First Feather Pattern

[Bruce Bump]

I got 1084 and 15N20 from Kelly that worked real fine for some small simple random and twist billets I made up at Indian George's with his press. Kelly will cut stuff to fit in flat rate priority mail boxes too. He'll shear the 15N20 to the same width as the 1084 too, at least that's what I did. The 1084 was 1/8" thick and the 15N20 1/16". I initially thought I'd double up the 15N20 so it was the same thickness as the 1084, but didn't end up doing that.

Phil, thanks for the reminder, Yes I did double up the 15n20 so it was approximently the same thickness as the 1084. It seemed to help the pattern allot.

 

[NickWheeler]

I forgot I had these pics on my laptop.

Might help some more

Initial stack of 1084/15N20 prior to being forge welded.

The billet was forge welded solid, then turned 90 and drawn out to 3 times its original length (another raw billet for refrerence)

The billet was cut into three pieces, ground clean, and is welded... ready to forge weld solid.

This is just the VERY START of W's. The billet was drawn out, cut, and stacked into 3 layers. You can see how the vertical lines are being deformed to what is starting to look like a C shape. More crushing and they look more likes C's. More stacking and crushing and they look like sideways W's.

 

[A C Richards]

I agree Bruce. I would also like to know how many layers in that shot. It is probably the perfect #. Bruce I have had a TIG welder for about 6 years now. I got one when my old boss upgraded his machine and my Son bought the old one. I bought it from my son ans would not be without one. I also have a nice mig welder with a spool gun so i am covered. Take a look at the inverter machines. The one I have welds smoother than the big Lincoln I used to have.

Mike, As far as settings go, I think I am at about 80amps DC. I just weld and if it isn't hot enough I turn it up. Warming up the billet can prevent shrink cracks. The biggest thing is to keep your electrode clean and sharp.

 

[Mike Krall]

Thanks a lot, Chuck...

Mike

 

[BrianThie]

This is turning into a very informative post. Thanks for starting this Chuck and to everyone else that has contributed. It seems to me that the physical size of the billet and the thickness of the pieces used has a direct relationship on the total layer count needed to get the same results.

I tried my hand with the feather pattern more of an experiment than anything to get my mind off of other things going on. It was done on a small scale starting out with 17 layers 4 inches long. The attached photos show 3 pieces that were cut off of the billet where I welded the ends together.

The first tile (bottom) is the end of the the initial welded billet after it was drawn out.

The next tile was the end of the billet after I cut the first billet into 4 and welded and drew out

The final tile was after I took the second billet cut into 4, welded and drew out, then cut into 5 welded and drew. At this point, I was at 80 layers. I stacked this into 3 for a total of 240 layers before I made my cut. The pattern at this point was much too fine. I should have stopped at 80 for the material and billet size I was using I think.

Brian

 

[A C Richards]

I agree. I have learned more by reading these posts than when i made the billet.

 

[Phil Dwyer]

Yep, sure leaning a lot! Now if Chuck, Bruce and Mick each tell us how many layers they did for the finished billet photos they posted we'd add even more to our database. Thanks so much guys. -Phil

 

[Mick Maxen]

This bar was made a few years ago, but as far as I can remember we had about 20-25 layers in the first billet. Then we followed Steve Dunn's "recipe". Steve mentions about doubling up the layers. I think he only did this for mass, to get large slices out of the last bar for a knife with a blade about 10" long, if I remember correctly.

Mick.

 

[Bruce Bump]

Yep, sure leaning a lot! Now if Chuck, Bruce and Mick each tell us how many layers they did for the finished billet photos they posted we'd add even more to our database. Thanks so much guys. -Phil

I tried to count Micks layers and came up with 100 or so.

Remember you count the intial 20-25 as one single layer.
Dont forget we are looking at the "end view" of everything.

I think he has about the right combination. I think mine has about 35. We havent seen Nicks yet.

 

[NickWheeler]

That of mine that were successful are long gone

But the next time I do one I'll post it... maybe it will be the jellyroll type

I agree Bruce... Mick's is spot on fantastic. :thumbup:

 

[A C Richards]

I am trying to remember exactly what it was. I thought it was 4-5 stack welds. But now that I think back it was probably 2-5 stacks and one 4 stack re-weld. (I was doing 3 different billets at the time). So that actually works out to 100 layers. Like Bruce pointed out you do not count the first 21 layers. If you did that would be equivalent to 2100 layers in a standard billet. It sounds like the 100 range is the target for this pattern.

 

[Phil Dwyer]

Well if I was confused before, I'm hornswoggled now.

If you don't count the first setup of alternating steels then your second layer (the first you count) only has three layers? And the billet increases incrementally by the power of three with each drawing out, cut and stack? So it would take 5 cut and stacks, which would be 3 to the 5th power (3, 3x3=9, 3x3x3=27, 3x3x3x3=71, 3x3x3x3x3=213) to equal 213 layers? I guess I can see not counting the first setup as layers if you rotate them after the initial forge weld. Do you still not count the first setup of layers if you DON'T rotate them after the intial forge weld?

