header & manifold tubing flares

[petertha]

I need to nail down the intake & exhaust tubes on my radial. Specifically the upset or flare. The flare has to contact the end of a counter bore hole in the head, then a threaded nut seats it. One tube for a short exhaust radius exhaust pipe. The other for induction header. Its essentially this same task, different engine
http://philsradial.blogspot.com/2013/08/tube-buckling.html

I bought some 5/16" brake line & also some 3003-O aluminum Versatube just to get a feel for the materials & tube bending. I probably should have just sprung for a commercial flaring tool, but visually it seems like the die blocks were quite thick, therefore that straight line segment would influence the shape of the header bedore it could start the bend.

On my first die I made a backside counter bore to accommodate the curved header. But the 0.100" grip band didn't have enough surface area to grip the tubing, it slid out prematurely. I noticed some commercial blocks had thread like serrations so I made another block 0.25" thick, tapped it 5/16-24, then drilled it one drill size under 5/16 tubing OD so the threads had just a bit of bite on the tube. It gripped much better, didn't move. But I was quite surprised how much clamping pressure it took in the vise to make the flare just using a dummy 45-deg form tool against the die chamfer in the brake line. The versatube was easier but I havent decided if using them on the exhaust is a good idea.

So I tried just peeing over the edge in small increments like Phil did, first with a small punch which was ugly. Then a crudely shaped roundy tool that showed more promise. The brake tubing material is quite malleable. I need to improve the technique but its heading in the right direction.

Do you guys have any tricks to pass on? Any experience using commercial flaring tools for model applications?

 

[petertha]

This is slightly de-grunged an exhaust tube from an OS 4 stroke. I've examined a few similar engines, most have an angle flare similar to what I'm attempting but this one is darn near 90-deg. Hard to discern what the material is. Some seemed silver plated over possibly copper based brake line. Another almost seemed like stainless which I may try but have heard horror stories.

 

[petertha]

Another reason I'd like to stay with an exhaust tubing alloy which is 'solderable' (or maybe TIG-able if I ever get one) is a future step would be a collector ring. Not quite sure what to make of the very bright metal looking ends that similarly go into the heads. They almost look like aluminum with a turned boss upset vs a flare. But then what? High temp adhesive to the rest of the exhaust assembly? I thought about silver soldering a ring onto the end of a formed tube but concerned it might fail under exhaust temp. Any thoughts on that?

https://www.ch-ignitions.com/fg-84r3-fg-90r3-radial-exhaust-ring.html

 

[WOB]

I built a Hodgson 9 radial and used 5/16" 3003 alum tubing for both intake runners and exhaust stubs. I used 1/4" inverted flare nuts by drilling them out to 5/16" and machining the flared end to 90 deg. to fit a flat bottom, threaded( bottom tap) recess in the heads. The tubing flared easily by cutting roughly to length and chucking in the 3-jaw. In the tail stock, I used a simple radiused flaring punch turned from drill rod on the lathe. It had a tubing bore dia. spigot that flared with a small radius to flat ( 90 deg) by the time it reached the 5/16" tubing OD. It was easy to flare the tube with a little lube and no danger of splitting the flare even when flared to about 3/8" dia. This gave me enough material to face off the flare and trim the OD to match the threaded holes in the heads. No leaks and easy assembly, worked perfectly. My bending jig used the same threaded recess as in the heads. All bends were 1/2" radius and were easily accomplished by filling the tubes with Cerrobend before bending. The bends were perfect. Exhaust temp on a small engine like the Hodgson is not high enough( lots of turbulent air flow around the heads) to cause a problem. I had no intention of using a collector ring so I don't know if 3003 alum is weldable or not. I do know that it is wonderful material to cold form and machine.

RWO

 

[DJP]

A nickel copper blend is preferred for auto repair of brake lines so it may suit this purpose as well. A thought for your consideration.

 

[petertha]

DJP yes this brake line is a nickel-copper alloy. Cant recall off hand if its Cunifer or Cupronickel or Cupro or if they are tradenames of essentially same alloy.


WOB, Thats encouraging you made a 90-deg kind of flange. That would be my preference too.

