Structural integrity of equipment shelf on casters

[matthuszagh]

I recently moved and ended up in a smaller space. I have a large, heavy-duty shelf that I place all my instruments on. This is old HP gear and therefore very heavy. I'd put a rough estimate at about 500-750lbs (225-340kg) total for all gear combined. I'd also like to be able to access the back of this gear without pulling it off the shelf. In my old setup, I allocated about 2.5ft behind the shelf, which was perfectly adequate. But, I can't really allocate that space in the new room. I decided to put the shelf on heavy duty casters so that I can keep it against the wall during normal use and then pull it out when I need to access the back of a piece of equipment.

I've attached an image of what one side of the shelf looks like on these casters. I didn't fully think this through and the structural integrity of it looks concerning to me. In particular, the left side (for which the wheel is not directly below the vertical support) looks like a heavy load would bend the thin gauge metal at the bottom of the shelf.

Some further info: both the casters and shelf on their own are more than enough to support the instrument weight. The casters have a lock that locks the wheel and swivel.

It seems like I have a few options. I'd love thoughts on these solutions, as well as any better solutions others come up with.

1. I can return the casters. I could then get replacements (I'm not sure what might work better) or go back to having the shelf be immobile. But I really like this space efficiency solution if I can make it work.
2. I could fix the wheels in a position in which they're always below the vertical supports. I think I could accomplish this with a zip tie. I could keep that in position with nuts on each thread.
3. I could prevent the screw thread (upper part of the caster) from moving away from the vertical position. I can think of two ways to do this. The first would be to tie the two screws together with probably a number of zip ties chained together. I could get extra strong zip ties for this. The second option would be to have something that maintains the spacing between the screw thread and adjacent vertical support. How to accomplish this is less obvious to me.

I'm slightly leaning toward option 2 if people think it would work. This has the downside that I could only pull the shelf away from the wall and push it back toward it (no motion along the wall). But that seems like an ok price to pay.

One possibility if I go with option one is to drill holes in the bottom of the shelf directly below the vertical support. Then, I could get casters whose wheel is directly below the thread. But, I haven't found anything like this yet.

[james_s]

That offset mounting is putting a huge amount of stress on the bottom plate and the end of the leg, I'm fairly sure it will bend, if it doesn't snap the weld. Can you put a piece of angle iron spanning both of those legs and attach the casters to that? Alternately maybe you could fabricate a mount so that the attachment plate is directly under each leg.

[matthuszagh]

That offset mounting is putting a huge amount of stress on the bottom plate and the end of the leg, I'm fairly sure it will bend, if it doesn't snap the weld. Can you put a piece of angle iron spanning both of those legs and attach the casters to that? Alternately maybe you could fabricate a mount so that the attachment plate is directly under each leg.

To clarify the angle iron idea. A single piece of angle iron would be used for both casters. It would extend from the inside of one vertical support to the inside of the other. I'd drill a hole in it where at the locations of the mounting holes in the bottom of the shelf. Then, the casters would attach to this. Is that the idea? I may be able to do this, but I'm not sure. I only have a simple drill.

In regard to the custom mount. I could probably drill a hole directly below the vertical support and mount the caster there. But, that would still put some torsional strain on the shelf.

[matthuszagh]

Here's a close up of the mount for better reference.

Another question: could ball transfers work here? I probably can't put too much pressure on the wood floor though.

[ajb]

I would definitely put something across both feet of the shelf AND move the casters out--preferably outside of the posts entirely. The problem with where the casters are now isn't just the dinky little feet they're bolted to but more importantly what it does to the stability of the whole thing.  When the caster is swung inward the contact area of the caster gets a lot closer to the center of gravity of the whole rack, which makes it less stable--a LOT less stable if that rack is more than a couple feet tall.  Sure, it's gonna be heavy as fuck with all of those boat anchors, so it would still take a lot of force to tip it, but I still don't like the look of it.  Moving the casters out so the pivots are right underneath of the upright would help, but still gives you a smaller footprint than when the rack was sitting directly on the floor, so if you want the same level of stability you need to move them out so that even when swung inward the *contact area* of the wheel is under the outside edge of the post.  That looks like it would put posts of the casters just a bit outside of the rack frame.

I would probably put a piece of structural channel across the bottom of the rack, with the open side up, and sized so that the feet of the rack fit inside of it.  Channel sections of that size will have substantial fillets on the inside corners, so you'll need to take that into account when sizing.  You could also use box steel, although that will make the rack taller which may not be desirable.  I wouldn't use angle because open sections don't resist torsion very well.  With an angle section, the neutral axis the angle section is closer to the corner, so it will twist more easily when the caster is swung inward than channel which has the neutral axis in the middle.