So where can I read and learn more? This is a bit like jumping in the deep end of the pool.

 

[NickWheeler]

Phil, you're not confused, you've got it!!! :thumbup:

Look at those pics I put up, the first set up was welding about 20 some layers solid, but as soon as I rotated it 90, it became one layer. So if you scroll down a bit to the pic taken after the initial billet had been drawn, cut, ground, stacked three times, welded, the end ground and etched.... you'll see that the end grain shot is just 3 layers.

This is only because of the type of patterning.

A regular billet, like random, ladder, etc... would not be turned 90 (for the most part) and those first 20 something layers would be part of the equation. (20, 60, 180, etc.)

 

[A C Richards]

Any of the mosaics are counted differently. I did a high density radial pattern (my avatar) that the initial layer count was 15. The first block had 30 radial layers, I arranged 32 blocks around 1 large block for one tile and 5 of those tiles in a blade. so 30x33x5=4950. But I just called it a mosaic. This was actually the first time i added them all up. Once you re-orient the pattern the layers make little difference except for pattern development like we have been discussing about being too fine etc. So if you look at Nick's post of the end grain the bottom block is basically the first layer. Then you have it right if you do a 3 stack re-weld. if you do a 5 stack then you are doing multiples of 5. That is why I got to 100 layers with 2-5 stacks and then a 4 stack. simple patterns are counted from the initial stack, 21x5x5x5=375. Are you totally discombobulated now?

 

[Steven Penner]

Good info...but more questions

The final thickness of the layers in the pattern are functions of:

thickness of initial layers
final thickness (height) of the billet
number of layers ( we all got that ok)....seemed to reach a consensus of about 100 layers for this pattern

These all combine to create the final layer thickness.


but we seem to be ignoring the initial layer thickness...

What initial stock sizes and thickness are each of you using?
because width of the origional stock becomes thickness of the "W"


Extreme examples:

10 layers each 1", compressed to 1 inch bar = .100 per layer...too coarse

10 layers each .010 = .100 thick bar & .010 per layer...a pattern



A 200 layer pattern with 200 layers in a 1/4" bar =800 layers per inch
A 200 layer pattern with 200 layers in a 4" billet = 50 layers per inch

It would make sense to me to standardize terminology in
Layers Per Inch

Thanks for this thread
Once i understand it all, I'll have to scale it down and do a miniature by hand like Dave has in mind.

 

[Mick Maxen]

As Nick and others have mentioned, it can get confusing with this type of pattern for how many layers or bits of steel. Let me see if I can explain it.

If you start out with 20 bits of steel in the first billet, as Steve Dunn has.
You weld the first billet of 20
Draw this out and cut into 5 and weld = 100 bits of steel
Cut in half, weld and draw out = 200 bits of steel
Cut in half, weld and draw out = 400 bits of steel
Cut this into 8 pieces, stack and weld = 3200 bits of steel

For layers it goes like this,
Weld the first billet = 1 layers
Draw this out and cut into 5 and weld = 5 layers
Cut in half, weld and draw out = 10 layers
Cut in half, weld and draw out = 20 layers
Cut this into 8 pieces, stack and weld = 160 layers

So following this process you have reproduced the first billet a 160 times and in that first billet was 20 bits of steel, 160 x 20 = 3200 bits of steel

I have only been awake for 1/2 hour, I hope this makes sense.

Mick

 

[ddavelarsen]

Well if I was confused before, I'm hornswoggled now.

...

So where can I read and learn more? This is a bit like jumping in the deep end of the pool.

Truer words were never spoken. I'm convinced now this ain't likely to happen working by hand, so now I'm going to try to find someone within a couple hours' distance that will be patient enough to work with me on learning this stuff. Meanwhile, you guys keep it up - this is pure fascination. Thanks!

 

[A C Richards]

I used .25x1.5 1084 (11 layers) and .062x1.5 15n20 doubled (10 layers). From what I understand Bruce use .125 1084 and .062 15n doubled. It does make some difference. How much I do not know at this point.

 

[Mick Maxen]

By messing about with thick or thin layers, the only difference you will get is contrast between light and dark, as far as I know.
For a dark pattern with flecs of silver/shiny areas use a thicker plain carbon steel. Or if you want shiny with flecs of dark, use thicker 15N20.
The pattern Owen and I made had layers the same thickness and I think the balance is about right.

Mick.

 

[Phil Dwyer]

i Forgot I Had These Pics On My Laptop.

Might Help Some More

Initial Stack Of 1084/15n20 Prior To Being Forge Welded.

The Billet Was Forge Welded Solid, Then Turned 90 And Drawn Out To 3 Times Its Original Length (another Raw Billet For Refrerence)

The Billet Was Cut Into Three Pieces, Ground Clean, And Is Welded... Ready To Forge Weld Solid.

This Is Just The Very Start Of W's. The Billet Was Drawn Out, Cut, And Stacked Into 3 Layers. You Can See How The Vertical Lines Are Being Deformed To What Is Starting To Look Like A C Shape. More Crushing And They Look More Likes C's. More Stacking And Crushing And They Look Like Sideways W's.

What Happened To The Photos???