- so when you made the flare in the lathe with a form tool, do you mean it had the male impression of the flange and it plunged in axially? Or you mean a hand tool that kind of worked over the edge of the female die maybe same principle as forming nose cones bowl shapes over a mandrel block form

- I tried chucking the tube but it was a fine line of crushing trying to grip it under the axial pressure but it was also a short test section maybe 1" long. Do you think a 5C collet would be better

- any tricks to properly 'uncoil' the Versatube to something approaching straight from the pizza box it was delivered in? I figured they must have used a roller but I don't really want to build a machine I'll probably never use again.

- what kind of gasket did you use between tube end & exhaust flange & same question for seal on induction flange?

- did you have to splice your induction tube with a tasteful looking coupler or you managed to bend in one shot?

- I bought a Ridgid tube bender which makes nice curves, no filler required thus far although it is a fixed minimum radius. I have seen guys use those plier looking tubing tools. I suspect they might make tighter radii but I wasn't aware they existed

 

[Cogsy]

I haven't used them for modelling but have made brake lines for trailers using the commercial flaring tools. Where I live we are required to double flare the ends of brake lines (I assume for strength of join) and this makes it far easier to grip the pipe as the first part of the flaring process quickly forms a shape that will not allow itself to press through the clamp. The resulting flare is virtually identical in appearance to a single flare but leaves a thicker 'face' on the flare which may or may not be desirable in your application.

 

[Lloyd-ss]

I take it that you don't want to use plain copper tubing? It is so easy to anneal by heating cherry red and then letting it air cool or plunging it in water. Comes out dead soft. And you can re-anneal it half way thru a bend if the copper starts to get work hardened. Some times even just the process of using a tubing cutter on copper can slightly harden it. But the annealing actually gets it soft enough that you can just lightly hammer the taper of a socket into the end of the tube to expand it. Same result as with an exhaust pipe expanding mandrel, just ona smaller scale.
I've used the commercial flaring tools like in your first post. The cheap ones have a cone that just by brute force pushes straight into the end of the tube to force the tube into a flare. The expensive ones have an undersized cone that rotates in a circular path around the inside of the end of the tube and gradually works the flare outward. That style doesn't exert a force that tries to push the tubing out of the serrated clamp that holds the tube.

 

[petertha]

Thanks Lloyd. Turns out plain copper is almost the same price as the high copper content dedicated brake line stuff so figured I'd sample it first & side step the annealing business. In both cases its pretty heavy stuff for the induction segments 5 x maybe 6 lineal inches although I didn't have firm aspirations of flying the radial. Seems like a big enough task getting the bugger built.

I was aware of the double flare but didn't really look into it too hard. I thought it was more of a specialized hydraulic seal but it does make a nice surface. Someone told me I might be able to rent the tool locally. That would be a good way to see how the profile looks in real life. It does look a bit more robust than the single lip trumpet. I need to check the final OD to see if I can accommodate it.

One difference between brake lines & model engines is the straight section aft of the flare end dictated by the thickness of the die block. It kind of extends the whole header geometry. I might be able to work something out but it does deviate from the plans which are quite a bit more compact curves. I thought I would make the bends & then do the flare. But maybe I'm thinking of this wrong - make the flare on straight & then do the bending. I just assumed the flare would interfere with the die shoe/block but I need to confirm that.

 

[petertha]

The expensive ones have an undersized cone that rotates in a circular path around the inside of the end of the tube and gradually works the flare outward. That style doesn't exert a force that tries to push the tubing out of the serrated clamp that holds the tube.

Interesting. Happen to have a link?

 

[Lloyd-ss]

Interesting. Happen to have a link?

The Ridgid 345 and 458 are nice professional tools. Something a tool rental place would have, but again, the too-thick grip-block for your application.


Seriously though, copper annealing is super easy, but you do have to polish the parts afterward because of the heat discoloration.

 

[WOB]

WOB, Thats encouraging you made a 90-deg kind of flange. That would be my preference too.

