×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Finding vertical CG of a large box structure

Finding vertical CG of a large box structure

Finding vertical CG of a large box structure

(OP)
We have mfg a large box structure that we need to find the vertical center of gravity. We just shipped it out and have access to it on the other side of town. What I am looking for would be a way to to estimate the vertical center of gravity location on it. The base of it is structural so we can lift it or weigh it with scales no problem, but the walls and roof are not structural enough to tip it or anything like that.

Any help greatly appreciated.

RE: Finding vertical CG of a large box structure

Pick it up, weigh all four corners.

Pick it up, place a spacer between thing and scales (creating an angle (a) of the base) at two adjacent corners and weigh again.

Calculate the amount the horizontal CG moved.

Vertical height of CG = Tan (a) / horizontal CG moved distance

RE: Finding vertical CG of a large box structure

(OP)
Thanks

RE: Finding vertical CG of a large box structure

crate it up on a close fitting crate, so the box doesn't have to support itself.

if you can,
1) weight the box on it's own (standing up), one number is sufficient;
2) weight the crate on it's own (laying down, horizontally), at least two numbers (from both ends) so you can find the CG of the crate (between the two ends);
3) pack up the box in the crate (so the box is supported);
4) lay the crate+box on it's side and weigh each corner, it's probably good enough to weigh the corners one at a time, so long as you don't move the crate;
5) now you can find the CG of the crate+box, and then the CG of the box (by removing the crate).

Quando Omni Flunkus Moritati

RE: Finding vertical CG of a large box structure

Mack:
Is this box on a trailer or a railcar, and the hauler wants to know the vert. CG? Calc. the CG and then ballast the load at deck height to bring the CG down below their limit.

MintJ’s method should work, it is technically correct. But, for small angles (a) of 5, 10 maybe even 15̊, the tangent and the sine of the angle (a) are so close, in value, that you may have trouble weighing accurately enough to do a meaningful calc., by his method.

Since we are not really playing 20 questions here, why don’t you tell us what the box is; its dimensions, weight, some of its construction details, etc. etc. so we might imagine how it could be lifted and tipped, etc.

RE: Finding vertical CG of a large box structure

(OP)
Okay then ..... let me try to explain. I work at CARCO industries as the engineering manager, and we just built an Arctic Lube Body.





Its getting shipped to northern canada and it just left our facility to travel across town for integration on one of the Kiewit chassis's, we get a call from the customer that we need to reveal the vertical CG so that paperwork can be submitted when it crosses the border. As long as we do our due diligence of trying to determine the CG (and add a couple inches to the found value for a SF) will allow it to cross the border.

If it had all been drawn in solidworks I would just go with what solidworks says, but we have a designer on staff how does a lot of this stuff in key creator (that's a whole other story).

RE: Finding vertical CG of a large box structure

So it's big and not particularly delicate and the CG should be a fare distance above the floor.

My method should be workable.

RE: Finding vertical CG of a large box structure

By comparison to some of the stuff that goes into northern Canada these days, this is pretty small and I cant see Canada customs having a qualified P.Eng checking the accuracy of C of G calculations or anything similiar. Give them a number that looks about right and all should be OK

RE: Finding vertical CG of a large box structure

I believe you can do it without weighing it.

Just rotate the box about one edge until the CG is just above the pivot point, Record the rotation angle a. Now do the same at the opposite edge and record the angle b. If the distance between the rotating edges is W then

h=W/[(tan(a)+tan(b)]

RE: Finding vertical CG of a large box structure

(OP)
BTW it weighs 17000 lbs and has pieces of it that go down 30", so its not something we can just tilt over on an edge.

RE: Finding vertical CG of a large box structure

From Mint..
"Vertical height of CG = Tan (a) / horizontal CG moved distance"

Nitpicking,I get

Vertical height of CG = sin(a) / horizontal CG moved distance

For small angles they are close

RE: Finding vertical CG of a large box structure

Zeke is correct.

I shouldn't do trig in my head without drawing a picture.

RE: Finding vertical CG of a large box structure

(OP)
Thanks guys. We are going out next week with wheel scales and will get it done.

RE: Finding vertical CG of a large box structure

Estimate your CG from your blue print for Pete sake. The Canadian border people will just want a number for their paperwork

RE: Finding vertical CG of a large box structure

(OP)
chicopee, I am sure we could do that ..... but the practice is good and establish me as a good engineering manager for this company. I am in my 3rd week and want to make a difference.

RE: Finding vertical CG of a large box structure

My bad,
I redid the height calculation for Mint's solution and now find that the
vertical height=y1=[x1cos(a)-x2]/sin(a), not [x1-x2]/sin(a) as I previously posted (inverted)
x1= original position x coordinate of CG referenced from rotating edge
y1= original position= vertical height CG
x2 = x coordinate of CG rotated
x1-x2= moved x distance of CG

This is, in general, a significant change since x1cos(a) is less than x1 and the numerator may be small, so any error in either numerator term could be important.

