Author Topic: Gamma1-VPA Stirling  (Read 12619 times)

Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Gamma1-VPA Stirling
« on: August 18, 2014, 05:26:26 PM »

This is the retrospective build log of an engine I've just nearly completed.
Retrospective as it is, I start with some engine photos and a video.


The "Gamma1-VPA" is a gamma-type Stirling engine of my own design. Mostly for
novelty, it features a variable-phase mechanism which allows the lead angle
between the displacer and the power piston to be changed.


I have always planned to do a build log, but it was to come afterward, after I
had proved that a Stirling engine of my own design would actually run.


With a log here in mind, I took photos during the build and made an effort to
rough out some notes of what had transpired. Sometimes I "fell down on the job"
and didn't keep notes. Some parts lack photo documentation.


In the next few days or weeks, I'll be working on the write-ups a part at a time
in roughly the order the parts were built. Some of the later parts lack any
notes at all, so I'll have to take cues from the photos to remember how they
were made.


There is still work to be done, such as making a new flywheel. The one seen
here was just for testing.


"Beauty shots" and a video.





<a href="http://www.youtube.com/watch?v=lpzM9CvXTVI" target="_blank">http://www.youtube.com/watch?v=lpzM9CvXTVI</a>

Questions welcome,


--Tim


Offline b.lindsey

  • Global Moderator
  • Full Member
  • *****
  • Posts: 13686
  • Dallas, NC, USA
    • Workbench-Miniatures
Re: Gamma1-VPA Stirling
« Reply #1 on: August 18, 2014, 05:54:16 PM »
What a nice engine Tim !  I love the geometric touches to the supports and cylinder exterior, etc. The variable phase angle is also another very interesting feature. Will look forward to seeing more on this one, even if retrospectively :)

Bill

Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #2 on: August 18, 2014, 05:58:45 PM »
Thanks Bill.
The geometric touches are to some degree CNC learning exercises, and only successful to some degree, as will be seen as the log develops.

--Tim

Offline mklotz

  • Full Member
  • ****
  • Posts: 2403
  • LA, CA, USA
    • SOFTWARE FOR PEOPLE WHO BUILD THINGS!
Re: Gamma1-VPA Stirling
« Reply #3 on: August 18, 2014, 06:12:44 PM »
Very attractive engine.

I'm sure we would all like to hear what you learned with the variable phase angle feature.  How about a shaft power versus phase angle plot?
---
Regards, Marv


Home Shop Freeware
http://www.myvirtualnetwork.com/mklotz

Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #4 on: August 18, 2014, 06:44:04 PM »
Hi Marv,

I have not learned much thus far, as the engine has not been running that long.

I actually did some first-order isothermal analysis resulting in some highly optimistic plots of shaft power vs phase angle, and
especially shaft power versus swept volume ratio. That is how I chose my initial design parameters for these things.

I had grandiose plans of fitting a pressure sensor and shaft position sensor for data acquisition so that I could directly compare experimental values to indicator plots from my first-order analysis.
There's noting to keep me from doing that later. I'd also want to rig a little prony-brake type dyno.

If there is any interest, I could do a separate thread with some information on how my analysis was done.

--Tim

P.S. Initial impressions are as the analysis suggested: phase angle does not make as much difference as you might think.
« Last Edit: August 18, 2014, 06:48:14 PM by tvoght »

Offline Roger B

  • Global Moderator
  • Full Member
  • *****
  • Posts: 4010
  • Switzerland
Re: Gamma1-VPA Stirling
« Reply #5 on: August 18, 2014, 06:49:57 PM »
Interesting engine and an instructive video.  :praise2: The power cylinder looks like the body of a high wattage resistor to me?

I would certainly be interested in your analysis and any power figures  :ThumbsUp:  :ThumbsUp:
Best regards

Roger

Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #6 on: August 18, 2014, 06:53:22 PM »
Roger,
The resemblance of the cylinder to a large power resistor was not lost on me, but it is in fact purely coincidental. The cylinder is made of cast iron and CNC milled to look like that.

Thanks for your comments!

--Tim

Offline MMan

  • Full Member
  • ****
  • Posts: 81
  • Hampshire, UK.
Re: Gamma1-VPA Stirling
« Reply #7 on: August 18, 2014, 10:31:55 PM »
Hi Tim,

I would like to see your analysis too. I want to understand how mine works and would be glad to learn from you.

All the best,

Mman.

Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #8 on: August 19, 2014, 03:46:21 AM »
Thanks for your comments MMan. I am trying to get together some material for a thread with notes on the analysis (hoping that noone
mistakes me for an expert in any way).

Construction will start at the hot cap.

A length of 1.25 inch diameter 303 stainless steel was centered in the 3-jaw
chuck, faced, center-drilled and then drilled with a 15/16 drill to a bit more
than the final internal depth of the hot cap.

I had reservations about starting with such a large drill, but I researched the
proper speeds and feeds and tried to stay in the correct neighborhood. It
worked just fine, with only a couple of belt-slips when I got too aggressive.
The photo suggests I might have gotten just a little too hot.



I used a 7/8 endmill in the tailstock and faced to the depth of the drill tip.



From here, a boring bar was used to flatten the bottom further, and to open
up to the desired inside diameter give or take a couple of thousandths.

Once bored, I measured the depth of the bore and then faced the end to
acheive the desired internal depth. Then I thinned down the walls to about
.050" thickness, to be thinned in a later operation on a mandrel.



Finally I cut off the cap at a length from the faced end calculated to give me
about a .050 end wall thickness. I used a cutoff tool until the noises became
unpleasant, at which point I shut off the lathe and finished the job with a
hacksaw.  Again, the end wall will be thinned down further with the part
inverted on a mandrel.
Hmm, I don't have a picture here of the part in that intermediate state.

Next, the workpiece will be placed on a special mandrel and the walls thinned
down further. The thinner the better, since I want heat to transfer well
through the end wall, but conduct badly lengthwise toward the cold end of the
engine. Thin walls will work in my favor on both counts.

Heat conduction is also one rationale for using stainless steel, as it is a
pretty awful heat conductor. Thats an advantage for preventing the heat from
conducting down the length of the hot cap, but a detriment for conduction
through the end.

More follows,

--Tim


Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #9 on: August 19, 2014, 03:49:16 AM »
Here's the mandrel I made to thin down the walls of the hot cap.
The mandrel fits in a 1 inch collet. It's drilled and threaded 1/8-27
(tapered pipe thread). The drill hole extends throughout the length of the
mandrel, but the threaded depth is just a little more than half the length of
the sawn slots.



The expander is a pipe cap with a socket hex head. The other end is drilled
and tapped for a 8-32 cap screw. The cap screw is shortened and then silver-
soldered into the cap. Now I can drive the expander from either end of the mandrel.
This allows me to tighten the mandrel from the rear when the hot cap is
covering one end. This expander design got me through this build, but it's
not really satisfactory, as I'll discuss later.



Here's how the mandrel is tightened from the rear (collet) end. by a left
hand turn to pull it further into the tapered threads.



In use. Ah, there's the missing photo of the roughed out part.


The end and side walls were thinned to about .015 inches. With this expander,
it was impossible to get the mandrel tight enough. The part would slip on all
but the lightest cuts, It took many very shallow cuts to reduce the wall
thicknesses.

The finished part.




Thanks for stopping by.

--Tim
« Last Edit: August 19, 2014, 03:52:36 AM by tvoght »

Offline Ian S C

  • Full Member
  • ****
  • Posts: 1065
  • Stirling Engine Maker Darfield Canterbury N Z
Re: Gamma1-VPA Stirling
« Reply #10 on: August 19, 2014, 12:32:01 PM »
Tim, looks like a well thought out motor.  I'v made hot caps that way from 316 stainless(all I could get), only difference mine have a thread cut in the open end, and around that area six radial holes for a C spanner. I'v gone to fabricated hot caps lately, thin walled tube, end TIGed on.       Ian S C
« Last Edit: August 19, 2014, 12:41:35 PM by Ian S C »

Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #11 on: August 19, 2014, 04:53:05 PM »
Thanks Ian. I've never machined 316 stainless, but from what I hear, I wouldn't have wanted to do this with 316. Seems it would have been quite a different story. Kudos to you!

The displacer shell was made in the same way the hot cap was. The final
step (the thinning down of the walls on the mandrel) was done after cutting
the same mandrel down to the smaller diameter of the displacer I.D. I won't
 belabor the point with more photos, except for this one of the finished shell.



The displacer cap is turned from aluminum. Here is the stock in a collet
with a shoulder turned back to make a sliding fit in the stainless shell.
It has been drilled and threaded 5-40 in this setup for concentricity (note
that the rod will screw in from the other end of this thread). A tool with
a sharp point was used to square up the shoulder and relieve the inside corner.



