Terry_Cowan (16K)




Announcement: Please enjoy your visit to our forums.

Threaded View
Conversational View
Thread Notifications: Off
ForumID: 109061855423990 ThreadID: 9990174032612 PostID: 9990176944855
typhoon

Posts: 5
Status: Offline

Hello, I'm new to this forum, very glad that such an area exists. Are there many advanced metal shapers here? I'm dealing with 2024-t3 on a regular basis, is this forum very much aviation related? Can anyone help me with the following? ...Is there a certain process to follow to make a reverse compound curve with the wheel? Is anyone aware of one?

Reply
Edit
Highlight
MoreInfo
Recommend
ghiafab

Posts: 95
Status: Offline

For a reverse, you need to stretch the edge of the panel, and then gradually taper the amount of stretch as you go into the panel. I think Kent White has a video that shows this...lots of aviation related stuff...be sure to look at the rest of his site while you are there:

http://www.tinmantech.com/html/tmvideos_list.html

John www.ghiaspecialties.com

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

Ian Slater wrote: Are there many advanced metal shapers here? I'm dealing with 2024-t3 on a regular basis, is this forum very much aviation related?

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Ian, We have several advanced metalshapers in the group and a number of aviation guys. The group is made up of all ages from all walks of life, with the common thread being the love of making things of beauty from metal. Lots of old hot rodders 50 and older and lots of bike enthusiasts, mixed in with sculptors, blacksmiths, war bird restorers, auto restorers and general fabricators... we are a real mixed lot (smile).

What is the size and shape of the reverse curve you are making? Can you post a photo? Maybe someone could post a series of photos showing how to do it on the wheeling machine.

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
unklian

Posts: 517
Status: Offline

A picture would be real handy for this sort of discussion.
My searches for a good example came up empty.

Reply
Edit
Highlight
MoreInfo
Recommend
youngwally2002

Posts: 8
Status: Offline

If the reverse curve is to one side of the panel, such as a hood leading to the windshield, then it is usually an easy task, for the main part of the panel is already shaped. The area where the change of shape is to take place should be marked so that no wheeling will go beyond it in to the panel. The panel will be upside down in relation to regular wheeling, and will begin with stretching the edge with three or four tracks, then close track inboard for about two inches and track back out again. Repeat this, but track inboard for about four inches, then track out again. If the new shape is further inboard, for instance(six inches), then start again with a two inch, followed by a four inch, followed by a six inch. You get the idea, for each two inches further in board, do a two, then four then six, and so on. This way, the edge gets plenty of stretching in relation to the inboard section. Start with a medium to light pressure between the wheels, and if the reverse shape does not progress fast enough, then increase the pressure. You will be using the flattest anvil wheel to suit the curve you are trying to get. Two points to remember, is that you will require a template of the reverse shape that goes well into the main panel, and that the cross shape of the panel is in the correct shape when checking, other wise you will get false readings from the template.

Reply
Edit
Highlight
MoreInfo
Recommend
admin

Posts: 146
Status: Offline

John Glover Wrote:
Two points to remember, is that you will require a template of the reverse shape that goes well into the main panel, and that the cross shape of the panel is in the correct shape when checking, other wise you will get false readings from the template.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

John,

Can you explain what you mean by "and that the cross shape of the panel is in the correct shape when checking"? I'm not sure I understand how we would get the false reading.

Thanks,


Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
ghiafab

Posts: 95
Status: Offline

Terry,

I think John G is talking about the form/arrangement/configuration of the panel. If you are checking the curve in one direction, make sure the curve that is perpendicular to it is not out of its desired configuration. If the reverse is front to back on the side of a car, changing the curve up and down will make the front to back read differently.

John www.ghiaspecialties.com

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

Well I finally got to spend a couple hours in the shop and worked on making a reverse curve.


This is just a freeform piece -- I'm not matching it to any template. Its just roughed in and still needs smoothing.

Making the reverse curve just as it is on an actual 1930s fender was pretty easy, because the crown does not go all the way to the part that meets the running board. At a point, the top of the panel goes fairly flat as it approaches the running board -- because running boards are flat.

Where I am still having a little difficulty is in making the part as if the running board were rounded. (Keeping a crown all the way through a reverse curve and into the running board, as if the running board were crowned) I'm almost there -- but not quite.

I realise there may not be any cars with round running boards -- but I would just like to be able to make that shape if need be. As soon as I go through the rest of the photos, I'll post a few more.

This was all done on the wheeling machine -- no bag, But I cheated with a shrinker/stretcher.

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
This post has been modified
ghiafab

Posts: 95
Status: Offline

Terry, you should do this more often! Nice piece. I look forward to more pictures. John Glover's two part "how to make a model A fender" video shows how he did a similar shaped fender. Well worth having in the video library.

John www.ghiaspecialties.com

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

At 01:05 AM 10/14/2004, Dave wrote:
Ah, yes come on Terry hurry up, can you also show the various lower anvils you use as well please.
---------------------------------------------

Dave.

I just loaded some of the images in an album:
http://allshops.org/cgi-b[...]5013509


I'll post each of them here in the forum with comments for discussion. Please note: This is just an exercise -- I'm not saying this is the "proper" way to do any of this. It is just how I managed to get it done in a "quickie fashion" -- and I would love to hear about other approaches.

As for anvil changes -- honestly -- I didn't change anvils at all. I just used what was loaded on the wheel. (lazy). The general rule is to use the most flat anvil possible and progress to the final shape.

