Nautilus out of an I-Beam - Questions on Bending

[Shootr]

I saw this on Pinterest and it is something I want to try and make a version of.



I'm nothing more than a flux core welder wielding hobbyist that has been tinkering for a year or two at most - so EVERYTHING I do is an experiment/learning experience.

This is a beam I picked up at a scrap yard - roughly 4" x 6" x 10' long:



I guesstimated the sections to be cut and arrived at 1", 2", 3" and 4-1/2" - the number of each dictated by the count on the original:

[Welder Dave]

Would have been much better not to have a galvanized beam. Breathing galvanized fumes is very bad. You need to have a proper respirator and exhaust in your shop/garage. Does the scrap yard have any non galvanized beams? You would be far better off considering the amount of work required to the beam. A couple cuts and welds on galvanized is one thing but you're looking at extensive cutting and welding for your project. The last thing you need is zinc chills.

[Shootr]

I plan on using a portaband saw to make the cuts, through the top flange, through the web, stopping just as I get to the bottom flange.

Question #1: What tooth type/count and FPM would you recommend to make the cuts?

Once (if) all the cuts are made, comes the hard part - bending each piece by spreading the cuts apart - how to do it.

I don't have a torch kit, but if the consensus is that it will be required to make the bends, then it's on the procurement table with my war department.

Question #2:

Since the cuts/blades are only 0.020" wide, I'm not sure what method would work or not.

* A 3lb hand sledge and good aim?
* Some kind of steel wedge to pound into the cut to start to spread it open?
* A long stout prybar to beat into the cut and try to muscle it open?

My layperson's thoughts include:

* It's probably the lightest I-beam I've come across, hell I can lift and carry the thing around fairly easily.
* The flanges are not particularly thick or beefy, seems to be as noob-friendly as a beam could be.

But every project and effort I've done has brought unforeseen surprises, this time I'm hoping this forum and it's members may be able to warn me off about things I hadn't thought of.

Thanks for looking and thanks in advance for any and all thoughts, ideas, and advice.

[Shootr]

Would have been much better not to have a galvanized beam. Breathing galvanized fumes is very bad. You need to have a proper respirator and exhaust in your shop/garage. Does the scrap yard have any non galvanized beams? You would be far better off considering the amount of work required to the beam. A couple cuts and welds on galvanized is one thing but you're looking at extensive cutting and welding for your project. The last thing you need is zinc chills.

I have a Honeywell half mask and I use the organic vapor/P100 particulate filters at work (copper mine crusher dust in confined spaces). Even though it's here in sunny Arizona I work with the garage door open and a fan blowing on me as I work on stuff. I think that should be adequate here, but this will be the first time I've worked on something this big full of zinc.

[Welder Dave]

It looks like the one you're trying to copy each individual section is welded. Would be good to have a table to lay it out on.

[OPUS FERRO]

Shootr

. . . so EVERYTHING I do is an experiment/learning experience . . .

Galvanized 'donor material' is a non-starter* . . . for your aspirational product.
Pitch - 'all of your good work' in the Dumpster - you are wasting your time . . .


Opus



* J.O. / Univ. Weld Inst. - 'You start with feces - you end with feces' . . .

[BOP101]

I looked at the photo for a while and it looks like there wasn't any bending involved at all. I counted approximately 41 I Beam slices that were then welded back together to form the "Nautilus". You will observe the bottom flange has weld on them. If bending was incorporated then there would be no need to weld on bottom flange. Just my .02

[tapwelder]

I also looks as if the beam only comprises a small portion of sections like 5 chunks? Then it spreads wider. Seems mathematical?

As stated there is no bending done there. And the sculpture is more complex than a so glad beam

I would cut. Form a template(s) and weld. Where a respirator and use a fan outside you should be fine.

You did state a "version of"... So carry on


Good luck

[VaughnT]

As others have noted, the original is cut and welded, not bent. Bending .20" thick steel isn't hard, though it's a thorough pain in the butt when it's 4.125" wide and you don't have an OA rig and a big table to lay it out on! It's doable, but a ton of aggravation.

I'd first draw out the scroll you want to form. You can use the Fibonnaci sequence to get a nautilus curl going, but that might not be the perfect look for the stock you're trying to form to it. A good starting place, though, and the golden ratio has been used for hundreds of generations.

[psacustomcreations]

I used a technique similar to what you are suggesting to make the piece below. I cut part way through the channel then bent using my 20 ton press. Doing it by hand was a no go. My piece of channel was not galvanized so I did not worry about the hazards.

