Help! Up against the wall

[sshire]

I’m about to order stainless tubing to replace the steel that came with the Rider-Ericsson castings.

The drawings spec a 1.85 diameter with a .0625 wall.

The closest tubing dimension that I can order, in the length that I need, is 2.0” OD with a 0.134 wall thickness.

Both the OD and wall have a +/- .010 tolerance.

If my calculations are correct, the worst case would be a 2.01 OD with a 0.124 wall. (+ on the OD, - on the wall)

Turning the OD down to 1.85, would leave a .034 wall which seems too thin.

Option 2 would be a 2.0” OD with a .156 wall. Again, worst case would be a 2.01 OD with  A .146 wall.  (+ on the OD, - on the wall)

Turning the OD down to 1.85 would result in a .056 wall. Better.

However, this option, going the other way with the tolerances (- on the OD, + on the wall)

This would be a 1.99 OD and .166 wall.

The result here, after turning the OD to 1.85, would be a .096 wall (which could be bored to .0625 if necessary)

My inclination would be to order the .156 wall tubing. Since worst case would be .056 wall thickness. The best case scenario with this tubing would be a 1.99 OD with a .146 wall.
After turning the OD to 1.85, I’d have a .076 wall.

If you’re confused, join the club.

Someone just tell me which tubing to order. Please!

[cfellows]

Given the uncertainty of the exact size you'll be receiving I would probably go with the .156" wall.  It may be more work but you have a better chance of getting the size you want out of it.

Chuck

[b.lindsey]

Yep, I agree with Chuck also Stan.

Bill

[mklotz]

The closest tubing dimension that I can order, in the length that I need, is 2.0” OD with a 0.134 wall thickness.

Both the OD and wall have a +/- .010 tolerance.

If my calculations are correct, the worst case would be a 2.01 OD with a 0.124 wall. (+ on the OD, - on the wall)

Turning the OD down to 1.85, would leave a .034 wall which seems too thin.

2.01 - 1.85 = 0.16 removed from the diameter so that means 0.08 removed from each side/wall...

0.124 - 0.08 = 0.044, not 0.034

Option 2 would be a 2.0” OD with a .156 wall. Again, worst case would be a 2.01 OD with  A .146 wall.  (+ on the OD, - on the wall)

Turning the OD down to 1.85 would result in a .056 wall. Better.

Again...

0.146 - 0.08 = 0.066, not 0.056

What am I doing wrong?  I've only had one cup of coffee and I have the distinct feeling that I've overlooked something obvious and will be very embarrassed when you reveal my error.

[sshire]

Thanks, Marv. Coffee or not, you're correct.
I had a page of all permutations for +- .01 of both OD and wall thickness.
The wall thickness range was .066 to .096

Not the neatest page of calculations. It's a wonder I posted any correct numbers.

The second question is that this tubing is for the displacer piston. Is the wall thickness (within the .066 - .096 range) that critical?

An aside:
Jerry Pournelle (you'll know who he is if you're a reader of science fiction) said that he would have gotten his PhD in Physics instead of Political Science if calculators had been available in 1964. He couldn't do the math.

[Tennessee Whiskey]

Stan, kinda like Marv here, except, I've absorbed way too much beer and hickory smoke, but, best I can remember from the last real job I had, there is a difference in pipe and tubing as far as O.D. and wall thickneses. If I can find my old briefcase I have some charts somewhere. We had to deal with it concerning piping vs. Instrument tubing. God, I thought I'd never forget that stuff. Thanks for the memory jog.

Cletus

[sshire]

Thanks, Mr. Ribs and Pulled Pork. This is definitely seamless tubing.

[mklotz]

The second question is that this tubing is for the displacer piston. Is the wall thickness (within the .066 - .096 range) that critical?

Perhaps for ultimate efficiency but, for hobby/demonstration use, I doubt it matters much.  The atomic level physics going on in the displacer chamber would probably need a supercomputer to model with sufficient fidelity to discover a difference.