But I just wanted to "whip" something out quickly -- so I began with an anvil that turned out to be the final shape. In the beginning it streaks the metal a little -- but in the end it all seems to smooth out just fine.

I began forming a curve in the panel and bending the edges over using a wheeling machine with an inner tube on the upper wheel. Not that this process was necessary -- it was just to demonstrate that it can be done this way. Using the rubber on the upper wheel limits the amount of stretching to a negligible amount. In some cases it would be very important not to stretch the metal. This process works fine in some cases, but I'm of the same opinion as Jim Bailie -- there usually seems to be faster ways of doing it:

It would have worked much better if I had taken the time to change the anvil to a small high crown to bend the edges over.

The marks on the panel are not meant to represent exact passes -- but rather a representation of the difference in the number of passes needed to form the gentle curve in the center of the panel in contrast to what is needed to bend the edge in a 90 degree bend.

Although it probably doesn't really make a lot of difference when using the rubber on the upper wheel -- I didn't go all the way to the end of the front because that area will be pushed over and shrunk in a different process anyway. If we were to have skipped the rubber and done this with a bare upper wheel, thereby stretching the metal -- then you certainly do not want to go all the way to the edge, since you would be stretching the area that you will later be shrinking.


This is just an ordinary inner tube, cut and pushed on the upper wheel. This one is rotten -- so I'm going to go buy a new one.



You can see how much contact there is by the shiny area on the rubber.


I just didn't have the patience to change wheels and continue with the rubber covered upper wheel, when I had a leather covered slapper handy and something to hammer it over. At least I'm still using the wheeling machine -- right? (smile)



So here is what we end up with.

More to come....

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
This post has been modified
392hemi

Posts: 231
Status: Offline

Joe Stafford Wrote:
Terry,
Just got this email, our server has been down. Hopefully it's not too late to use my input.
Typically 2024 does not respond well to wheeling. It has a rather high Brinell hardness rating somewhere around 120. Yield strength is also somewhat high, comparable to Alloy 6061.00 Couple that to a moderate temper and you can understand the difficulty this individual may be experiencing. However, a slight reverse on the compound can be attained before work hardening renders the material unworkable by turning the work material up side down and wheeling the edge 1" into the center of the work piece and then wheeling out to the edge. Then wheel back in 2" to the center and wheel out. Repeat this process and go 3" in and then back out to the edge. Repeat this processs over and over going in as far as necessary until you achieve the desired result. As always, practice makes perfect.

Hope this helps.

Joe

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Thanks Joe! I can see how turning the workpiece upside down could work where there is little or no side-to-side curvature to the panel. On a related topic (but not using 2024) -- I'm still trying to deal with a shape similar to one half of a cylinder made into an "S" shape. ( I'm using 3003 H-14).

Turning that upside down would probably not work so well, since it would tend to flatten out the side-to-side shape. I'm playing around with it and posting it in this forum. Wish me luck (smile)

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

Dave wrote:
Ah, a big thanks for the album Terry, I have been eager to make some front fenders like this, for my 37 chev but just trying to squeeze in the time. You have given me some good insight thanks.

There was a mention about the john glover video (though I am planning on purchasing some of the videos very soon) can some one tell me how much more detailed does john go into regarding making the fenders that he does?

Also it is was my understanding that these fenders where typically made in separate parts I have seen a pic of them shown them to be all broken into different sections, from what you have shown Terry you maybe able to get away with making maybe two sections one for the main fender and the other part that reverse curves back to the engine area to act as the bolt up section. Hmm. Thanx, Dave, Mt Macedon, Australia
===================================================================

Dave,

John goes into great detail. As for how many pieces you can get away with -- remember the piece I'm experimenting with is much smaller than an actual fender and I'm not having to match anything. So there is a big difference in what I'm doing in this thread and in making a full-sized fender. Sometimes you will want to divide up a panel just to make it easier to handle, since you can weld two pieces easier than you can struggle with getting the entire piece in a wheeling machine. As many have said... sometimes it comes down to whether you are a better welder or a better shaper (smile). It also may depend on how much equipment you have available. If you have a shop full of Marchant shrinkers, Pullmaxes, Eckolds, and power hammers -- you can probably cut down on the amount of time it will take to make a part in less pieces.

If your equipment is limited -- it may make more sense to divide it into easier-to-chew bites.

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

Unfortunately I didn't get many photos of this process -- but this is shrinking with the wheeling machine. It is much like tuck shrinking -- except you form the puckers by pushing down on the panel while running just the edge through the wheel. The trick is to tip the anvil so it isn't pinching an stretching the outer edge of the panel. Also back way off on the pressure so it will allow the metal to thicken. You can even put tucks in with a tucking tool and work them down with the wheel.



You don't get a real pronounced pucker as you do with a tucking tool -- but you get something like in the photo below:



Shrinking with a wheeling machine is a rather laborious process, so I just finished up with the Marchant:


The shrinking almost magically arcs the piece (I love this part). But it is so much of a shrink that it puts a little kink or bulge above the shrink. However, this is easily dealt with. The wheel smooths it out with no problem. Just be sure to have very light pressure so it will compress the bulge rather than stretch it and make it worse.



Now -- speaking of a shrink... if you have never seen one up close made by a kick shrinker -- here's what they look like:



You can see the stipple marks on either side and you can even see how the metal thickens. The nick in the middle was caused by my dies not being exactly right. The area marked "C" is approximately what is affected (compressed and thickened). "A" shows how much thicker the material is after one kick on the shrinker. "B" is the original thickness of the panel.