FYI, the formulation easily derives from rotating the CG vector,viz
cos(a) -sin(a) x1
sin(a) cos(a) times the vector y1

y1=vertical height of CG

RE: Finding vertical CG of a large box structure

Making a mountain out of a mole hill will not establish you as a good engineering manager for this company as it make more work for very little gain. A good engineering manager makes decisions based on the availability of material at hand without adding extra unecessary expenses. Proposal from the above responders will create extra expenses for little gain and when such resources can be used for other problem areas.

RE: Finding vertical CG of a large box structure

(OP)
chicopee,

that is exactly what I am saying.

-have designer use key creator to determine the CG of each of the installed components; 2 days which is $1600

-get the CG for the mfg structure from mexico; I can't even guess how many days this would take

-the alternative; take two guys with four wheel scales, blocks of wood, digial camera, and tape measure; 1 day lead time and cost is about $500; its effective and its gets it done cost effectively

Which one would you choose?

RE: Finding vertical CG of a large box structure

(OP)
chicopee,

that is exactly what I am saying.

-have designer use key creator to determine the CG of each of the installed components; 2 days which is $1600

-get the CG for the mfg structure from mexico; I can't even guess how many days this would take

-the alternative; take two guys with four wheel scales, blocks of wood, digial camera, and tape measure; 1 day lead time and cost is about $500; its effective and its gets it done quickly

Which one would you choose?

RE: Finding vertical CG of a large box structure

"-the alternative; take two guys with four wheel scales, blocks of wood, digial camera, and tape measure; 1 day lead time and cost is about $500; its effective and its gets it done cost effectively"

And how sure are you that the equation that you need to perform this task is accurate? After all, you got 3 answers for that and why would you use any one of them without proof? What I am saying,in general, it is very dangerous to rest a design (or calculation in this case) on the results of forum contributors who are giving you their honest opinion. But is that sufficient ?

RE: Finding vertical CG of a large box structure

(OP)
We have already ball parked it at 30 to 32" from the truck frame top. We don't have the actual weight, so going though this exercise has validity in just that. I was going to use all the calculations and see how they come out.

RE: Finding vertical CG of a large box structure

Looking at your web site, your ball park value looks about right. I guess your Artic lube body fits on a truck frames. Nice.

RE: Finding vertical CG of a large box structure

I am going to disagree with any criticism of not getting the vertical CG correct - at least within 3- 6 inches of actual vertical position. Horizontal CG is vital also when lifting 17,000 (or 700 ls!) of an unknown load. The riggers will look at YOUR marking and use them. You can be off a little, but you can be wrong when somebody else's crane and people and trucks and forklifts are using YOUR painted mark for THEIR lifts.

Long truck trips through the wilderness need adequate CG info for loading and moving.

RE: Finding vertical CG of a large box structure

By the way, don't you have it in 3D CAD? With proper and accurate internal eqpt loadout of the internal GG's of the all of the stuff inside, you should be able to the internals, the frames, the walls, and the floor/roof added together.

RE: Finding vertical CG of a large box structure

(OP)
I don't think key creator has this capability. Its not parametric CAD application. What ever we come up with we are going to add a couple inches as a SF.

RE: Finding vertical CG of a large box structure

could it be higher than the geometrical center ?

how much validation do you need ?

Quando Omni Flunkus Moritati

RE: Finding vertical CG of a large box structure

(OP)
No it won't be higher than geometric center. The inside is full of lube/oil tanks that only come up about 2/3's of the height of the structure. I don't think we need much validation, the owner of the company has experience with this stuff and he is sure that the number we come up with will pass any requirements.

RE: Finding vertical CG of a large box structure

so if you think the geometric center is reasonably conservative ... why fuss it any more ?

Quando Omni Flunkus Moritati

RE: Finding vertical CG of a large box structure

(OP)
Its a good exercise, and documenting it gives the ability to determine this again on other projects.

RE: Finding vertical CG of a large box structure

Do you have a crane with spring scale? Lift up one side (SLOWLY!) with the crane until the weight reading is below some threshold, 1,000lbs, 2,000lbs? Whatever you feel is safe, crane still supporting enough weight to ensure you don't topple the thing over. Treat that as the tipping point, measure the tilt angle, and use trig to get the vertical COG. It will be as conservative as your spring scale threshold.

If you wanted to further reduce the chance of tipping, and increase the accuracy, you could plan to use the spring scale weight as part of the calculation. Modified method of the "tilt and weigh 4 corners" mentioned earlier. If you've got X,000lbs on the crane, X,000 on the ground, know the horizontal COG, and can measure the tilt angle, from there it is physics 101.

I'd like to think that this is a pretty elegant solution. Depending on the details it works out as the best option for accuracy, man hours (only thing quicker is a WAG - wild ass guess), engineering justification, and safety. You could eliminate ALL assumptions and peg the vertical COG dead on if you wanted, but as already mentioned, probably not worth the trouble.

I'd pull one side up with a spring scale until I had an appreciable angle and a clear change in weight distribution, then plug it in.

