Subscribe or Renew Membership Subscribe Renew

Layout of curved concrete wall?

rondeandesign's picture

We have a job in our architecture office where the designer has placed a 80 foot long curved concrete wall leading up to the entrance. How should I dimension this for the contractor to layout? Currently the draftsman noted the radius of the arc (88'-6") and the length of the wall. Should I provide orthogonal dimensions instead?

In general how do contractors layout this type of geometry in the field?

(post #103923, reply #1 of 28)

Locate the radius point and give somebody a 100' tape measure and a can of spray paint..I just watched it all being done from a roof at one one of the largest horse farms in KY.

Parolee # 40835  

(post #103923, reply #2 of 28)

The site won't allow such a long radius to be pulled, there are trees and buildings nearby.

(post #103923, reply #3 of 28)

My cell phone has GPS, that'll get him close.

Parolee # 40835  

(post #103923, reply #4 of 28)

Give the layout to a surveyor and have him stake out the curve.  He should be able to take the CAD drawing and punch it into his total station.

(post #103923, reply #6 of 28)

I would want a location for a chord drawn with points on the curve relative to that at five foot intervals.



Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!



Oh Well,

We did the best we could...

(post #103923, reply #5 of 28)

What if the radius is five hundred feet? That hundred foot tape won't help much.



Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!



Oh Well,

We did the best we could...

(post #103923, reply #7 of 28)

It wasn't. He said 88'6'' IIRC.

Are ya mad at me Honey?

Parolee # 40835  

(post #103923, reply #8 of 28)


You got me confused with Gunner?


how did I miss that radius? Haven't had a drink in three days.



Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!



Oh Well,

We did the best we could...

(post #103923, reply #9 of 28)

Oh C' know you love a 4 day beard..LOL

Yup, GPS is the rule, I set my phone on my corners, wrote down the long/lat and moved witha 100 tape to the next corner...took a reading, and wrote that, did the math,,I was with in 2''.

Thunder thumpers baad, I gott a go...later my friend...I am about to lose m

Parolee # 40835  

(post #103923, reply #22 of 28)

And when the house is in the way?

If you didn't have time to do it right the first time, how come you've got time to do it over again?

(post #103923, reply #23 of 28)

He didn't indicate that in post#1.  So I ran with what I saw.

Parolee # 40835  

(post #103923, reply #24 of 28)

I'm sorry I kinda jumped on ya there, but you did sort of state the obvious :). When the house IS in the way, the math can get confusing. My method involves laying out the cord and continually bisecting and calculateing the arch rise for each new cord. Its time consuming and one of the other post's solution sounds much better, but , honestly, I don't quite get it yet.

Sharpening my pencil.

If you didn't have time to do it right the first time, how come you've got time to do it over again?

(post #103923, reply #26 of 28)

Not to be overbearing, but all I need is the delta (central angle or to a surveyor the "I") and you can two tape any horizontal curve, reverse, spiral, or compound.

Guess you could figure the delta by finding the arc of a 100' R, (.01745333 R 360)then figuring the delta of 1' of that arc, then multiplying that by the arc (88.5') which would give you the delta.  Then for the chord, it's 2R sin of 1/2 delta.  Then break the arc into 4 segments, and calc an O/S to the projection of the approaching tangent by breaking them into 4 separate triangles.  Pull the chord, pull the o/s set the point.  Swing a 90, set the o/s.

That will be $300 please.... ;-)



(post #103923, reply #27 of 28)

No problem.

But for $300, I'd like a sketch, glossary, formulas, alternative methods and a few examples of each.

The night school course in geometry is only $145, but I'm willing to go a little more, to avoid eating into my T.V. time.  :)


Still sharpening my pencil

Edited 5/9/2007 4:19 pm ET by karp

If you didn't have time to do it right the first time, how come you've got time to do it over again?

(post #103923, reply #28 of 28)

Geez, you want it with ice-cream too?

Anyway, the data is:

C=85.64, Delta=50 deg 42' 24", T=47.38, R=100.00', Arc=88.50.

One "two taping" method.  You'll have to convert decimals to feet.

Project a straight line past the proposed curved area. Set points at these distances from the PC, 21.94, 42.82, 61.60, 77.39.  Pull a chord of 22.08 from the PC, then 2.44 over from the first point set, and set the first point on the face of the wall.

Pull the second (from the PC, as all will be), of 43.89 and 9.63 over, then 65.16, and 21.23 over, and finally at the PT, 85.64 and 36.67 over.

Surveying is simple because it all comes down to triangles.  If you know how to solve'em, there's no problem you can't solve in the field.  It'd have to be simple for me to understand it for all these years.

Is the check in the mail?  If so, thank you for using Microsoft products....



(post #103923, reply #25 of 28)

You and your phone may have a new sideline...I just finished talking to a surveyor who wanted an extra 300 bones for a GPS siting...just finding true south, no curves!

Outside of the gates the trucks were unloadin',
The weather was hot, a-nearly 90 degrees.
The man standin' next to me, his head was exploding,
Well, I was prayin' the pieces wouldn't fall on me.