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
This post has been modified
392hemi

Posts: 231
Status: Offline

More on the reverse curve...

After wheeling in both directions to form a compound curve, I used the shrinker to shrink the edges and went back and wheeled more to increase the crown. It is amazing how much change a little shrinking will make.



I kept on increasing the crown and shrunk farther down the edges to get this. Now the remaining edges (near the running board) will have to be stretched to allow the reverse. That's where it begins to get a little tricky.



As I began stretching the edges -- it started looking a little scary. Unlike the cooperate effort I had when shrinking the other end -- this end was putting up a fight. There is resistance from the center of the panel. If I had left that end of the workpiece flat, then the stretching would be yielding more promising immediate results. (I'll set up a sample to demonstrate what I'm saying here and post a photo later) But since I had curved the part all he way to the running board end -- that makes things more difficult. Just stretching the edges is not enough because due to the crown, there has to be a little stretching take place on the inner portion of the part as well.



Below is another view of the distortion caused by the stretching:



Remember -- this is just a learning exercise -- not a step-by-step of how to do this. (I'm learning as I go).

The distortion on the edge is easily dealt with by simply "encouraging" the part to go in the desired direction.



The edge comes back in-line when the reverse is put in place with a slapper. Doing it this way is all fine and dandy if the running board end is to be flat. But simply bending the metal in this manner does not stretch the metal where it needs it, in order to continue the side-to-side curve. Notice how the metal on the left is resisting -- that's because the edge has not been stretched yet. If we were trying to make this piece "properly" -- both edges should have been stretched before trying to encourage the reverse. Actually it would probably be best to alternate by stretching a little and then work on the reverse and back to stretching, rather than doing all the stretching at one time as I did.



Just for the heck of it, I did the stretching on the other side with a cross-pein hammer. The pein on this hammer is sharper than it should be, but even the deep indentations it makes are fairly easy to even out with a little planishing. The planishing can be done with a planishing hammer or on the wheeling machine. Even most of the stretching can be done on a wheeling machine, but probably the best way to do that would be prior to bending the edges over.



Well -- I've probably lost some of you, simply because some of this is a little difficult to describe. If anyone has any questions -- or better yet some suggestions -- let's hear them. As I've said before -- this is just a exercise (not a lesson) so input is more than welcome. Suggestions of better or alternate techniques are especially welcome. If there is something that you don't understand, there are probably a hundred or more others wondering the same thing -- so ASK THE QUESTIONS!!!!!

This is a discussion -- not a lecture... so let's hear from you guys.

BTW -- I did some work on this piece with the air hammer yesterday, trying to get the side-to-side curve to follow through to the end... and the results are pretty ugly. I'll post them later. But in doing so and after thinking about it at 3:00am this morning when the neighbor's dog woke me up... I think I have figured out the error of my ways. I think a light came on -- but we will see (smile) More later...

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

John B Wrote: I feel a bit awkward chiming in here because i have no e-wheel experience, all my shaping experience is with hammers. After covering that point i will say regardless if i was using a e-wheel or hammer i would have pushed (reformed) the flanges outward on that fender and flattened (reformed) the panel as much as possible before stretching it lengthwise. You wouldn't be able to do this at the very front (because it's a ball) but that's OK because the reverse curve isn't there. Terry got into trouble because the stretching has to go deeper into the center of the fender than just the flanges. For lack of a better way to describe it the tracking pattern or the hit pattern is 2 triangles with the tips nearly meeting in the center of the fender. How much you move that pattern up and down the fender and how much more you stretch the outside edges versus 1/3 of the way in from the edge is strictly up to the shaper's experience or lack of it <grin>. Quite some time ago a fellow metalshaper posted "Reverse curves are easy" after taking a shaping class. If you just making art work, in other words shaping things to please the eye it's not too bad. It's making a panel where the reverse curve that has to fit a buck and the radius of the reverse curve is constantly changing that gives even experienced shapers problems. Make a cowl (gas tank) top for a 28/29 model A with no problems and i will be impressed. Also because of Scott Knight's class i never start a job with forming (rubber) tooling but that's just the way i learned the basics of shaping, it's pretty hard to argue with a shaper's actual results or finished job even if he didn't do it exactly as you would. ~ John Buchtenkirch

+++++++++++++++++++++++++++++++++++++++++++++++++++++++

John,

Thanks... What you are saying, pretty much confirms what I was hoping I had figured out. Your description of the tracking pattern is exactly what I had come to think was the solution.

Going into it -- I wasn't thinking of that area needing to stretch so much in relation to the center. Its pretty easy to work yourself into a corner if you aren't seeing what really needs to happen. I was first thinking that the center has to stretch... but it finally dawned on me that its just the opposite.

The shape I'm trying to made is similar to half of a tube. I finally pictured it as a tube. If you bend a tube, it will krinkle on the inside of the bend... so either the inside of the bend would have to shrink -- or it would have to stretch more and more from the center outward. Duh!

Okay -- now I'm going to give it another try.

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
unklian

Posts: 517
Status: Offline

"The shape I'm trying to made is similar to half of a tube. I finally pictured it as a tube. If you bend a tube, it will krinkle on the inside of the bend... so either the inside of the bend would have to shrink -- or it would have to stretch more and more from the center outward."