Quote (so if you think the geometric center is reasonably conservative ... why fuss it any more ? )


Because he is the engineering manager, not the shipping manager. And if this works out well he can write a procedure, this could be a chargeable item for a customer - dry and wet COG's verified.

RE: Finding vertical CG of a large box structure

procedure could easily be "use geometric center, if appropraite" and bill the customer the same ...

Quando Omni Flunkus Moritati

RE: Finding vertical CG of a large box structure

Will keep that in mind next time an interesting problem comes up: "Why bother, put a conservative number down and move on." I would be interested in hearing the engineering justification for dry and wet COG being the same number though. Have a cup of coffee and maybe reconsider?

Neither of us knows the value this information might have to the end user. It goes on a truck, is vertical COG limiting any operating parameters? Does the end user need to confirm max incline it can climb, max safe speed, safe braking force, govt regulations if it needs to travel on a public road, blah blah blah, all the boring "no, because we can do what we've always done" questions that a silly engineer might ask. Sorry for being pedantic. dazed

RE: Finding vertical CG of a large box structure

Mack:
The two center sills in the floor framing of your lube body which will mate with the truck chassis sills, should be fairly husky structural members. Maybe you can weigh it and lift it from these. Weighing it on/at the four end points of these sills should give you a pretty good CG in plan. Accuracy in scale positioning and these measurements and in the four actual weight readings is important for good final calcs. and results. If you were to put the lube body up on two saw horses and on these two center sills, this should allow clearance for the parts that hang lower, and still allow you to roll the lube body. Then lift on one sill, and roll the body until all 17k is on the two scales on the other sill; then draw a plumb line (hang a plumb line) from that pivot point, up the ends of the body; do this same thing on the other sill, in the other direction, draw the plumb lines each end; where the two plumb lines cross on each end should be the height CG. Of course, this requires that you support the lube body so you can get to these balance points on one sill, without the body actually rolling over, out of control. This larger rotation will give you the most accurate results without any trig. calcs.

In my first post the point I was trying to make is that a larger angle of rotation will yield better calculated results because your weight readings and angle measurements are not so small and critical. I still think that with a small angle of rotation you may not be able to weigh accurately enough at the four corners to make an accurate height calc. A larger rotation would likely yield better results. Both plan location and height location of the CG will change btwn. dry and full wet weight conditions. Maybe you want to bracket these extremes.

I’m with 1gibson on this, this is not a trivial problem, or bit of info. But, only you know what degree of accuracy you need to achieve. maybe do this whole process several times, in opposite directions, or on different lube bodies to see that you get some consistency in your results. Something you can really hang your hat on. And, they type of equipment you have at that shop will determine how you actually do this.

RE: Finding vertical CG of a large box structure

Pendulum still sounds simpler. Suspend the box. Measure time for one swing cycle. Calculate distance from point of swing to cg. The rest is measuring distance from point of swing to bottom of box support structure and subtract calculated pendulum length.

Ted

RE: Finding vertical CG of a large box structure

"Pendulum still sounds simpler. Suspend the box. Measure time for one swing cycle. Calculate distance from point of swing to cg. The rest is measuring distance from point of swing to bottom of box support structure and subtract calculated pendulum length".

Not a simple pendulum. You need I.

RE: Finding vertical CG of a large box structure

Do not need I. Assume mass is concentrated at the center of gravity.

Ted

RE: Finding vertical CG of a large box structure

I like the pendulum idea, and it does sound simpler, but maybe not for something this size. By the time you rig something up that is solid enough to let swing around safely, you've got a good amount of labor and material cost sunk in there. Then you still have to convince the shop that swinging a 17,000 lb box around is a good idea.

RE: Finding vertical CG of a large box structure

"Do not need I. Assume mass is concentrated at the center of gravity."

With assumptions like that, I would take Rb's center of box as better.

RE: Finding vertical CG of a large box structure

we could put the box into a fixture and apply horizontal load and adjust the position of the load until the box translates (ie doesn't rotate) ...

Quando Omni Flunkus Moritati

RE: Finding vertical CG of a large box structure

This might be too late, but it sounds like you want to perform an inclining experiment, like we do when we build ships. We have to experimentally determine the VCG for stability purposes. For something like this, there is a precedure for inclining the object in the air. You will need to know what the object weighs, and good methods for that have been discussed above.

The process works by moving a known weight over a known distance, and measuring the resulting inclination. you have to pick the item up and sling it with a beam make sure that it can only rotate about the longitudinal direction. Then you have to make an apparatus mounted to the top (or bottom) where you can move the known weight a known distance. you might need a small leveling weight to ensure you start from a level position.

You measure the resulting angle with a pendulum or a digital inclinometer.

Then you use an equation that says GM= (known weight x known distance)/(tangent of the resultant angle x the weight of the entire object) . For an incline in air, the GM is the distance from the hinge point to the VCG.

See ASTM F1321 for the process to perform the inclining in water. Same principles apply for in air.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources