As a stair-builder, I'd appreciate a little bit of narration/or commentary on a stair modeling demo

I’m a stairbuilder and I do the kind of complex circular stairwork that shows up in a few YouTube demos. Rio Ryne has several very impressive ones. Even if I slow down the video I"m having trouble seeing and understanding the tools and extensions he is using. I’d appreciate any thoughts/suggestions on how I might determine the tools I need to be using within SketchUp and some strategy to apply them.
In the past I have tortured SketchUp into modeling somewhat reasonable representations of a circular stair build, However there clearly is a path with the newer tools to get results more eficiently and with much better results. Is there anyone who offers consulting or advice to help me navigate this process, It is important for me to be able to develop my own design ideas and represent them to architects and clients. The attached photo is an example of my work and the sort of project I’d like to be able to model. .



I’ve watched one of Rio Ryne’s stair videos before and I think they show quite clearly which extensions they’re using. If you want to know about a specific part, post a link to the video with a timestamp and someone here is bound to know which extension they’re using.

I’ve watched a little bit so far, but it’s interesting. I saw Fredo Scale used for curving it all. I can’t speak to the difference between Fredo Scale and True Bend for that task.

Beautiful work! Your work that is.

1 Like

Thanks for that. I’m going to give the time stamped linked video a try.
I don’t know if I have the stamina for this, but it seems like my learning (and discussing) the processes involved could be a useful discussion for others.
Some of the salient points for me:

  • tread nosing locations are usually not truly radial and are adjusted to generate a consistent rate of climb at the “travel line” (often the plan c-line) of the stair. This is easily developed in 2D cad then imported in SketchUp. There are other code issues connected to the minimum width of the narrow end of the winder treads, etc.

  • In the past I have imagined using cylindric surfaces with defined radii within SketchUp and then wrapping an imported 2D generated layout on to the cylindric surfaces. The faceted geometry used in SketchUp does not play well with that strategy. The cylinder segments need to be determined by their intersection with the push/pulled surfaces of the treads. What Rio and a few others show using either Fredo Scale or True-Bend is the ability to adjust those interconnected segments, maintain connectivity and also adjust tread widths at their inside and outside plan radii.

  • I am more than happy to let go of my primitve modelling strategy, However I want to maintain control over the design geometry and accuracy of the model. I may need to let go of some small discrepancies with handrail height, baluster spacing, etc,

  • There seems to be quite a bit of overlap in function between extensions. I’d like to settle on which ones best fit the situation

Does this make sense?
Thanks for your comments,------the journey continues

1 Like

Makes lots of sense. The bulk of my experience with curving stairs was from the '80’s when I worked for a firm that did multi-million dollar houses. Since then, I haven’t had such budgets on my own, but continued making interesting stairs with more affordable straight segments. Still, with winders and all, it can be challenging. One was in Fine Homebuilding (July 1998, pg 80-81). I did use a York Spiral Stair once (another Maine stair fabricator) seen here and here. That was prefabricated, but modeling it in SketchUp was my first try at curving stairs and helices with no plugins.

When it comes to certain difficult geometry problems, especially with arcs and circles, I tend to work stuff out in PowerCADD first and then bring in templates into SketchUp to continue into 3D. Somethings hold off until SketchUp resolves them in 3D. I wouldn’t try wrapping the 2D drawing onto faceted faces, though, as you said.

I always thought code cared about the edge of nosing, and then risers went where they needed to go relative to the nosing, but I’m an architect, not a fabricator.

My first woodworking job was with the business that became York Spiral and Duratherm Window many moons ago! Good company, and your installation looks great…
I’ve had some success since last I chimed in here. Yours and McGordon’s replies prompted me to look closer at Rio"s video. I’m attaching pdf’s of some preliminary results with Shape Bender. The info was there to be had.
Anyway, I want to keep this thread going because already it’s helping define the areas where some SU ace might make a suggestion.
The “stair plan” was generated in A-cad and imported into SU. The “full stair bend attempt” shows my initial setup incorporating both straight and curved sections of the stair, The bizarrely skewed plan section below the portion to be curved was worked out based on what I computed the inner radius arc lengths would become when that plan was bent at its centerline. The centerline is critical since it is what allows maintaining a consistent tread width there. Anyway, Shape Bender did not seem to want to incorporate straight portions along with curved.
“successful bend” is the result of leaving out the straight sections. What resulted is not perfect, but it is close enough to be an acceptable representation. Oddly, the bent stair came out with all curved risers (ie. - the front edges of the treads curved in plan) that I then needed to straighten out. I wish I had printed out a view to show - - maybe later.
And the 4tth pdf - “successful w straight sections in place”.

Random thoughts:

  • Rio’s version of a stair is marvelous but/and that continuous rib below the stair would be a fabricator extravaganza - probably best accompished w plaster work. And all curved risers on a stair like this? A few graduated curved ones at the bottom (or top if useful) would count for a lot. On the other hand, he is demonstrating the versatilityof SU with extensions. I appreciate the way he attacks and resolves the molding/trim issues that arise.