It's obvious;when you know the answer.

Breaking down into it's basic elements,
and exagerating the effect your trying to achive,
can help visualize what needs to happen.
Ask,what does the metal want to do ?
If it wants to tear,it needs to be stretched.If it wants to buckle,then it needs to shrink.
But if you can't shrink in one place,you might have to stretch somewhere else.

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

As I said earlier -- I made a pretty big mess when I went in entirely the wrong direction with the air hammer...



In a fit of temporary insanity and a lapse of good common sense... I stretched the center, thinking it would continue the curvature I wanted. DUH! WRONG! It is real clear from the photo above, that this piece is reacting exactly as a tube would if you tried to bend it in this fashion. If only I had visualized a tube being bent before I went berserk with the air hammer! Oh well -- like I said from the beginning... this was meant to be a "learning exercise" -- and I'm certainly learning from my mistake (smile).

It is very obvious that stretching was exactly the opposite way to go -- so I set up a stack of bags to hold the part so I could pound the stretched areas down.



I was able to actually shrink most of the stretched areas back down by simply pounding them down on the sand bags. The stretched areas were trapped on all sides by the surrounding metal, which kept the metal from going anythere but back into itself when hit.



The areas on the end that were not trapped on all sides, formed puckers, like this which were knocked down, further shrinking the wrongfully stretched metal. This is the same process as tuck shrinking. If you are not very familiar with tuck shrinking... here are a few links that may help you:

http://allshops.org/cgi-b[...]5631440

http://allshops.org/cgi-b[...]9223184

http://allshops.org/cgi-b[...]2229151



Even more shrinking was necessary -- so larger puckers were made with a tucking tool and knocked down.



Now things are looking up... Most of the wrongfully stretched area has been corrected and the outer edges have been stretched more as they should have been in the first place. Its just like John B said. As I stretched the outer edges of the piece I kept John's "triangle" visual in my head.



The result may not be perfect -- but at least the side-to-side contour is fairly consistent now, and the exercise has taught me a valuabe lesson about compound reverse curves.



If you want to see an enlarged version of any of these photos, they are all housed in an album -- here's the link:
http://allshops.org/cgi-b[...]5013509

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
This post has been modified
typhoon

Posts: 5
Status: Offline

Thank you all so much for the information!, You've given me a lot of ideas to play with. I think the first thing is going to have to be starting off with 2024 0, trying something new with hard material was a bad idea, I'll try some of the trick you've pointed out and see what happens!, that brings up another question.... are there many people here with information about negative effects of flame annealing aircraft materials to be used for structural purposes? I suppose ideally using a furnace would be the way to go, but it's also very expensive. If I burn off carbon from a piece of 2024 T3, it should give me a half hour or so before the material begins to age harden, is there any negative effect from adding carbon to the surface of the aluminum? and will the item in question be sure to return to the T3 or T4 state?

Ian Slater

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

While looking through the album of when I visited Dutch Comstock's shop, I came across this piece that I made while goofing around with Dutch's equipment.







For a larger view of the above pic:
http://allshops.org/commu[...]827.jpg

We cut it in half so I could practise TIG welding on it.

This reverse maintains the side-to-side curvature just like the one I was trying to do on the fender-like shape.

But there is a big difference in how they are shaped.

This one is stretched in the center... and there is a logical reason. Would anyone care to explain what that reason is?

I know John B knows the answer (smile).

This situation is similar to a cowl.

There is a major difference in the "cowl" and the "tube"... what is it?

Terry C


Reply
Edit
Highlight
MoreInfo
Recommend
youngwally2002

Posts: 8
Status: Offline

As I see it, the half of the completed panel that has the reverse shape in it, could be formed by stretching both ends only, (The welded edge, and its opposite edge) to get the desired shape. Any stretching in the center portion would only negate the edge stretching. What have I missed?

Reply
Edit
Highlight
MoreInfo
Recommend
392hemi

Posts: 231
Status: Offline

John wrote:
As I see it, the half of the completed panel that has the reverse shape in it, could be formed by stretching both ends only, (The welded edge, and its opposite edge) to get the desired shape. Any stretching in the center portion would only negate the edge stretching. What have I missed?
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

John,

I don't think you missed anything -- I was probably just not clear in what I'm trying to say.

I think where I used the word "center" -- that was a poor choice of words. My apologies. Hmmmm... let's see how does one say that clearly?

Okay...I didn't mean center of the piece -- I meant at the centerline where I stretched the piece, which would be on the centerline on the "A" end below:



Yes, both ends "A" and "C" were shaped by stretching. Although "B" is not actually centered, it is between where the compound curve "C" ends and the reverse begins. So if that is the area you are calling the "center" then yes, any stretching in that area would certainly negate what had been accomplished. What I think you are calling the center would be sort of a neutral area.

To further explain how this piece was shaped... the area marked "D" is where the shrinking dies were used and the rest of the side up to and including the area marked "E" was just bent over (no shrinking or stretching). The reverse (the area between "A" and "B")was done by stretching with linear stretching dies.

I hope that explanation makes it more clear.

The point I'm trying to make is that unlike the tube shaped end of the fender-like piece which could not be stretched down the centerline (as John B pointd out)-- that this piece is, in fact, stretched along the centerline and adjacent area.

One method stretches the edges and not on the centerline -- and the other method stretches on the centerline, but not on the edges.

I'm on my way out to the shop now... I'll see if I can do another sample that that makes more sense (smile).