In your instance, the example was cut completely apart then each piece was welded together While this sound like more work, it is actually the easier and correct method. Trust me on this. It will be very hard to cut part way through then try to get all the bends correct. Very hard.

If you are dead set on using your galvanized beam, this also would then make the complete cut and weld method more appropriate. You can cut the beam with your portaband or a cut off wheel. Then take the sections outside away from everything. Put the sections in a bucket with a lid. Pour some muriatic acid in the bucket and close the lid. Wait 20 minutes then remove. If the galvanized is not off, scrub a bit with a wire brush and soak again. Wear face and eye protection the whole time. Once the pieces are clean, dunk in a baking soda and water bath followed up by a fresh water bath.

Getting an I beam that is not galvanized will save you a few hours of work. Maybe use this one for a future project.

Layout your design on the floor or table using the golden ratio as Vaughn and Opus mentioned before. Tack weld both sides of the flange so it does not warp and curl more than you want. If you start at the small pieces and weld one or two at at time, you can then use a grinder to smooth the welds so they are not seen.

[psacustomcreations]

To add to my post above, if you are going to use the galvanized beam, leave it in one piece first. Take 10 minutes and run a grinder all over it a few times. If you can remove 40% or more of the galvanization in 10-15 minutes, you will save hours of soak time. That also saves time for the muriatic gases to get where you don't want them and rust everything in sight.

[Louie1961]

I also looks as if the beam only comprises a small portion of sections like 5 chunks? Then it spreads wider. Seems mathematical?

Not a question you asked but a long time ago I learned the mathematical formula that describes the curve of the nautilus. I don't remember it off the top of my head, but you can probably google it. It should be easy to find and will help you lay out your cuts more precisely.

[dmcnally]

Not a question you asked but a long time ago I learned the mathematical formula that describes the curve of the nautilus. I don't remember it off the top of my head, but you can probably google it. It should be easy to find and will help you lay out your cuts more precisely.

I think you mean the Archimedes spiral. Here's one link and here's another.

Dave

[Shootr]

Wow, profuse thanks to everyone (sans one) for taking the time to detail many great observations and advice. I want to address each in a separate reply so if a conversation breaks out, I can keep it straight lol. Again, thank you all.

[Shootr]

It looks like the one you're trying to copy each individual section is welded. Would be good to have a table to lay it out on.

I looked at the photo for a while and it looks like there wasn't any bending involved at all. I counted approximately 41 I Beam slices that were then welded back together to form the "Nautilus". You will observe the bottom flange has weld on them. If bending was incorporated then there would be no need to weld on bottom flange. Just my .02

When I stopped by my welding supply shop, they also noted that the original looks cut and welded. My thought was that if it could be bent instead it would look cleaner. He too had doubts about being able to spread each cut symmetrically and keep the overall shape plumb and straight.

The other fear I have is warp - I'm still learning how to control it as best I can, so trying to weld 40 +/- pieces straight seems like worse odds than bending it .

[Shootr]

I also looks as if the beam only comprises a small portion of sections like 5 chunks? Then it spreads wider. Seems mathematical?

As stated there is no bending done there. And the sculpture is more complex than a so glad beam

I would cut. Form a template(s) and weld. Where a respirator and use a fan outside you should be fine.

You did state a "version of"... So carry on


Good luck

As others have noted, the original is cut and welded, not bent. Bending .20" thick steel isn't hard, though it's a thorough pain in the butt when it's 4.125" wide and you don't have an OA rig and a big table to lay it out on! It's doable, but a ton of aggravation.

I'd first draw out the scroll you want to form. You can use the Fibonnaci sequence to get a nautilus curl going, but that might not be the perfect look for the stock you're trying to form to it. A good starting place, though, and the golden ratio has been used for hundreds of generations.

Not a question you asked but a long time ago I learned the mathematical formula that describes the curve of the nautilus. I don't remember it off the top of my head, but you can probably google it. It should be easy to find and will help you lay out your cuts more precisely.

I think you mean the Archimedes spiral. Here's one link and here's another.

Dave

The two links Dave posted are what I've been researching. In my opinion, the nautilus shape falls in between the two - looser than the Archimedes example, but tighter than the Fibonacci equation. The "golden spiral" link had another link that explains phyllotaxis (wow am I getting schooled!!!) which is closest to what I am trying to achieve.