  • Rio’s strategy with treads and nosings involved applying them to the tops of the risers after push/pulling them down to allow for addition of the treads. For whatever reason, I’m going to attempt to create the nosing projection and finished riser face by copying the top front edge of the tread/riser down for the thickness of the tread and push/pulling the newly created lower face back to create the tread nosing. Because the tread/riser sections are not rectangular I’m hoping I’m going to find that extension push/pull tool that will get me there without too much fuss.

  • Most of Rio’s tread/risers sections are the same component and thus an edit of one changes the others. With my layout each of the tread riser section are unique… I haven’t yet converted them to individual groups.

  • When some of the spiral generating tools first started to show up for SU there wasn’t much they had to offer to me. Spiral stairs are a much simplified cousin of circular stairwork. Also what might work as a “circular stair here” place holder in a designer rendering is not what I need. It now appears that with the newer extensions I can get the correct amount of accurate detail. Shape Bender and Eneroth Upright Extruder particularly relevent. .

I’m going to post more on this project., and hope I get more useful feedback and discussion. Maybe there’s someone else building stairs w SketchUp.

Ooops forgot tthe attachmentsFull stair bend attempt.pdf (29.2 KB) stair plan.pdf (34.3 KB) successful bend.pdf (31.8 KB) successful w straight sectns in place.pdf (37.4 KB)

Wow, I’m more confused by how that works.

Shape Bender vs. True Bend is probably something to look into, but I’m no expert on either. @Box is the best person I’ve seen with True Bend. Maybe he can add some insight.

Judging by what’s happening in plan, I might be tempted not to use a bending approach. Very interesting problem in any event.

Back again.
I should have checked the overall dimensions of that bend that I’d labeled “successfu"l. Its overall plan centerline radius was oversized by 2”. Happily, I quickly and easily re-did the bend and all the important dimensions came out spot on. I have no idea what went wrong the first time.

I still have that weird curved riser result but that is easily corrected. In the future I will likely leave out the tread and riser surfaces and simply bend the 2 stringers; their saw tooth upper edges would allow for quick drawing in of the treads and risers.
Also, the advantage I see in the bending strategy is that allows for cosistent center-line spacing

See the attached.more successful bend.pdf (57.6 KB)

When it comes to stairs I follow the rule of 17, tread +riser equals 17". It works well for outdoors such as landscaping.
Outdoors a low riser works well such as 5" riser with 12" tread.

There’s also,
2 risers and 1 tread = 24" to 25"
1 riser x 1 tread = 70" to 75"


This a stair I built. Not only is the walk line consistent but so is the inside and outside stringers.
stair_mitered_nosing_with_blocks.skp (1.7 MB)

1 Like

Was that done with your stair builder plug-in? Very clean one way or another.
I see from your edit it was!

No - not completely.

However it was helpful. My stair builder plugin works with continuous curves - currently it does not handle change of pitch etc. It does help with certain features.

The 2D to 3D utility can certainly speed up various aspects.

1 Like

The issue for someone who does much custom circular stairwork professionally is that one needs to have the options that are afforded with changes of pitch and plan curvature. The challenge for me is to be able to edit and model the design when it falls outside simpler configurations.
Anyway here is my latest challenge. ------and it’s probably fairly simply addressed. I’ve edited the model to eliminate the curved risers/treads (see “more successful bend” pdf above). To make the tread nosing overhang I push/pulled the riser faces back by 1". This left funky geometry where the “pushed” riser surface (for pretty obvious reasons) did not really intersect with the curved vertical surfaces of the stringer.
Any suggestions on how to simply extend the pushed riser surfaces to intersect with the stringers? I’ve tried some basic manual editing. but I’m getting lost in that weid curved surfaces world. I suspect there’s a tool or strategy I’m missing. Is this where Su4u weld applies?pushpull to make stair nose-.pdf (45.4 KB)

Off the top of my head, I don’t think I would push=pull the riser back to get the nosing. I might try drawing the profile of nosing and riser and extruding it as a linear object. Make it overly long (runs past the stringers), and ask SketchUp to intersect the surfaces. Just a general strategy.

I think you’re right. There’s a second stair that sits directly above the one I’ve drawn already; same plan except different tread to tread rise and one additional tread. I’m drawing it now, (as Rio Ryne did and I should have) with the risers as the controlling locator and then applying treads on top of them. Somewhere along the line I saw an extension video that seemed to affirm my first approach, but damned if I can find it.
FWIW, - - when trying to edit the nosings on this first stair I did try extruding or push/pulling through the curved stringer face and I couldn’t get the intersect tool to cut or define edges on the stringer… It’s that curved surfaces thing that I haven’t gotten a handle on how to edit.

Hi volute54,

Thats a beautiful staircase indeed, however, I thought there were highly detailed bespoke parametric cad software programmes to assist for this type of work ?
Interesting links below:-

Thanks for the suggestion. I’m aware of those programs .

The programs that will easily produce what I’m after are expensive and make sense for a shop with a much larger production capacity. They are also aimed at multi-axis CNC production of custom railings. I do that work without CNC - old school. ( I do job out some of the work to CNC) The shop lay-out and materials spec is not really much of an issue. 2-D cad works fine for that, Just want to better and more easily represent to the client and architect what the final result will be. FWIW, Sketchup seems to be getting more capable of getting me there.

1 Like