless you do wheelie's or jump busses.
Reply:I am good at overkill.  If you want something built, I can sure make it heavy and resilient.  The trick is to build something just strong enough for your needs (trying this with my cargo carrier)The Lord has declared, "This is my work and my glory--to bring to pass the immortality and eternal life of man"  Moses 1:39Link: My name is John, and I am a member of The Church of Jesus Christ of Latter-Day Saints.-- ColdCreekWorks.com --
Reply:Last night I emailed my uncle the engineer. He sent me some equations and verified math. He cautioned me about using FEA software as it can make dangerous assumptions.  He did say the models look good.I decided (along with my uncle's blessing) that 0.5" x 2.5" was a good compromise. Not too much deflection and enough capacity. Tonight, I cut some scrap and bent up a test part. Even though my scrap piece had some thin spots it was very strong.  I was able to deflect it with a long wrench in the vice but unable to bend it. Each time it returned to the original shape. This is the same results I found using the models.It become very obvious why offsetting the wheels are nice, It's only 1.5" (2" total lift) from the floor with a 5" caster.Now the real question... do I want to spend money on 0.5" x 2.5" bar stock Thank you to everyone who helped with suggestions and critique as I noodled on this. Attached Images
Reply:Originally Posted by jdh239Actually, I like the springiness about it.  It was a side-effect I didn't anticipate, but the springiness helps to absorb the vibrations and makes for a decently quiet compressor (at least compared to my little 12 gallon Craftsman).
Reply:Yeah, 1/2 inch thick plate 2.5 inches wide really isn't going anywhere for a 500 lb load.  Especially when there are 6 of them sharing the load.  As to buying the 1/2 material ...  Do it once and do it right and then you don't have to worry or think about again.  Or fix it or beef it up down the road.You could get 'fancy' and work out some gusseting and then use thinner material, but for the plain bent offset brackets using 'beef' instead of 'fancy' works fine too.  The best laid schemes ... Gang oft agley ...
Reply:I honestly do not think the 0.5 x 2 inch bar will deflect in that configuration .117 inches under a 100 pound load. 500 pounds maybe. I do a lot of machinery moving and we use homemade pry bars. Metal bar, anything to get you under some machinery. And I can lift very heavy machinery with even a quarter inch plate over a two inch span. And not bend the plate. So the half inch plate will certainly do it. When you bent that bar, did you hear your equipment whining? Ha-ha.        Sincerely,             William McCormick
Reply:Everything I was discussing comes from the "V" break. And it is true that in the "V" break your pressure can double. Because you are only putting half the weight of the press upon the two points in the "V" block. So it pays to go heavy. The angle of your bracket makes up for a lot of the stress that a 90 degree bracket would leave you. You are no longer capable of putting the full bending force on either bend. So the flex ends up in the long piece. Right in the center.        Sincerely,             William McCormick
Reply:Originally Posted by William McCormick JrI honestly do not think the 0.5 x 2 inch bar will deflect in that configuration .117 inches under a 100 pound load. 500 pounds maybe. I do a lot of machinery moving and we use homemade pry bars. Metal bar, anything to get you under some machinery. And I can lift very heavy machinery with even a quarter inch plate over a two inch span. And not bend the plate. So the half inch plate will certainly do it. When you bent that bar, did you hear your equipment whining? Ha-ha.I just drew it up quick I did not have a good feel for the actual sizes. When I saw your actual piece. I looked at the size of your tires. And then your bracket at least by estimating. And I see that when the wheels are turned out. That there is over eight inches of leverage. That could easily turn your 100 pounds into 200 pounds. I also believe that it puts a lot more stress on the long bar. So you might need every bit of that 2.5 inches if you load that table up.        Sincerely,             William McCormick Attached Images
Reply:Hope I'm not re-stating something previously mentioned, but here goes...What if you used lighter material, a right angle, and let the assembly flex until it contacts the side of the cabinet, using the cabinet itself as a structural element.just a thought...MillerMatic 252, HTP 221 w/cooler, Hypertherm PM45, Lincoln IdealArc 250 AC/DC"I'd like to believe as many true things and as few false things as possible"
Reply:I had no idea of the size of the brackets. Until I saw your finished product. I assumed they had about 2.5 inch flanges, and maybe a 4 inch long connecting piece. When I saw the done deal, I realized what you were going for. My "V" block is four inches across. So I based it on that. If you could shorten them up a little. It does indeed make a big difference. Like this I could see it working for a total 500 pound load. Jumping small obstacles. Ha-ha. But even with my idea of wheels and only a one inch difference in distance from the bending point. That is pushing it for half inch material for any substantial load.        Sincerely,             William McCormick Attached Images
Reply:Originally Posted by frieedHope I'm not re-stating something previously mentioned, but here goes...What if you used lighter material, a right angle, and let the assembly flex until it contacts the side of the cabinet, using the cabinet itself as a structural element.just a thought...
Reply:Creating the right angle tube connections as mitered joints willhelp you in a bunch of ways---red diagonal lines  Additionally, if possible, nesting this right angle tube inside the anglewill allow for more stress distribution-- black lines     FEA is a great tool to get a feel for the design concept in a static sense.Once even a slight amount of elastic or plastic distortion occurs in a joint,then the statics and dynamics have changed and will continue to change,often quite dramatically. http://weldingweb.com/attachment.php...1&d=1272053518 Attached ImagesBlackbird
Reply:This is not as elegant as your design, but, it would be exceptionally strong.You could bolt it to BOTH the bottom and the sides.PS excuse the crude drawing.  I still trying to learn old 2D Autocad - it is an upgrade from my old Amish Cad (pencil). Attached Images
Reply:To be honest if possible the best is to put the wheels inset and under the table. I have a few hydraulic tables that have the wheels underneath the base. And they are pretty stable. Especially considering they have such a small foot print. And the table overhangs the base. Two wheels are locked straight though for stability. All our large metal storage tables have the wheels just under the corners. But ideally to keep the height low, you could make a support to go inside the table, and that would really do it. You don't kick the wheels while walking past or around the table either.        Sincerely,             William McCormick
Reply:forhire, I feel your pain, reallyI was wondering if you intend to fasten these 6 casters to the bottom of your cabinate, or build a cradle that the cabinate will be placed in and fastened in say 4 places?Matt