[tapwelder]

The two links Dave posted are what I've been researching. In my opinion, the nautilus shape falls in between the two - looser than the Archimedes example, but tighter than the Fibonacci equation. The "golden spiral" link had another link that explains phyllotaxis (wow am I getting schooled!!!) which is closest to what I am trying to achieve.

You are likely correct about the tight Fibonacci spiral. I have found it difficult to maintain a manageable size if you create it from a pattern. It takes off/grows fast. Note the fib scroll is just connected 1/4 circles of increasing size. So, gauges are easy to make for you to hold against your sculpture.

I once did some work for a wrought iron dealer. Aside from trying to make a business selling only antique wrought iron his fatal mistake was buying it from the Middle East. The Archimedes scroll appears to be most common there. So, his stuff was not appealing considering what folk are used to seeing. Quality and workmanship and traditional joinery were there, but the scroll (spiral) look was different. It was not difficult to determine regions from where items came from by design.

Good luck

[Shootr]

If I'm allowed to stay on this forum, you'll soon see I have no shortage of weird off the wall ideas...

So what if I cut into the bottom flange a little, making a score line if you will. Then put a piece of stout angle iron under the beam at the bend point, and use something like this to wail into the saw cut to try and spread it.



Possible need to scab on a wider wedge that would be the full width of the beam? Or get really fancy and upgrade to a hydraulic/manual log splitter and modify it to work!

[tapwelder]

I would not expect difficulty bending is your reduced the flange to a bout 1/8 inch. The wedge would work.

Got heat? Then you.could just cut away the web.

The difference between .2 and .125 is substantial when bending manually.

[cwby]

I think it could be bent. If you draw out a pattern on cardboard & bend one segment at a time matched to your template, then tack a brace across the open end to hold it in place, then cut & bend the next segment, & so on. I would try to start with the tightest bends 1st - more leverage for those tougher bends. A come a long would be my tool for force & I would start with a riser in the middle to hold the cable at enough angle to get leverage to get it started (think truss with cable as rafters). Once you get the curl started there should be enough leverage to make the rest of the bends.

[BD1]

That's slick for sure. I'm gonna have to check my I-beam stock for length and thickness.
Saw cutting would definitely be cleaner then cutting and welding. Depending on beam size and your cutting tools I would saw cut.
As a reminder, you can cut straight from top flange, through web AND into the inside of bottom flange. That inside saw kerf will make forming the bend easier.
A stout welding table would help and a come a long. A car or truck rim may work for a jig to bend around. Lots of options.
I'm looking forward to seeing it IF you try it.
Good luck.


Sent from my iPhone using Tapatalk

[Shootr]

I would not expect difficulty bending is your reduced the flange to a bout 1/8 inch. The wedge would work.

Got heat? Then you.could just cut away the web.

The difference between .2 and .125 is substantial when bending manually.

I think it could be bent. If you draw out a pattern on cardboard & bend one segment at a time matched to your template, then tack a brace across the open end to hold it in place, then cut & bend the next segment, & so on. I would try to start with the tightest bends 1st - more leverage for those tougher bends. A come a long would be my tool for force & I would start with a riser in the middle to hold the cable at enough angle to get leverage to get it started (think truss with cable as rafters). Once you get the curl started there should be enough leverage to make the rest of the bends.

That's slick for sure. I'm gonna have to check my I-beam stock for length and thickness.
Saw cutting would definitely be cleaner then cutting and welding. Depending on beam size and your cutting tools I would saw cut.
As a reminder, you can cut straight from top flange, through web AND into the inside of bottom flange. That inside saw kerf will make forming the bend easier.
A stout welding table would help and a come a long. A car or truck rim may work for a jig to bend around. Lots of options.
I'm looking forward to seeing it IF you try it.
Good luck.


Sent from my iPhone using Tapatalk

Would any of you have advice on what tooth count and blade speed I should try with the portaband saw?

[dmcnally]

Would any of you have advice on what tooth count and blade speed I should try with the portaband saw?

The general rule I learned at the junior college I went to is: 3 saw teeth across the thickness of the item being cut. For the 0.2" web (straight up and down) that works out to 15tpi. If you go flat across the top of the I beam, then you'd want a more agressive blade (4-6 tpi). If you cut the top of the I beam at an angle, then the 15tpi blade would work okay.

I use Lennox blades in my horizontal bandsaw. They have guides to help you choose the right blade on their website.

Good luck,
Dave

[OPUS FERRO]

Shootr


'non sequiturs' have consequences . . .

Either - you are 'technically [a fool] challenged' - or you are Spam . . .


Opus

[Robb M.]

test post