Thanks to both Johns, maybe we can get this explained so everyone can understand.

Terry C

Reply
Edit
Highlight
MoreInfo
Recommend
66tiger

Posts: 84
Status: Offline

I think John means to build it in two pieces where the stretching would be over the two yellow striped ares in the attached photo.

-- Arden



Attached Files:

  • 9960101277827.jpg (8 kb)

  • Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    youngwally2002

    Posts: 8
    Status: Offline

    Thanks Arden, you have it right. John G.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    392hemi

    Posts: 231
    Status: Offline

    Arden said:
    I think John means to build it in two pieces where the stretching would be over the two yellow striped ares in the attached photo. -- Arden

    John G. Said:
    Thanks Arden, you have it right. John G.

    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


    But the part was very easily made in one piece. It is a rather small part and the Pullmax and shrinking dies and air hammer made short work of it. It was cut in half after it was finished. Dutch cut it in half just to make me weld it back up for welding practice. Maybe John B was right -- maybe it was a poor example.

    I think I'll back up and finish some examples I began today before I go any father with what I'm trying to get across, since it looks like its just getting more confusing.

    Thanks Arden and John G. Hey Arden -- maybe you could translate for me too (smile)

    Terry C

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    392hemi

    Posts: 231
    Status: Offline

    Bill Reid Wrote:
    hope you guys don't stop..I'm learning
    =========================================================

    Bill,

    We aren't stopping... just going to re-load (smile). I'm working on several illustrations so I can properly explain the points I'm trying to make. This is just a little more difficult to visualize than some of the more simple techniques. So before I open my mouth again and risk more confusion, I'd like to get some graphics done, that will explain better than I can in words.

    For instance, I'm making some tunnels out of sheet wax and using the "thumb Marchant stretcher" to show two completely different approaches to obtaining similar reverses.


    I'll also follow up with aluminum. Just give me a few days to finish up... I'm working this in between "putting out fires" daily so it will take a while.

    Terry C

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Ian, I would stay away from trying to flame anneal 2024. You'll most likely end up with cracking immediately. I've built gun turret fairings for a B-29 bomber where the radius changes with the belly of the plane. I use the softest 2024 I can get. If you want it hard after you can have it done professionally where they do a soak cycle in a large oven. On reverse curves the best way I've found to figure out you baseline pattern is to find the center of your part along the length ( like a fish skeleton with the backbone being your centerline ). Then make the ribs ( I use half inch strips of paper )and tape them to the backbone at regular intervals and cut off the ends a the edge of the part. Once you've taped the entire part, remove it and you can lay it flat for a pattern ( keeping the back bone the in its 2 dimensional curve ). You will also see what areas need the most stretching by how much the ribs move at the ends when you lay it flat. Don,t wheel your baseline just bend it to the curve of your part. On the wheel, I use a lot of force at the begining with a sharp radius wheel, then slowly back off the pressure and go to a larger radius wheel that fits the final part I'm trying to make. As you back off pressure your part will smooth out then you can smooth the center. Good luck. You can see some of my work at http://home.comcast.net/~seattledl/B_29_Before_After.html
    the last pictures on the page as you scroll down. You can see me and my shop at http://home.comcast.net/~seattledl/B_29_Who_we_are.html
    Also the last picture on the page, black sweater guy.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    unklian

    Posts: 517
    Status: Offline

    Welcome Brian.
    Here are those links in clickable form.
    http://home.comcast.net/~[...]r.html

    http://home.comcast.net/~[...]e.html

    Looks like some serious metalworking.

    Right below the box where you type your text,it says "Add a Poll,A Link,or an Image"
    If you click on the blue "A Link",then paste the addy into the box,it adds the necessary code to make the links clickable.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    392hemi

    Posts: 231
    Status: Offline

    Brian wrote:
    On reverse curves the best way I've found to figure out your baseline pattern is to find the center of your part along the length (like a fish skeleton with the backbone being your centerline).
    ==============================================================================

    Brian -- that is an excellent analogy. Just having that visual really helps me analyze a reverse curve.

    I'm also interested in your comment about beginning with a sharp radius lower wheel and backing off the pressure as you progress to a radius that fits the part. Do you do that on most projects -- or just reverse curves?

    I haven't made any more progress on my reverse curves yet -- I got sidetracked with the tear drop thread which I'll post a few pics later.

    It has been quite a while since I've read John Glover's book The English Wheel, so I drug it out today and read the section on reverse shapes. If I had read that section before I started the reverse exercise I wouldn't have tripped up on stretching in the wrong area. Oh well -- when I learn through a mistake -- I think I remember it longer (smile).

    Brian's fish skeleton analogy along with an illustration and explanation in John Glover's book have both helped me get a better visual of the reverse curve. BTW John's book is available from one of our sponsors -- Metalcraft Tools: http://metalcrafttools.co[...]os.html

    So Brian -- welcome to the group and thanks for the analogy. I'm looking forward to seeing more photos of your shop and your work (and more analogies).

    Terry C

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    392hemi

    Posts: 231
    Status: Offline

    Group,

    Since many of the comments on this thread have been on the Yahoo group (which does not automatically post here on the MSA web site forum) -- there are some holes in this thread. I try to move pertinent comments here to the MSA web site forum -- but I just don't have the time to catch them all. I'm playing catch-up today moving over the ones below and trying to fill in the blanks. A few days ago I had most of these setup to move over and some additional comments of my own to post, but I lost them all when I had to re-boot. So I'm backtracking this morning, just trying to fill in the blanks then I'll probably begin a fresh new "reverse curve" thread next week.