[sshire]

Since this is the first hot air engine I wanted to be sure before ordering a $35, 6" piece of tubing.
Additionally, my Grizzly Atom Analyzer (G475443)  is waiting for a part under warranty.

[Mosey]

Stan,
Take a look at Zoro Tools for tubing sizes that may match what you need better than domestic sources, and for good prices. I bought some from them to make a draw tube for my jewelers lathe. Watch out for stuff made in India.
no, I don't own them, haha.
zoro.com
Mosey

[sshire]

Rob
Thanks. I've gotten things from them but never thought to look there for tubing

[Tennessee Whiskey]

Okay, look at 1.5" Sch5 seamless stainless PIPE. 1.9" o.d and .065" wall. I think the polish will take up the difference. I've welded Sch.5 before and you had to make sure there were no scratches in you lens, because they looked bigger than the joint, you. Didn't know which to follow

Old Pork Butt

Or as Marv said, in local government terms, what's a few thousands here OR there going to matter

[mklotz]

According to Wikipedia, Pournelle has a BS/MS in [spit] psychology and a doctorate in [spit^2] political science [the ultimate oxymoron], which means he shouldn't be allowed anywhere near a calculator or, in fact, normal people.  I'm not much for science fiction but I used to read him regularly in Byte and there he seemed remarkably human-like for something with those credentials.

Porky mentioned digging up his stuff on pipe versus tube.  For the benefit of the lurkers who may read this my ancient writing on the subject is offered below.  [I really have to stop wasting my time on lurkers.]


              PIPE AND TUBE

What is the difference between pipe and tube?  At first glance, the naive
might say "none".  After all, they're both just hollow cylinders.  That's
wrong, of course.  There is a difference (in the metalworking arena) and I
hope to clear up a bit of the confusion here.

For any hollow cylinder, there are three important dimensions - the outside
diameter (od), the inside diameter (id) and the wall thickness (wt).  Since
these three are related by a simple equation:

      od = id + 2 * wt

one can completely specify a piece of pipe/tube by supplying any two of these
numbers.

Tubing is more frequently used in structures so the od is the important
number.  Strength depends on the wall thickness.  So tubing is specified by
the od and the wt.  Very logical and simple to measure.  The id is simply
whatever falls out of the equation above.

Pipe is normally used to convey gases or fluids so the internal
cross-sectional area (defined by the id) is important.  It's therefore not
surprising that pipe is specified by the id.  Although anyone who's ever done
any plumbing knows that the id on the pipe label is only a *nominal* id.  As
an example, a (nominal) 1/8 wrought steel pipe will typically have a
*measured* id of 0.269 (schedule 40) or 0.215 (schedule 80).  (More below
about those schedule numbers.)

While the designation for tubing is straightforward, that for piping is
obscure for some perverse reason unclear to me.  All pipe of a given nominal
size has the *same od*.  An abbreviated list:

nominal
size      OD

1/8      0.405
1/4      0.540
3/8      0.675
1/2      0.840

Now, the folks (ASME?) who codify this stuff, in an effort to make things
difficult for us, instead of specifying the wall thickness directly, decided
to use (seemingly arbitrary) schedule numbers to specify the wall thickness.

For instance, a (nominal) 1/8 schedule 40 pipe will have a wall thickness of
0.068 (id=0.269) while a 1/8 schedule 80 pipe will have a wall thickness of
0.095 (id=0.215).

And, no, these schedule numbers do not reflect a constant wall thickness.  For
instance, a (nominal) 1/4 schedule 40 pipe has a wt = 0.088 while the same
pipe in schedule 80 has wt = 0.119.

Schedule numbers range from as small as 5 up through 40, 80 (common) to as
high as 100, 120 and 160.  There may be others.  This is not my area of
expertise.  Larger schedule numbers correlate one-for-one with thicker walls,
which seems to be the only predictable thing about schedule numbers.

To the best of my knowledge, there is no mathematical relationship that can be
used to translate schedule number into equivalent wall thickness.  You're
forced to consult a table.  Machinery's Handbook has such a table for wrought
steel pipe (pg. 2378 in the 23rd edition).  Do these tables also apply to pipe
made of other materials (e.g., plastic)?  I don't know, but I doubt it.  That
would be too simple.  Since I don't want to make a career out of plumbing
minutiae, I'll let you research it for your application.