You may also want to add some diagonal bracing between the uprights and the shelves as well. 

[matthuszagh]

I would definitely put something across both feet of the shelf AND move the casters out--preferably outside of the posts entirely. The problem with where the casters are now isn't just the dinky little feet they're bolted to but more importantly what it does to the stability of the whole thing.  When the caster is swung inward the contact area of the caster gets a lot closer to the center of gravity of the whole rack, which makes it less stable--a LOT less stable if that rack is more than a couple feet tall.  Sure, it's gonna be heavy as fuck with all of those boat anchors, so it would still take a lot of force to tip it, but I still don't like the look of it.  Moving the casters out so the pivots are right underneath of the upright would help, but still gives you a smaller footprint than when the rack was sitting directly on the floor, so if you want the same level of stability you need to move them out so that even when swung inward the *contact area* of the wheel is under the outside edge of the post.  That looks like it would put posts of the casters just a bit outside of the rack frame.

I would probably put a piece of structural channel across the bottom of the rack, with the open side up, and sized so that the feet of the rack fit inside of it.  Channel sections of that size will have substantial fillets on the inside corners, so you'll need to take that into account when sizing.  You could also use box steel, although that will make the rack taller which may not be desirable.  I wouldn't use angle because open sections don't resist torsion very well, and when the caster is swung in perpendicular to the side of the rack it'll want to twist.  Since channel is symmetrical it won't have that problem. 

You may also want to add some diagonal bracing between the uprights and the shelves as well.

This is a nice idea. And good point about moving the casters out. I hadn't considered that. I found this (https://www.mcmaster.com/1262T45-1262T453/). It's a bit pricey (maybe I'd be better served by a home depot or similar) and the 3' long section is about 4" more than I'd want for the 2' wide shelf (it's about 4" from the center of the caster wheel to the outer edge). This is making me wonder though if casters with the wheels directly below the mount point would be better.

Also I should mention that this shelf is quite tall (7') so the stability definitely matters.

[beanflying]

Something like this solution. Weld or get welded a sensible weight plate (3/16" - 5mm) directly under your verticals so you are not putting a torque on the plate and your wheels need to be moved closer to the centre.

Better option is have a look at these sorts of Casters 250kg rated each and wind down feet directly under the centre when you get to where you need to be so the casters are off the ground for much more stability and you also have the ability to level your stand.

Only arrived yesterday but for my new welding table. Plenty on offer on Amazon or evilbay eBay auction: #373720715093 EDIT there is also a slightly smaller (3" high) 150kg set but locally these were the same price.

[matthuszagh]

Something like this solution. Weld or get welded a sensible weight plate (3/16" - 5mm) directly under your verticals so you are not putting a torque on the plate and your wheels need to be moved closer to the centre.

Better option is have a look at these sorts of Casters 250kg rated each and wind down feet directly under the centre when you get to where you need to be so the casters are off the ground for much more stability and you also have the ability to level your stand.

Only arrived yesterday but for my new welding table. Plenty on offer on Amazon or evilbay eBay auction: #373720715093 EDIT there is also a slightly smaller (3" high) 150kg set but locally these were the same price.

These look awesome! Thanks for the recommendation.

[Berni]

Yeah that plate really needs some reinforcement for that sort of load.

My own solution for this was to buy a 42U server rack with caster wheels included. The thing is pretty huge, it is 70cm deep to accommodate even the deepest of test gear and weighs almost 100kg without the cover panels. The side panels are steel plate thick enough for you to stand on and probably weigh 30kg each. The frame is a cube of welded box section and has triangles welded in the bottom where the caster wheels sit, includes bolts that you can unscrew out to jack the rack up from the wheels and level it. The thing is probably designed to hold half a ton of equipment and the smooth bearings in those caster wheels make it easy to move with one hand on smooth level floor (I have it in a brick building, so the floor is very solid). And the best part? I scored it locally for 70€ and a 250km drive both ways. Even came with a power distribution bar, some power cables, shelves, rack slides..etc

It also makes the whole stack of equipment look nicer as its all built into this huge white metal box with all the cable mess hidden inside.

[rs20]

I know some people are irrationally against wood, but I'd consider just bolting/screwing the shelving down to a couple of wooden beams, and then attach the casters to the beams. Mostly because the appropriate piece of wood would be cheap and easily available at my corner hardware store and easily "machined" (i.e. sawed and drilled) compared to steel.