    It would help me a lot and help keep these "keeper" threads in order, if you guys would post them from the MSA web site instead of Yahoo. That way they will be in the proper order and preserved with the images on the MSA site -- and they will then get forwarded to the Yahoo list. The MSA forum is located here: http://allshops.org/cgi-b[...]ssages&

    ================================================================
    Sat, 16 Oct 2004 12:28:52 Danny Pascoe Wrote:

    John whats the chance of you doing a basic part like Terry did but you using the hammer and showing a sequence of how you might interpet the same part using a hammer. I know my self and may others in the group do not have any hammer experance. Mine is limited to english wheel, hammer & dolly, and some planishing hammer. Could you please just as a comparison for us non power hammer people thanks -- Dan Pascoe

    ===========================================================================================

    Mon, 18 Oct 2004 21:31:John Buchtenkirch wrote:
    Here are the hammer patterns to make a reverse curve in a long panel such as a cowl or a good portion of a fender. The things to note are the hits are narrow ovals (directional dies) so they are stretching the fender lengthwise and they are overlapping much more at the outer edges of the panel. The drawing is just to transfer the required stretching/shaping idea from one craftsmen to another, how much more to stretch the outside edges will get as compared to 1/3 of the way in will have to be determined by the panel being made. I have also given my best shot (i have zero E-wheel experience) at drawing tracking paths for a E-wheel. In hindsight i don't think the tracks should have touched in the center of the fender but i would like to hear what the experienced wheeling guys have to say.

    In Terry's second shaping sample he shows a smaller part which i feel is a poor example to use. Why --- well i have put duck tails on several motorcycle fenders by just flaring out the end of the fender with directional dies. While i did end up with a reverse curve it's a technique that would only work near the edge of a panel and wouldn't be practical for making that model A cowl or a gentle reverse 1 foot from the end of a fender. I'm hopeful that i have helped more people than i have confused with this, for me it's hard to put some of these ideas into writing. ~ John Buchtenkirch





    Tue, 19 Oct 2004 16:05:42 David Szkilnyk wrote:
    Hang on, sorry donít have a Model A cowl handy, but isnít a model a cowl nothing more than a tapier. I would figure that you would start with \ / and create your curved edges and properly stretch the sides so that the cowl curves back up (taken advantage curving nature of stretching and shrinking). I reckon it would take a bit of juggle to to get it right a paper pattern would help here. -- Dave



    ==================================================================


    Mon, 18 Oct 2004 23:43:51 Terry C wrote:

    John,

    I think I just moved over to the confused camp (smile). From your first description I thought I understood what you were saying, but now after seeing the drawing, I'm not real sure.

    And yes sometimes these things are difficult to put into writing, obviously I'm not doing so well at it on this thread!!

    What I am trying to explain is two completely different applications with two completely different approaches, but I don't seem to have gotten off to a very good start... so let's forget that for the moment, since its getting confusing... and lets back up and discuss the cowl only.

    Can you clarify where your pattern will be on the Model A cowl, please? I don't understand exactly what it is that would not be practical for a Model A cowl. If you can explain that a little more I would appreciate it.

    I assume the early cowl such as the one below is what you have in mind. -- Thanks, Terry


    ==================================================================================

    At 10:09 AM 10/19/2004, John B wrote: The lines i drew at the top and bottom of the page are the front & back of the cowl or the inside & outside edges of your fender when looking at it from the top. What i should have mentioned is that you take that pattern and keep shifting it over an inch till you cover the whole panel, to me that was just a given but i shouldn't have assumed so for everyone else. You had the right idea stretching the flanges on that fender, you just needed to stretch nearer to the center with a fade away stretch which is the best i can think of to describe it at the moment. The main thing to remember is you stretch opposite edges of a panel and the stretch fades off to nothing in the center to get a reverse curve. Are we OK now ??? ~ John Buchtenkirch


    TC: I thought you meant to shift the pattern over -- but I wasn't sure.

    Now I see what you are saying... but I don't think you see what I'm saying (smile)... but I'm not going to attempt to clarify my point just yet, since it would probably be more confusing at this point. without a proper illustration.
    ===========================================================================

    Wed, 20 Oct 2004 15:26:22 John B wrote:

    Terry, I would not be able to make that cowl top by stretching the sides (narrow ends) of the panel but being i have never done a reverse curve on a panel that long i will wait for your illustration to see what you mean. My objection to your second smaller sample piece was because many times you can get away with things near the edge of a panel (like the duck tails and the resulting reverse on the Harley fenders) that just wouldn't work thru out a larger panel. The problem was with the reverse on your sample fender, i probably shouldn't have mentioned the cowl and you shouldn't have come up with that other panel. There has been very little response to this thread so i'm thinking we have managed to confuse a bunch of people ? What did John Glover think of my guess on the E-wheel tracking pattern ?? ~ John Buchtenkirch

    ========================================================================================

    At 02:26 PM 10/20/2004, John B wrote: My objection to your second smaller sample piece was because many times you can get away with things near the edge of a panel (like the duck tails and the resulting reverse on the Harley fenders) that just wouldn't work thru out a larger panel.

    TC: That's true... so I'll start all over with a fresh beginning to see if we can get off to a better (clearer) start.