I can only guess that the schedule number relates to some burst pressure and
thus the relationship to wall thickness is non-linear.  But that's only a
guess - anyone who knows the real story please correct me.

Why do you care?  Well, beyond the satisfaction of simply knowing some obscure
metalworking stuff, this should help you in selecting and specifying hollow
cylindrical elements for that project you have planned.  It should help you to
understand why you won't have much success trying to bend tubing with a pipe
bender.  On the latter, the bending dies are sized to the (constant) pipe ods
mentioned above.  It's unlikely they'll fit any tubing you buy since tubing od
generally comes in straightforward sizes like 1/4, 3/8, 1/2, etc.  If you want
to bend tube in a pipe bender, count on making some purpose-built dies - a
tricky lathe job.  Or buy a tubing bender.

[fcheslop]

Used stainless electrical conduit before with no ill affect .Just dont know the sizes you guys use
Its a bit tough going but works
Home depots often have stainless goodies
cheers

[gerritv]

Marv and other, thank you for all the details, it has given me hope of making a MT2 tailstock for my Taig lathe. Now that I know about pipe I no longer have to machine from solid!
 

Gerrit

[IanR]

I'm not an expert on hot air engines, but I think it's generally recommended to keep the wall thickness of the displacer as thin as possible, to minimise heat retention or conduction. HTH
Ian

[Tennessee Whiskey]

Here is an old pipe fitters rule concerning pipe: 6" and below is measured by the ID, as the schedule increases, so does the OD. 8" and above is measured by the OD, as the schedule increases the ID decreases. Kinda weird ain't it. I welded a ton of 4" SS heavy wall and it was 4" ID with an OD of 5.375". 

Porky

[sshire]

Who I thought that system was a good idea? Sounds like a ploy to keep pipe fitters employed since no one else could figure it out.

[mklotz]

Marv and other, thank you for all the details, it has given me hope of making a MT2 tailstock for my Taig lathe. Now that I know about pipe I no longer have to machine from solid!

Have you considered buying and modifying a Morse taper socket...

http://www.use-enco.com/CGI/INSRIT?PARTPG=INSRAR2&PMAKA=214-8705&PMPXNO=945794

[PStechPaul]

Here's a nice list of specifications for schedule 40 and schedule 80 pipe:
 

So it looks like, for Schedule 40, the ID matches the nominal size closely from about 2" to 12", and the wall thickness (t) for a pipe size (s) is very close to 0.1*sqrt(s). It's not as straightforward for schedule 80, although it is close to 0.17 * sqrt(s). It may be more linear with respect to actual ID, but that gets a bit more complicated. 

[b.lindsey]

Stan, in this case as with most Stirlings, I don't think the wall thickness is a critical as keeping the weight of the displacer piston light. As you may recall,  this was the one part I used the supplied tubing for and painted it with hi-temp black paint to minimize corrosion. If and when I do tear it down and replace the displacer piston, I would also consider aluminum for that. The waist of the hot cap is another matter. that section of the hot cap is usually made as thin as possible to minimize heat transfer as mentioned already.

Bill

[sshire]

Bill
Using stainless only for the cylinder and hot cap tube would simplify this whole issue. Aluminum is an intriguing idea and a less expensive option for the displacer piston  The issue woud be attaching the end caps. No TIG welding. With the thin wall, single point threading doesn't seem like a viable option. So, either JB weld or high temp Loctite? 

[gerritv]

I have yes. But the ones found so far have a hardened tang and/or body, which means I can't machine that part. Looked at both taper sockets and extensions. I might order one of each when in Florida for the winter, and then anneal the area I need to machine.

Thanks,
Gerrit

Marv and other, thank you for all the details, it has given me hope of making a MT2 tailstock for my Taig lathe. Now that I know about pipe I no longer have to machine from solid!