[ajb]

Much as I love McMaster-Carr, they're not the place for stuff like this. Try onlinemetals, or maybe xometry. Or if you have a local metal supplier they often sell drops from custom cut orders, sometimes as low as scrap prices, and you might get lucky. You could probably use aluminum instead of steel if that's easier to find at a good price.

And yeah, you could probably use wood too, if you don't mind the extra height it'll add

[jpanhalt]

1) Materials:  PLaces like Online Metals are very overpriced (2x to 3x) in my experience.  Find a local supplier or distributor.  As one example, last Winter, I got 8 feet of  3"x5" x 1/4" piece of hot roll angle for $50 out the door.  Location: Macedonia, Ohio.

2) As for casters, under heavy and particularly high tall loads, be sure they are double-locking.  That is, the swivel is locked too.  As an example, casters on scaffolds are always double locking.  Do you really need swivel casters?  Also consider low profile machinery moving rollers.

[mazurov]

Just for reference, here's my "overflow rack" for equipment used less frequently. Working very well for about 5 years.

In addition to what others have said, use larger diameter casters. Makes it easier to move around. See next post.

[mazurov]

Next post.

[matthuszagh]

Thanks for all the terrific advice.

My current plan is to use a structural channel below each pair of feet. This will get bolted to the shelf at each of the two existing shelf feet holes. The structural channel will also extend a bit beyond the edges of the shelf. Then, I'll attach leveling plate casters to the structural channel such that when the caster is rotated inward, the center of each wheel is below the corresponding shelf vertical support.

As I understand it, the main stress points will be some torsional strain on the bolts used to fasten the caster to the structural channel and the channel to the shelf. The vertical supports will also experience some torsional strain when the caster wheel is swiveled outside the shelf. I believe these strains will be ok. With regard to the bolts, these strains aren't really particular to my setup. In terms of the shelf, the vertical support is essentially a structural channel itself and I therefore expect it should handle a modest amount of torsional strain well.

Unfortunately, I don't have equations to back up any of this, or that tell me the physical requirements of the structural channels, bolt diameters, etc. But, the radius from the center of the caster mount to the center of the wheel is only 1.5", so I expect this will all be ok. If anyone has a better intuition of the numbers here (or just experience indicating whether this is strong enough) that would be appreciated.

[ajb]

As I understand it, the main stress points will be some torsional strain on the bolts used to fasten the caster to the structural channel and the channel to the shelf. The vertical supports will also experience some torsional strain when the caster wheel is swiveled outside the shelf. I believe these strains will be ok. With regard to the bolts, these strains aren't really particular to my setup. In terms of the shelf, the vertical support is essentially a structural channel itself and I therefore expect it should handle a modest amount of torsional strain well.

The bolts holding the shelf to the channel and the threaded posts holding the casters to the channel will basically be in pure tension, so you don't have to worry about torsion on them at all.  When properly tightened the bolt will have thousands of pounds of tensile preload that will overshadow an reasonable load from the equipment on the rack.  With such a small clamping distance trying to 'bend' the joint between the rack and the channel or the rack and the caster will add to that tensile load rather than impart any bending load on the bolt itself due to the geometry of the mating parts.  The bending forces are going to be transmitted through the mating faces of the caster/channel/rack foot.  Speaking of which, if you use the existing holes in the feet of the rack (where the posts of the casters are now) there will be some torsion in that flat tab when the casters are to either side of the rack frame.  Hard to say if that would be an issue or not just from the photos.  Drilling a new hole in the foot directly inside of the post and putting in a bolt there would better transfer the moment load from the weight of the rack being offset from the caster into the post.  Depending on how thick that foot is you might want a slightly oversized heavy washer on top of it to hold it flat and tight to the channel, and reduce the chances of that foot bending.  This is all very handwavy without knowing the actual dimensions of everything.

[wizard69]

You are talking a quarter ton here so yo really don't want to fool around.   I'm not sure what the framing size is here but it looks to be a bit narrow, this leads me to fear the toolbox problem.   What is that you might ask,it is when one drawer is loaded so heavily that the whole box rolls over when that drawer is pulled out.   On top of all of that those sorts of shelving framing systems are not designed to be moved around, it is surprisingly ease to get a leg to collapse just from sliding an unloaded unit.   There is a fantastic Youtube video that shows an entire row of much heaving racking systems collapsing when a fork lift barely makes contact.  Hopefully this link works: and this is just one example.

