What's new
What's new

Curved stair railing fabrication

Bigrhamr

Aluminum
Joined
Feb 24, 2004
Location
LaCenter, WA USA
I don't know if anybody here does this kind of work but with all the technical minds I thought I'd give it a try. Also posted on one of the blacksmithing boards.

I've got an idea brewing for a jig for curved staircase rails that I want to bounce off you guys. The idea is to be able to go out and measure, set up the jig and take it back to the shop where it would work as an actual model of the stairs. Hopefully avoiding multiple trips out to check fit parts and onsite bending and tweaking.

The idea goes something like this: On a stair tread lays a piece of square tube. In the end of the tube that goes out to the front edge of the tread goes a piece of square solid, it's a slip fit so it telescopes inside the tube and there's a set screw to clamp it in place. Out on the end of the square solid there's a round hole going though it vertically. On the other end of the tube that goes up against the next tread there's a piece of round solid welded to it and going straight up. The round solid is a slip fit for the size hole bored in the square solid on the other end which also has a set screw. Clear as mud so far? Build as many of these units as you have stair treads. Lay one on the bottom step, take the next one and slide it's round hole over the round bar on the bottom one, rotate it to match the curve and clamp the set screw down. That gives you the height of each tread and the curvature. Then you use the telescoping square part to set the width or length of each tread. Repeat until you hit the top. Now take the whole unit back to the shop and clamp it to some type of upright posts and you have a model of the staircase standing in your shop where you can fit a rail to it.

Any comments, sugestions or reasons why that would or wouldn't work? Anybody even know what I'm babbling about?
 
I don't know if anybody here does this kind of work but with all the technical minds I thought I'd give it a try. Also posted on one of the blacksmithing boards.

I've got an idea brewing for a jig for curved staircase rails that I want to bounce off you guys. The idea is to be able to go out and measure, set up the jig and take it back to the shop where it would work as an actual model of the stairs. Hopefully avoiding multiple trips out to check fit parts and onsite bending and tweaking.

The idea goes something like this: On a stair tread lays a piece of square tube. In the end of the tube that goes out to the front edge of the tread goes a piece of square solid, it's a slip fit so it telescopes inside the tube and there's a set screw to clamp it in place. Out on the end of the square solid there's a round hole going though it vertically. On the other end of the tube that goes up against the next tread there's a piece of round solid welded to it and going straight up. The round solid is a slip fit for the size hole bored in the square solid on the other end which also has a set screw. Clear as mud so far? Build as many of these units as you have stair treads. Lay one on the bottom step, take the next one and slide it's round hole over the round bar on the bottom one, rotate it to match the curve and clamp the set screw down. That gives you the height of each tread and the curvature. Then you use the telescoping square part to set the width or length of each tread. Repeat until you hit the top. Now take the whole unit back to the shop and clamp it to some type of upright posts and you have a model of the staircase standing in your shop where you can fit a rail to it.

Any comments, sugestions or reasons why that would or wouldn't work? Anybody even know what I'm babbling about?
 
on the few stairs that ive helped build the guy that i helped would divide the total amount of "total rise" by 6 1/2" or 7" to see the total number of steps needed. then he would adjust the "rise" of the individual steps.
all steps need to be the same height so you dont have a "tall" last step.
the fixture you describe will probably work fine just make sure the top stair attaches at the right place hehe...jim
 
on the few stairs that ive helped build the guy that i helped would divide the total amount of "total rise" by 6 1/2" or 7" to see the total number of steps needed. then he would adjust the "rise" of the individual steps.
all steps need to be the same height so you dont have a "tall" last step.
the fixture you describe will probably work fine just make sure the top stair attaches at the right place hehe...jim
 
What if it moves on you in transporting it back to shop? Sounds unstable, hard to handle. Maybe tack it together instead of set screw together. Might save some heartburn/egg on face.

John
 
What if it moves on you in transporting it back to shop? Sounds unstable, hard to handle. Maybe tack it together instead of set screw together. Might save some heartburn/egg on face.

John
 
Want to reverse engineer? Sounds like you need a Faro arm.. then you can just take the 3-D data back to your office on your computer.

97533168.jpg


TMD
 
I agree, it sounds like it could easily get jarred, and you would never know it.Plus, your jig cant account for things being out of square, and every house in the world is. You are assuming that all the angles are 90 degrees.