    JB: The problem was with the reverse on your sample fender, i probably shouldn't have mentioned the cowl and you shouldn't have come up with that other panel.

    TC: Actually, the cowl is a very good example to use. It may not be a good example for your particular point -- but certainly a good example to discuss reverse curves.

    JB: There has been very little response to this thread so i'm thinking we have managed to confuse a bunch of people ?

    TC: Probably so... but the lack of response is probably more so because no one is willing to step out in public and risk being wrong in front of thousands of peers. It is much easier and less risky to discuss lighter subjects. I'm still learning too -- so I don't mind being wrong in public as long as we figure out what is right.

    JB: What did John Glover think of my guess on the E-wheel tracking pattern ?? ~ John Buchtenkirch

    TC: We didn't get that far -- maybe he can chime in here.

    ====================================================================
    Wed, 20 Oct 2004 Terry C wrote:
    Dave,
    (Correct me if I'm wrong), but I think what you are saying is that part of the shape of a cowl is the result of the way the blank is cut. In other words, the front of the cowl will not be cut as wide as the rear -- so that when you pull the sides down, the cowl automatically tapers down to the front. So without any shrinking or stretching, the cowl is already higher at the rear. At that point you have a curved axis from one side of the car to the other -- but a straight axis front-to-back. Then from there all that is needed is a little stretching, to obtain the gentle reverse. (Am I even close?)

    Where we are getting this thread confusing is in the use of terms such as "center" (which I incorrectly used) and edges (without a drawing or description which tells us which edge or which center we are talking about.)

    It sounds like you and John B are talking about stretching in different areas... John B is stretching along the front and rear of the cowl -- and you mentioned "properly stretch the sides"... Well oddly enough -- either way will get a similar result. It is six of one and half a dozen of another.

    But as I said -- I'm not going to attempt any further explanation without good illustrations to make it clear.


    In a conversation with John Glover, he summed up our problem with this thread when he said "I think the confusion could be lessened if ALL explanations and drawings were to be illustrated with reference points, and very clear instructions. For instance, you could be saying. "Stretch the center", and some people would be confused as to what area is the center. Now if you were to say," Stretch the edge in area 'B" for two, or three inches either side of the center line, then there would be no confusion."

    He makes a very good point (especially on this particular topic), so from here on out on this thread, I will have a referenced illustration before I attempt to explain any more. I'll be working on the illustrations over the next few days.

    Terry C
    ===============================================================

    10-20-2004 Kent White Wrote:

    Terry,

    This is a pair of reverses that I made for an old airplane, on our Air Shaping Hammer using 2024 .040 aluminum, which we do not weld for aircraft applications, but it will be riveted.

    On our new Air Hammer Operations DVD I show the pattern process, blanking and marking the flat stock, and working an even more severe reverse for the Hughes H1 replica, step by bloody step. I use a flexible pattern which goes into place simulating the reverse exactly, and then the same pattern lays flat on the sheet for blanking.
    While it may not be the absolute koolest pattern method on the mud ball, it works for me.
    Kent




    ====================================================================

    Thu, 21 Oct 2004 22:14:51 John Buchtenkirch wrote:
    The wing farings are either Hellcat or Bearcat. I have another photo of a table at Grumman loaded with reverse curve parts, hopefully i will find it for a future post. George Burkhardt (was the senior hammer man till he retired & Grumman employee #17) drew me the hammer pattern i posted a few days ago. He told me you had to figure out where your neutral line was (more or less the midpoint where there is no stretching) and the ratio of the stretch at the edges compared to the neutral line determined how fast the panel turned.






    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    typhoon

    Posts: 5
    Status: Offline

    Brian,

    I suppose if 2024 cracks immediately, there would be no way to keep it in a airworthy state by using flame annealing techniques. Looks like I'll have to find someone with a large oven if I want to bring the material back to T4 after forming. Very nice work on the tail section, it's amazing to see some of the forming being done by the people on this site, it'll keep me motivated while I curse at my reverse curves! Thank you for the tips as well, it seems as though there are more ways to skin a cat that I thought.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Ian G,Terry, Ian S
    Thanks for the warm welcome and Info. This Forum format is new territory for me so bear with my confusion with threads and links and all. I am fairly new to 2024 but have done some research and have had much help from Kent White and former Boeing Engineers and workers, the guys that actually built the planes. Right now I building some Alloy flairs for a Corvette Race car and I'll try and figure out the steps of downloading pictures of my progress. So far today (Sunday)I have about 4 hours into the process. I've made an outside edge boundary and made my fish bones ribs for curved shape for the rear tire and the outside boundary for the front tire. I'll take some pics tomorrow (oops already here)and see how it goes. I have until Wednesday to get all 4 fenders done for a race. Wish me luck.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    392hemi

    Posts: 231
    Status: Offline

    Group,

    I haven't forgotten about the compound curve thread. Its just that every time I'm about work on it -- I get waylaid into doing something else.

    I'm going to try again tomorrow to get something done on it.

    Terry C

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Well group,
    I posted the pictures today of the fender flares I did for a very good customer of mine. Hope this helps with the ideas I tried to convey before. I posted the pictures to an album in the metalshaping catagory. Terry will help me figure out how to move them over to this forum later (right Terry ?).