So no don't put wheels directly on these types of storage frames.   If you want moveable storage (a good idea really) build a rigid and robust frame upon which the racking legs can be mounted.   Make sure the center line of the wheels is well outside the perimeter formed by the racking system legs.   That has to be in any orientation of the wheels.   Your stability will quickly go to hell if the legs are applying weight when the wheels close together.   The taller the rack the bigger the problem and the risk of the simplest thing resulting in a significant tipping moment.

Inertia really sucks, all you need is a defect in the floor or something as simple as a wire tie laying in your way, that hangs up a wheel.     The more top heavy you are the higher the likely hood that you could roll over the entire rack.   Keeping your center of gravity low and the wheels wide will help to reduce concerns here.   I bring this up due to seeing what happens when lesser carts have "lost it', in this case you have a lot more in the case of weight and height potential.

So be safe and engineer a safer way to move the rack around.

[EPAIII]

E-Bay casters with NO BRAND NAME??? And just how much weight and $$$ value are you going to support with them? When it collapses, who are you going to sue? And after your stay in the hospital from the injuries when it fell on you, who gets the medical bill?

You are in the US. You can get NAME BRAND casters from many places. And those places will still be there when both you and I are pushing up flowers. For Pete's sake, get some good casters instead of going cheap. And get some with BIGGER wheels so they don't stumble on even the smallest defect in the floor. Get some with PUBLISHED SPECS. - a real spec. sheet, not a bullet list.

I will have more on the overall design after I finish reading all the responses.

Something like this solution. Weld or get welded a sensible weight plate (3/16" - 5mm) directly under your verticals so you are not putting a torque on the plate and your wheels need to be moved closer to the centre.

Better option is have a look at these sorts of Casters 250kg rated each and wind down feet directly under the centre when you get to where you need to be so the casters are off the ground for much more stability and you also have the ability to level your stand.

Only arrived yesterday but for my new welding table. Plenty on offer on Amazon or evilbay eBay auction: #373720715093 EDIT there is also a slightly smaller (3" high) 150kg set but locally these were the same price.

[EPAIII]

First, although that YouTube warehouse collapse was scary, I don't think you need to worry about that level of failure. They obviously overloaded the shelving, probably by a lot and just a small amount of damage to a, just ONE vertical set it all off.

The first thing that jumps out to me is your choice of putting the mounting point for the swivel axis of the casters INBOARD of the legs. That is a big problem because, with the wheel swiveled by normal movement of the rack, the base is decreased on the leading edge of that motion. This makes it a LOT easier for it to topple over if even a small bump is encountered. The weight of all that equipment high above the floor provides a lot of momentum up there and when the base stops, the top will not even hesitate. And OVER SHE GOES. If you draw the lever arms you will see that as it starts to tip over, the front vertical leg will actually get LOWER and the weight itself will add to the instability.

You need to have the wheels OUTSIDE of the base of the rack or, at their worst possible position, under the outer edges of the base. This means that your horizontal members (angle or channel) needs to extend at least by the radius of the castor wheels PLUS half the width of the plate of the castors that mounts to that horizontal member. It should also be that far outside of the left and right sides, but the rack is probably wider than it is deep so you have a bit more fudge room there.

And those horizontal members should be either welded to the legs or UNDER the bottoms of those legs. If you insist on bolting them to the sides, use at least grade 5 bolts, not the common hardware store grade 3 ones. Grade 8 would be much better.

The casters: Swivel casters are a bad choice. I would use fixed casters and bolt them on at angles that allow the rack to swing in an arc if that is needed. Fixed casters will always contact the floor at the same point relative to the rack so they can be mounted on shorter horizontal members. This would be a LOT safer.

The rack itself: The photo does not show the entire rack, but I will bet that there is very little cross bracing on the sides and NONE on the front and rear. It probably relies on the strength of the rather small joints as seen at that rather wimpy cross piece in the photo and the small, 45 deg gusset plate that is partially visible at the top edge of the photo. Those aren't much. I would add some 45 degree X style cross braces from the front to the back legs. And also the same at the rear. That will add a great amount of strength and stability to the overall rack. If you use screws to mount the fronts of the equipment in the professional rack-style, put them both at the top and the bottom of those front panels. That will help brace the front. I do not recommend using shelves or trays for the equipment in this, inexpensive style of rack.

After saying that I do not suggest shelves, I will say that one shelf at the bottom and another one at the top will provide cross bracing in the horizontal plane. These could be sheet metal or plywood.