How many curved railings do you do a year?
I have done a couple, and found that if it is not a standard radius of a circle, a paper template works ok for the x-y. the z is where things get a little hairy- because carpenters idea of precision is often just "within a quarter inch".
I use big sheets of cardboard- sometimes up to 4x8 feet, and trace the actual step profiles on them.
But I have also found that trips back and forth for fitting are pretty inevitable.

but hey- go ahead and build it, and tell us how it works- it isnt much money worth of materials, nor should it take very long to build.
 
I agree, it sounds like it could easily get jarred, and you would never know it.Plus, your jig cant account for things being out of square, and every house in the world is. You are assuming that all the angles are 90 degrees.

How many curved railings do you do a year?
I have done a couple, and found that if it is not a standard radius of a circle, a paper template works ok for the x-y. the z is where things get a little hairy- because carpenters idea of precision is often just "within a quarter inch".
I use big sheets of cardboard- sometimes up to 4x8 feet, and trace the actual step profiles on them.
But I have also found that trips back and forth for fitting are pretty inevitable.

but hey- go ahead and build it, and tell us how it works- it isnt much money worth of materials, nor should it take very long to build.
 
Only done a few sets of steel stairs,all pretty straightfwd...........however have put FULLTILTBOOGIE zillion $ wooden stairs in many a fine home.To the point that for 3 years went from one house to another doing nuthin but for a pretty big homebuilder co.

The curved stuff gets a trifle complicated.And am afraid that I'd only confuse you because of degree of difficulty in explaining.Its one of those things that if someone showed you a few of the tricks you'd pick right up on it.I will give you a few must do's and some general stair rambling.

First and foremost you need a finish floor to finish floor(FF)measure.Sounds easy but the problem is the cust. thinks that they can change floor cvrings after we've done the calculations.So it imperative that they sign off your FF-FF measures at the time of contract.I've found that being a touch "pricky" while discussing floor highths usually gets the point across......you know,threaten bodily harm if they dare change design specs on ya.Next is overhead obstructions and early on the site plotting where they are.Again sounds easy but the code Dic.....uhhhh inspection guys aren't playin by the same rules.Some eroneously calculate the obstruction perpendicular to stringers.Which when you think about it don't work.The old way was to visualize a rock dropped from overhead obstruction and it better travel "X" amt. of distance before striking tread.Just be sure and make a call to the local dic.....uh inspection office and see what their take on this confusing issue is.Take all your measurements and calculations......go take a break......then come back and measure it again.Also be dang sure and get EVERY possible measurement,no matter how seemingly insignificant.You'll be back at the shop trying to mock up and all it'll take is one little minor measure to stop progress.This was where my sons always come in handy.Having them along on measuring always made it more complete,kinda hard to describe.....Its been awhile since bein in stair world have no doubt left some impotent stuff out.
 
Only done a few sets of steel stairs,all pretty straightfwd...........however have put FULLTILTBOOGIE zillion $ wooden stairs in many a fine home.To the point that for 3 years went from one house to another doing nuthin but for a pretty big homebuilder co.

The curved stuff gets a trifle complicated.And am afraid that I'd only confuse you because of degree of difficulty in explaining.Its one of those things that if someone showed you a few of the tricks you'd pick right up on it.I will give you a few must do's and some general stair rambling.

First and foremost you need a finish floor to finish floor(FF)measure.Sounds easy but the problem is the cust. thinks that they can change floor cvrings after we've done the calculations.So it imperative that they sign off your FF-FF measures at the time of contract.I've found that being a touch "pricky" while discussing floor highths usually gets the point across......you know,threaten bodily harm if they dare change design specs on ya.Next is overhead obstructions and early on the site plotting where they are.Again sounds easy but the code Dic.....uhhhh inspection guys aren't playin by the same rules.Some eroneously calculate the obstruction perpendicular to stringers.Which when you think about it don't work.The old way was to visualize a rock dropped from overhead obstruction and it better travel "X" amt. of distance before striking tread.Just be sure and make a call to the local dic.....uh inspection office and see what their take on this confusing issue is.Take all your measurements and calculations......go take a break......then come back and measure it again.Also be dang sure and get EVERY possible measurement,no matter how seemingly insignificant.You'll be back at the shop trying to mock up and all it'll take is one little minor measure to stop progress.This was where my sons always come in handy.Having them along on measuring always made it more complete,kinda hard to describe.....Its been awhile since bein in stair world have no doubt left some impotent stuff out.
 