    Brian Nordby http://allshops.org/cgi-b[...]5782107

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    unklian

    Posts: 517
    Status: Offline

    You can add pics to your posts in these forums by right clicking on the pic you want,
    select "properties",left click and drag to highlight the address,
    right click and select "copy";
    now go to your message,and where it says "Add a Poll,a link,or and Image",click on "Image",
    then right click on the pop up window and select "paste".Then click "OK".
    It's easier than it sounds,much easier if you have 2 windows open at the same time.

    The pics you posted are pretty big,so I just copied one of the thumbnails.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Thanks Ian,


    Brian

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    This post has been modified
    unklian

    Posts: 517
    Status: Offline

    Cool You can also use it to add more than one pic in the same post,



    or to position text between pictures.



    Which can be helpful.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Oops,
    Sort of screwed things up with pics. Posted smaller pics to album and deleted larger pics not realizing it would mess up the thread.
    So did I leave you all speechless? I was hoping to get some feedback or other methods for doing this stuff. Is there a better way to do it?

    Hope I haven't offended anyone.

    Brian

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    ghiafab

    Posts: 95
    Status: Offline

    Hi Brian,

    Interesting way of doing flares. I have three different ways to flare a car so far. The first one is the traditional method of shaping metal to a buck:

    http://allshops.org/cgi-b[...]1521468

    the second is stretching the existing sheet metal (hard to do on a Vette), the third is to make a wheel opening form from 3/8" square stock, roll flat panels over my leg, weld to the the car, then stretch some shape into them:

    http://allshops.org/cgi-b[...]8836765

    I developed the last two methods fo do-it-yourselfers. Thanks for posting.

    John www.ghiaspecialties.com

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Thanks John,
    I just posted another Album this time in the aircraft catagory.

    This job was nightmare getting things to fit right. I must have made three or four different pattern and 2 or 3 side pieces that were scrap all out of very pricy 2024 (no welding). Albums at http://allshops.org/cgi-b[...]7406551
    The sides are really subtle reverse curves, the ball is slightly larger than the sides of the plane. The ball actually pivots inside the side pieces or what we referred to as the lips. They (the lips ) have a bead on edge with a thick piece of felt to keep the pieces from sliding together.
    Thanks for letting me share this stuff.
    Brian

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Group,
    Also added painted flare photos to the flare Album.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    ghiafab

    Posts: 95
    Status: Offline

    Hi Brian,

    Beautiful work! Did you use an epoxy based product to bond the flares to the body? You really have your reverses down. The airplane is amazing.

    John www.ghiaspecialties.com

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    392hemi

    Posts: 231
    Status: Offline

    Brian,

    Great job on the flares and the Ball Turret Section! I have to ask... since the car is fiberglass -- I'm curious why the customer didn't want the flares in fibergalss also?

    Thanks for posting the photos.

    Terry

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    John,
    The flares were attached with pop rivits. I would have doubled the number that are in there due to the amount of abuse these cars get, but thats what happens when you have someone else finish the job. The car was at the race track before the flares showed up from the painters. Thanks for the compliment, John.

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Terry,
    Mostly time was the factor, the car had to leave for Monteray by Wednesday afternoon and I started on Sunday. With fiberglass the whole body would have to go to the paint shop. Hard to carry a freshly painted body on to the plane. Also, the Owner wants me to make a set of Alloy Flares the exact shape of the original fenders so he can switch back and forth between slicks and regular race tires. Obviously money is not an issue. I would like to make molds from the flares and put fiberglass copies up for sale. But I'll let the customer decide.

    Brian

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    lusc10

    Posts: 1
    Status: Offline

    A beautiful job Brian. What did you have to start with. Did you make from plans or from origanal parts? Is that a bead in the end section that wraps around guns. I bet that was no fun after make those parts. Whos airplane will it be on? Kermits? Thanks for the great aircraft stuff. Looking forward to more albums.

    Randy Tait

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    metlman

    Posts: 14
    Status: Offline

    Randy,
    I started with the original parts that had been smashed and torn by the forklift. I straightened them as much as I could. Then after much deliberation it was decided that we needed to build it to fit the inner trunion ring. I'll post more pics later (pre digital camera) of the construction of that. It was a combination of working from the old parts and the blueprints. Yes thats a double bead about 3/4 inch wide which we had to machine just for this job. I had it made to fit my Pexto bead roller and set up a jig so I could run a straight edge. The plane belongs to the Museum of Flight here in Seattle. You can see more pictures of the whole plane at http://home.comcast.net/~seattledl/
    I jokingly referred to it as Double D in the pics.

    Brian

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    392hemi

    Posts: 231
    Status: Offline

    Well -- I didn't do this reverse curve with a wheeling machine -- I used an air hammer (simply because I'm better with an air hammer than I am a wheeling machine). This began as a round eleven inch disk.

    Once I get a better understanding of what this takes -- then I'll go back and try the same thing with a wheeling machine.





    All I was attempting to do was to make a saddle-like shape -- or a taco shell with curled up ends. While I was playing around, I made one end larger -- and it began to look a lot like a turkey. So now I know how to make a basic turkey shape (smile).

    Next, I want to make a wooden buck with exact curves that I will have to match. That should prove to be interesting.

    Terry C

    Reply
    Edit
    Highlight
    MoreInfo
    Recommend
    This post has been modified


    Delete Checked
    Search
    Forum Options


    Members Currently logged in: 0 -- Members and Guests Online not logged in: 13
    # of members viewing this thread: 0
    # of guests viewing this thread: 1

    Members viewing this thread:

    All Members Online:
    See where they are...