Have you considered buying and modifying a Morse taper socket...

http://www.use-enco.com/CGI/INSRIT?PARTPG=INSRAR2&PMAKA=214-8705&PMPXNO=945794

[Ian S C]

For the displace hot cap on my version of James G,  Rizzo's Dyna (a horizontal BETTA motor), I got a 3" length of 2 1/2" 316 stainless steel, and carved it out of that, the hot end is about .015" thick, the bore 1 3/4".  The open end has a internal thread that matches the other half of the cylinder.  Since then I'v been trying to use thin wall stainless tube with an end TIG welded on.
DON"T use aluminium for the displacer, (a) it conducts heat too fast hot end to cold.         (b)you stand a good chance of a melt down, I'm a slow learner, I melted 2 aluminium displacers on one motor before I changed to stainless steel.          Ian S C

[sshire]

Thanks for that info, Ian.
Just how hot do these engines run?


6061 Melting Point   582 - 652 °C   1080 - 1205 °F    

[b.lindsey]

Ian, I certainly can't dispute your experience, but IIRC Jerry Howell calls for aluminum in all of his Stirlings, at least those I have built. While it may be true that aluminum will come up to the internal temp of the hot end faster, any metal will eventually do that. The displacer piston's purpose is to push the hot air from the hot end up to the cooler end and then back to the hot end as it cools again. Keep in mind the hot cap transfers heat via conduction...hence stainless along with a minimum waist wall thickness limits that conduction. The convection that happens within the displacer chamber is less efficient. IN addition Stan's RE engine has a water jacket around the power piston to assist in keeping things cool. There is also a thermal insulator between the hot cap and the base casting which helps as well. Even after an extended run, the base casting of my RE never gets hotter than I can touch, and all of the castings are aluminum.

Stan as for joining the end caps to aluminum tubing, JB weld for a good seal and some adhesive properties, but I would also put in a small screw every 90 degrees in both the top and bottom caps for a mechanical connection. Small flat heads ( say 1-72 or 2-56) shouldn't interfere with the small clearances between the displacer piston and the side walls.  Just my 2 cents.

Bill

[sshire]

Bill
I had read much Stirling info before beginning this build. You are correct re: Jerry Howell's use of aluminum displacer pistons.
I found, at Online Metals, 6061 seamless tubing
1.875 OD with .06 wall. A cleanup pass on the lathe should get me near a 1.85 OD with about a .058 wall. Close enough.
I like the JB Weld + a mechanical connection.
That seems to solve the displacer piston question.
The cylinder and hot cap will be 304 SS tubing. To be ordered today along with some SS flathead screws.
Look for a build log post this evening.

[Ian S C]

bill, an example of aluminium displacer:  on mk 1 of my free piston GAMMA motor,  mild steel hot cap, aluminium aerosol can displacer, I had 3 melt downs(a bit of a slow learner) until I changed to stainless steel.  Most of my motors have the hot end (about 1/3) blue.  I will admit that I try to push the temperature as high as I can. 
     This is a hot cap, made from a bit of 316 bar,  the bore made first, including the thread on the open end, then it was placed on a mandrel, and the out side taken down to about .020", this runs well at red heat.   
      Of late I'v been making the hot cap of thin walled stainless tube with a flange attached at the cold end with high strength/high temp shaft lock adhesive, the end is sealed with a disc TIG welded on.  Same sort of construction for the displacer.     Ian S C

[tvoght]

I have done the same as Ian, turning hot cap and displacer from sold stainless rod (303 not 316!).
I don't weld anything though, I bore them first to depth and flatten the bottom of the bore, then cut them off leaving a blind end.

They are then placed on an expanding mandrel that tightens from the chuck end, so that I can leave the blind end in place for turning down the
thickness of  the walls and end. Wall thicknesses of .02 and less are surprisingly easy to achieve. The parts are amazingly light and pretty to look at.

I hope to show all this soon in a retrospective build log for the machine I'm working on.

--Tim

[sshire]

I found this on the stirling engine forum. The concept of an aluminum displacer acting as a regenerator.


http://stirlingengineforum.com/viewtopic.php?t=641#p1659