I've built a few spiral stairs myself. I've found the simplest way to template for the rail was to build a laminated pattern right on the inside and outside stair jack.

A spiral staircase follows a helicoid. A properly constructed hand rail follows the spiral exactly and duplicates its geometry in three dimensional space. Any significant error is readily detected by the uninformed eye making the rails appear clumsy and amateurish.

My procedure for making a spiral staircase is a bit overkill but you get a precision fitted template you can depend on. The process goes like this. I laid down a plywood strip thin enough to confrom to the jack, marked it, cut it out a bit oversize, and used that to mark and cut a number of 40" long arcs of 1/2 to 3/4 plywood - enough to make about 10 feet of template for each jack. Then I laid up these slightly oversized arcs and routed them to the stair jacks with a flush trim bit. Laying the arcs directly on the jack and butting them end to end in about 4 plies I glued and clamped them in intimate contact with the jack so the template takes on the true spiral of the jacks.

From this you get a rigid template that's easily transportable. From it you can make up and fit the spiral hand rail so that at final assembly and finishing everything fits correctly and looks great to the keenest eye.
 
I've built a few spiral stairs myself. I've found the simplest way to template for the rail was to build a laminated pattern right on the inside and outside stair jack.

A spiral staircase follows a helicoid. A properly constructed hand rail follows the spiral exactly and duplicates its geometry in three dimensional space. Any significant error is readily detected by the uninformed eye making the rails appear clumsy and amateurish.

My procedure for making a spiral staircase is a bit overkill but you get a precision fitted template you can depend on. The process goes like this. I laid down a plywood strip thin enough to confrom to the jack, marked it, cut it out a bit oversize, and used that to mark and cut a number of 40" long arcs of 1/2 to 3/4 plywood - enough to make about 10 feet of template for each jack. Then I laid up these slightly oversized arcs and routed them to the stair jacks with a flush trim bit. Laying the arcs directly on the jack and butting them end to end in about 4 plies I glued and clamped them in intimate contact with the jack so the template takes on the true spiral of the jacks.

From this you get a rigid template that's easily transportable. From it you can make up and fit the spiral hand rail so that at final assembly and finishing everything fits correctly and looks great to the keenest eye.
 
Yeah, I realize the setscrews could be the source of a major headache. I wanted to get away from welding onsite since its usually inside a finished house and sparks can be a bad thing. Maybe I'll try the screws then take it outside and tack everything before it gets hauled around.
I've done quite a few curved rails and several stair rails but so far when I find out the job is a curved stair with x,y & z to worry about I mumble something about being too busy for such a piddly little job. :rolleyes: BTW I'm talking about just a curve and not a spiral. On the curved ones I've done I used the cardboard template and/or a story board both of which worked just fine. The job at hand right now is a curved stair that follows the walls of a five sided room, none of the sides are the same length so the radius constantly changes. I set out to measure it and make cardboard templates at one point and after an hour or so of head scratching decided to retreat and regroup which is what I'm going to do now. Don't wanna have nightmares about this thing :D
 
Yeah, I realize the setscrews could be the source of a major headache. I wanted to get away from welding onsite since its usually inside a finished house and sparks can be a bad thing. Maybe I'll try the screws then take it outside and tack everything before it gets hauled around.
I've done quite a few curved rails and several stair rails but so far when I find out the job is a curved stair with x,y & z to worry about I mumble something about being too busy for such a piddly little job. :rolleyes: BTW I'm talking about just a curve and not a spiral. On the curved ones I've done I used the cardboard template and/or a story board both of which worked just fine. The job at hand right now is a curved stair that follows the walls of a five sided room, none of the sides are the same length so the radius constantly changes. I set out to measure it and make cardboard templates at one point and after an hour or so of head scratching decided to retreat and regroup which is what I'm going to do now. Don't wanna have nightmares about this thing :D
 