- so when you made the flare in the lathe with a form tool, do you mean it had the male impression of the flange and it plunged in axially? Or you mean a hand tool that kind of worked over the edge of the female die maybe same principle as forming nose cones bowl shapes over a mandrel block form.

Yes, I plunged the form tool axially into the tube end. It took very little pressure to make the 90deg flare.

- I tried chucking the tube but it was a fine line of crushing trying to grip it under the axial pressure but it was also a short test section maybe 1" long. Do you think a 5C collet would be better.

Now that I think about it( it has been several years) I believe I did hold the tube in a 5C collet.

​

- any tricks to properly 'uncoil' the Versatube to something approaching straight from the pizza box it was delivered in? I figured they must have used a roller but I don't really want to build a machine I'll probably never use again.

I just used my hands to straighten the tube. It is fairly easy to get it straight enough to enter the collet and is even better after closing the collet.


- what kind of gasket did you use between tube end & exhaust flange & same question for seal on induction flange?

No gasket needed. The soft tube easily crushes to fit the head ( 6061 alum) recess gas tight.

- did you have to splice your induction tube with a tasteful looking coupler or you managed to bend in one shot?

I made all tubes in one piece.

- I bought a Ridgid tube bender which makes nice curves, no filler required thus far although it is a fixed minimum radius. I have seen guys use those plier looking tubing tools. I suspect they might make tighter radii but I wasn't aware they existed

CHECK MY COMMENTS IN THE ABOVE QUOTE BOX.
There was no commercial bender that could make 1" dia. bends with a 5/16" tube, so I made a simple steel block with the head recess machined into it with a half-round section right next to it. A piece of flared tube secured in the recess with the inverted flare nut came out tangent to the half round. It was fairly easy to bend the Cerrobend filled tube 90 deg. over the half-round. I held the block in the bench vise and used a piece of steel tube out of the scrap box as a lever to bend the tube. There was no spring back and I made all 9 bends with no losses ( lucky). I just found soome pics that I had forgotten:

The last pic shows a tube ready to bend. I actually used steel flare nuts, but brass will work just as well if you cannot find steel readily. I think I had to order the steel nuts from Mcmaster. My local auto parts store only had a couple of brass nuts in stock.

WOB

 

[Lloyd-ss]

WOB, a picture is worth a thousand words. Slick!
Lloyd

 

[stevehuckss396]

I thought about silver soldering a ring onto the end of a formed tube but concerned it might fail under exhaust temp. Any thoughts on that?

Silver solder would work fine without trouble. I have both brass and stainless steel exhaust systems silver soldered together and never any problems.

 

[petertha]

Thanks Steve. Did you use gaskets on exhaust side & if so what kind of material?

 

[petertha]

I improved my assembly a bit & got some good preliminary results. I kept the basics of the 0.25" thick forming plate so I can position it practically at the end of an existing tubing bend. I made the chamfer a bit wider face & smoothed the bore/transition a bit. I then made a small frame to push in a 45-deg forming plug. The center bolt is 10-32. Flaring the aluminium was quite effortless & controlled. I can feel the dead end quite reliably. The plate gripped the tubing well with just a hint of marring & OD reduction. In this example I left more tubing material prod than necessary so the flange was correspondingly larger OD at 0.415", but it formed well without splitting. So I think I have a workable solution just controlling the material stick-out. Time to make nuts & see how they seat on the flange.

 

[petertha]

Closeup. I would still flatten the face so it makes good seal.

 

[petertha]

Well the nut turned out like it was supposed to design wise. It has a matching 45-deg ID bevel to mate/seal the 45 deg tube flare angle. It also goes around the bend radius made by the tubing bender. I'm kind of a rookie threader but slowly getting the hang of it. For now pretend this plate is the cylinder head, haha.

 

[petertha]

If I cant make the funky 3D multi-bend induction tubes in one go (likely) I tested a bailout option mating 2 individual segments. I machined an aluminum coupler & used JB weld on the joint. Seems very strong I'm not even sure I need it rigid though. Maybe I can find some thick heat shrink tube or thin walled silicone tube & bond it. It wont see any appreciable heat. Ignore the 2nd flare, it was just a scrap piece of tube.