The aluminum cap was sawed off and placed in the 3-jaw chuck with sawed end
facing out, and then faced to length, leaving a top flange. Here's a shot of
a trial fit.




Thanks,

--Tim


Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #12 on: August 20, 2014, 12:02:08 AM »

The Cooler came into the shop as a length of 2.25 inch 6061 Aluminum bar. 
I sawed off a length with generous chucking stub, and centered it up in the
4-jaw.  The outside was turned just to clean up, and the end was faced and
a boss formed where the displacer rod packing gland would go.
I then tried to cut a fin with a .063 inch HSS cutoff tool.


This photo shows how the cutter skewed off to the side. I realize I could have
fiddled around and maybe gotten things straightened up, and then I started
thinking about cutting 10 or 11 spaces using this method. I quickly went to
plan 'B'.





Plan B. CNC. This is something I had tried before and gotten decent results.
The cooler stock was clamped upright between v-blocks in the mill vise.


The program was to lower a 3/32" slitting saw to each fin space position and
to circle the saw around the piece in decreasing diameters until the fin depth
was reached. The process took a while, but was fairly stress-free and the
results are pretty good.







I'd gotten this nice lathe faceplate a couple of years ago but not used it til
now. I started by facing off the new plate to get it true to the lathe. Then I
threw together this tooling plate with clamps which would clamp on the topmost
thick flange of the cooler (clamping on the first fin).






After centering, I cut off the most of the chucking stub and faced the part to
length, followed by drilling all the way through the part with a drill of about
.23 inches diameter.
Then I drilled with a 7/8" diameter Silver and Deming bit, taking the tip of
the bit to the depth of what will be the blind bore of the Cooler.
I bored to the correct depth and diameter, trying my best to get a flat bottom.
This left a .23" hole at the far end, which was then reamed to .250". The
displacer rod packing gland will fit in that reamed hole.
Not shown is the shallow recess I cut around the open end of the chamber. The
flange of the hot cap will sit in the recess.





I had an off-the-shelf expanding mandrel of 1", which happens to be the
internal bore of the cooler. I put the part on the mandrel and was able to
access the socket head screw that tightens the mandrel by reaching the key
through the reamed gland hole. Thus mounted, I cleaned up the outside to be
concentric with the inside. That left a nice finish on the fins, just needing
some deburring later to crispen the edges.





The Cooler is not finished, but that's as far as I'm going for now until some
other parts are finished.


Thanks for looking in,


--Tim


Offline tvoght

  • Full Member
  • ****
  • Posts: 929
  • Indiana
Re: Gamma1-VPA Stirling
« Reply #13 on: August 20, 2014, 05:16:47 PM »
To make the Bearing Stand I started by rough-sawing a bar of 3/4" 6061 Al.
The first milling operations were at the Bridgeport. The work was clamped in
the vise and two reference holes drilled. The hole for the Bearing Block was
first drilled with a large bit:



And then bored using the boring head. I've finally gotten the knack of
using these telescoping gauges to a really decent accuracy.



The bored part was clamped at the CNC mill. The paper pattern was used to
help me get the clamps in a non-interfering position. After clamping, it was
torn away. This photo was taken before I realized that it must be clamped atop
a sacrificial plate (that's seen in the following photo).



After centering up on the bore, a program was cut loose on the work. It cut
the outline contour at many steps of depth. One thing to note is that I just
cut to the finished dimension with the cutter cutting on both its sides (slot
cutting).  I have since concluded this is a mistake, as will be shown.



Here is the finished contour (before any deburring of course).



And here's the ugly truth. The cut appears to have been chewed by shop-rats.
I blame this on cutting the contour with the end mill cutting on both its sides.
Future work will attempt to rough the cut first, with a finish pass cutting
only conventionally.



I used files and sand-paper to clean up as best I could. A general roughness
here will have to be a part of this engine's personality. Chalk it up to
experience (or lack thereof).

Til next time,

--Tim


Offline MMan

  • Full Member
  • ****
  • Posts: 81
  • Hampshire, UK.
Re: Gamma1-VPA Stirling
« Reply #14 on: August 20, 2014, 05:41:06 PM »
Hi Tim,

I ran into the same problem of recutting in a slot making the plate for my Bas, looked much like yours.

I had to remake for other reasons and what I did to avoid this was cut first with a 9.5mm cutter and then, once cut to full depth, go round with a 10mm cutter. The 10mm cutter is 0.25mm per side (or 10 thou) bigger and so a reasonable finishing cut and cleans up the marks left from the first cut.

Mman.