5 sided rooms with changing radius curves- this is why we love architects so.
Maybe if you make it in sections- then you only have a few joints to worry about. Cause I dont see how you are gonna get the whole jig out the door when it is all in one piece.
As far as welding inside goes- Tig is the only way to go. We do all our site welding Tig, and although we always use a fireproof blanket, its really pretty spark free. I used to just use a scratch start torch hooked up to the trailblazer in the back of the truck- the torch was $150, and I already had 100 ft leads. But now I have a little cart with a radiator, high freq unit, foot pedal, hooks for lead and torch, and a tank of argon mounted on the back. In the shop, a 304 inverter sits on top, and that powers it, but pull off the power supply, and the whole thing hooks up to the trailblazer. Instant tig on site, with all the bells and whistles. Quiet, quick, good looking welds with no cleanup, and minimal fire danger. Once you try it you will never go back to mig or stick inside a building.
We did a job a while ago where we were working inside a completely finished library, with all the smoke and fire alarms on, carpet down, hardwood handrails installed, and we snuck in and installed a bunch of metalwork inches from all those finished surfaces.
 
5 sided rooms with changing radius curves- this is why we love architects so.
Maybe if you make it in sections- then you only have a few joints to worry about. Cause I dont see how you are gonna get the whole jig out the door when it is all in one piece.
As far as welding inside goes- Tig is the only way to go. We do all our site welding Tig, and although we always use a fireproof blanket, its really pretty spark free. I used to just use a scratch start torch hooked up to the trailblazer in the back of the truck- the torch was $150, and I already had 100 ft leads. But now I have a little cart with a radiator, high freq unit, foot pedal, hooks for lead and torch, and a tank of argon mounted on the back. In the shop, a 304 inverter sits on top, and that powers it, but pull off the power supply, and the whole thing hooks up to the trailblazer. Instant tig on site, with all the bells and whistles. Quiet, quick, good looking welds with no cleanup, and minimal fire danger. Once you try it you will never go back to mig or stick inside a building.
We did a job a while ago where we were working inside a completely finished library, with all the smoke and fire alarms on, carpet down, hardwood handrails installed, and we snuck in and installed a bunch of metalwork inches from all those finished surfaces.
 
OH now I understand. You have a design problem not just a layout and fabricate problem. I had a stair mfg plant for many years where we made primarily curved wooden staircases, I have dealt with this type of problem before. That aside I will try to explain what your up against.
The pitch of any section of stair, curved or straight is the rise of one tread, divided by the run of one tread, or for a section of a given radius the rise of the that section divided by the run of an EQUAL number of treads. I'm sure you know all this, its only a recap and introduction to the real problem.
In this case every time the radius changes the run changes through that section, but hopefully the rise per tread remains the same. Each section has a differant pitch angle. Realistically and artistically they cannot just be connected togethor. There has to be a way to transition from one to other at an inflection point or ideally through an area where the change occurs. In an ideal world the orginal designer would have this is taken into account when the staircase was first designed. There are ways to facilitate this transition by progessivly varying the run of the treads approaching and leaving the area of transition. If it has not been done, which is probably the case here, then there is a real problem as to how to correctly join these sections of varying pitch togethor.
There is a book on stair design by a man named Mannes, that clearly deals with this design problem, very informative, but takes a bit of study.
This subject is a bit complex to continue with here but that's what you're up against. I am near you in Seattle. Contact me off line if you wish, I will be happy to talk further.
 
OH now I understand. You have a design problem not just a layout and fabricate problem. I had a stair mfg plant for many years where we made primarily curved wooden staircases, I have dealt with this type of problem before. That aside I will try to explain what your up against.
The pitch of any section of stair, curved or straight is the rise of one tread, divided by the run of one tread, or for a section of a given radius the rise of the that section divided by the run of an EQUAL number of treads. I'm sure you know all this, its only a recap and introduction to the real problem.
In this case every time the radius changes the run changes through that section, but hopefully the rise per tread remains the same. Each section has a differant pitch angle. Realistically and artistically they cannot just be connected togethor. There has to be a way to transition from one to other at an inflection point or ideally through an area where the change occurs. In an ideal world the orginal designer would have this is taken into account when the staircase was first designed. There are ways to facilitate this transition by progessivly varying the run of the treads approaching and leaving the area of transition. If it has not been done, which is probably the case here, then there is a real problem as to how to correctly join these sections of varying pitch togethor.
There is a book on stair design by a man named Mannes, that clearly deals with this design problem, very informative, but takes a bit of study.
This subject is a bit complex to continue with here but that's what you're up against. I am near you in Seattle. Contact me off line if you wish, I will be happy to talk further.
 








 
Back
Top