[beanflying]

E-Bay casters with NO BRAND NAME??? And just how much weight and $$$ value are you going to support with them? When it collapses, who are you going to sue? And after your stay in the hospital from the injuries when it fell on you, who gets the medical bill?

You are in the US. You can get NAME BRAND casters from many places. And those places will still be there when both you and I are pushing up flowers. For Pete's sake, get some good casters instead of going cheap. And get some with BIGGER wheels so they don't stumble on even the smallest defect in the floor. Get some with PUBLISHED SPECS. - a real spec. sheet, not a bullet list.

I will have more on the overall design after I finish reading all the responses.

Something like this solution. Weld or get welded a sensible weight plate (3/16" - 5mm) directly under your verticals so you are not putting a torque on the plate and your wheels need to be moved closer to the centre.

Better option is have a look at these sorts of Casters 250kg rated each and wind down feet directly under the centre when you get to where you need to be so the casters are off the ground for much more stability and you also have the ability to level your stand.

Only arrived yesterday but for my new welding table. Plenty on offer on Amazon or evilbay eBay auction: #373720715093 EDIT there is also a slightly smaller (3" high) 150kg set but locally these were the same price.

Instead of being a chest beating twat McMaster Carr sell the identical units which is where I got the Fusion model from. As I am on the other side of the planet McMaster is NOT a SOURCE. SO CAPITAL LETTER OFF !

Additionally there you go another SOURCE WITH SPECIFICATIONS https://www.californiacaster.com/footmaster-leveling-casters/

[matthuszagh]

I created a CAD model of the shelf on casters. Hopefully this makes it easier for people to assess whether this will be stable or not.

The CAD models of the casters and C channel are correct. The model of the shelf is an approximation just to give a rough idea. There's a good picture of the shelf on McMaster-Carr, where I bought it (https://www.mcmaster.com/4670T715/), as well as its specs. The casters are also from McMaster-Carr (https://www.mcmaster.com/24125T25/). I'm getting the C channels from a local metal shop. The specs are: HR Standard Channel C4 X 5.4 (H4 x W1.584 x FT 0.296 x WT 0.184).

[beanflying]

The only issue with rigidity is going to be side to side in particular if you try and move the unit. So either long cross braces or maybe some additional small angles as sketched on the shelves. The braces are a more ugly aesthetic but they are the much simpler and can just be made from steel strap. How concerned you need to be with this will depend on how the shelves are attached to the uprights and if they are just a clip in then be VERY concerned.

[matthuszagh]

The only issue with rigidity is going to be side to side in particular if you try and move the unit. So either long cross braces or maybe some additional small angles as sketched on the shelves. The braces are a more ugly aesthetic but they are the much simpler and can just be made from steel strap. How concerned you need to be with this will depend on how the shelves are attached to the uprights and if they are just a clip in then be VERY concerned.

You identified one of the inaccuracies of my model. The actual shelves and attachments look a bit different. The McMaster-Carr link shows this pretty well. I've attached an image here too. For reference, the vertical part there is almost 6" tall. It feels sturdy to me, but I'm no expert.

I can maybe add long cross braces if really necessary, though I'd like to avoid it if possible.

I've added a photo showing the edge of the shelf support.

To address your point, I guess these would qualify as just a clip in. But, I don't see any way they could come out. And, the shelf is rated to 6,400 lbs. Any reason this would be an issue?

[vk6zgo]

I know some people are irrationally against wood, but I'd consider just bolting/screwing the shelving down to a couple of wooden beams, and then attach the casters to the beams. Mostly because the appropriate piece of wood would be cheap and easily available at my corner hardware store and easily "machined" (i.e. sawed and drilled) compared to steel.

That is exactly what I thought of when I read the OP.

[Berni]

These are quite good shelves that they grab on 3 points, gives it some extra side to side stability because it holds more of a right angle.

I tend to find most shelving units a bit wobbly so i like to screw one top corner into a wall or wedge the whole thing into a corner of the room so provide extra stability. But if you are moving it on wheels you will regularly be applying sideways forces to it, the right way is to grab it at the bottom when moving it, but it is much more convenient to grab it at arm height. Your particular shelves look nice and robust (much better than any of mine) but id still be worried about side to side forces when its heavily loaded.

Id say the easiest way to take out any sideways wobble is the suggestion from beanflying . The red X shape on the back does not need to be actual beams. You can just run a steel wire in that shape and pull it tight. This constrains the length of diagonals of the structure and will make it very rigid.

As for holding the wheels id use a hefty chunk of wood unless you are handy with a welder.