The Village Woodworks, Inc

Chapel Hill, NC


We'll have a kid Or maybe we'll rent one He's got to be straight We don't want a bent one He'll drink his baby brew From a big brass cup Someday he may be president If things loosen up

(post #103923, reply #10 of 28)

if you can get the draftsman to lay it out in CAD from square reference lines then the layout man in the field can then plot those spots, i have done similar with bridges and overpasses, the radius is so great it would be impractical to scribe an arc.

in CAD it can be set up as a grid, and the respective points of the radius marked in square measurements. i would request (if it was me doing the layout) several points of reference so i could lay each point of the radius out from at least two different grid offset points.

if you lay it out in the field from two or more different starting points, and you arrive at the same finish point you can be pretty sure you are correct. the double layout is just a double check in the field. you may or may not need to do them all if it looks like it is right, but i would want to have the grids and offsets in CAD before i went to the field just in case. better to have more than you need than not enough.

(post #103923, reply #11 of 28)

I think your technique sounds best. I also spoke with a contractor working on another project from our office today and he suggested the same thim. Basically I'm going to turn the arc into a series of chords with orthogonal dimensions to the endpoints of each chord.

Thanks for everyone's advice.

(post #103923, reply #12 of 28)

read what segundo and I said again.

A series of chords will still be an arch that is hard to lay out in the field. Each chaord and verticx will be relative to the other ones so every error will compound.

What you need is ONE chord the full eighty feet. Divide that chord into four or five foot long sections and draw a line perpendicular to the chord intersecting each of those points.
Then define the length of each of those intersecting lines from chord to arc.

That way all points defining the arc spring from one common line that is separate and independent of the location of the actual structure and easy to pull from



Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!



Oh Well,

We did the best we could...

(post #103923, reply #13 of 28)

Thanks for clarifying. I understand and will do as you described; I just couldn't explain it as well as you did.

(post #103923, reply #14 of 28)

i couldn't explain as well as you did either! you da man!

(post #103923, reply #15 of 28)

group effort guys.
I thought I had explained it fine the first time but apparently not - then you did a whole lot better with your description which I admired,but then his responce showed that something was still lacking so I filled in the blanks is all.



Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!



Oh Well,

We did the best we could...

(post #103923, reply #16 of 28)

How about doing the math to figure the diameter needed for a smaller disc to hang from a 40' crane that will cast a shadow the diameter needed. Then just trace it on the ground with a really pointy stick.

It's easy to be smart when your drink Oklahomas poultry river water.


Edited 5/4/2007 9:20 pm by jagwah



(post #103923, reply #17 of 28)

These are all good suggestions, however there is one other method that will work if you can't set the radius.  As a surveyor, I've provided this method to a lot of concrete guys that didn't have them staked, but only given the prints.

Have your draftsman give you the offsets from each chord that you calculate from a straight line starting at the Point of Curvature (PC), and ending at the Point of Tangency (PT).  The offsets may be calculated at any point along the arc, or at needed grade points.  The points are set by going calculated distances on the straight line, then pulling over the o/s distance, then pulling the chord from the previous point, or the PC, when you start from there.

After the points have been set on the wall face, the contractor will need to set o/s stakes for line and grade that won't be destroyed.  These may be calculated from the same line.

Hopefully the Arc, Chord, Radius, and Delta (central angle) have been provided. 

Hope this helps.



(post #103923, reply #18 of 28)

Not intending to hijack this discussion but...

How would you form up this wall? Would you bother to curve the footings or do them in small sections? What about bottom plates and whalers?

I've got a long wall with varying diameter curves that I have been intending to do in stone, but think would be better in concrete.

(post #103923, reply #19 of 28)

Depending on the height.  We usually staked the footings which were usually 6" wider on either side, then the steel was placed and made the face easy to curve.  Once the footer was in we didn't usually have to come back.

I'm no concrete guy, but I've seen curved walls using bent ply with a 2X bracing for both sides and ties though the wall.



(post #103923, reply #20 of 28)

that is an excellent question! i would use the jigsaw to create top and bottom plates for the panels (or forms) for sure, then i would use thin plywood to cover the curved forms in layers.

the whalers are definitely the hard part, and of course it all depends the size and thickness of the wall, but horizontal whalers would have to be cut by jigsaw out of ply material and would be very time consuming.

i would opt for stout vertical members between the radiused plywood plates, 4x4 material and then taper ties or snap ties between the vertical members. basically no horizontal whalers and extra stout vertical studs/whalers.

all you are trying to do is keep the concrete from leaking out (plug the holes) and keep the forms from collapsing. bracing, cableing, kicking, all would be employed depending on circumstance, and it would be a good idea to have a coupla carpenters on pour watch ready with material and tools in addition to pour crew. make sure the vibrator guy has a lot of experience and keeps a close watch, he can make or break you. fast in and slow out with the vibrator, and if it looks like shes gettin ready to blow turn it off quick.

(post #103923, reply #21 of 28)

Kinda like this?  Footings curved also.  Front and rear walls needed to have the same curves for the bar joists to work.


PAHS Designer/Builder- Bury it!

PAHS works.  Bury it.