Tesla Compressor

Moderator: Mike Everman

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Tesla Compressor

Post by fastnova » Sat Nov 22, 2008 10:27 pm

After reading Nick’s thread I was inspired to experiment a little myself building a Tesla gas turbine.
I decided rather than go strait for a gas turbine I would design and build a small (ish) compressor using disks like a Tesla turbine and spin it up with a rather nice hyperactive nitro engine I have lying around.
This way I could experiment with different designs of disks and measure the pressure and airflow created. I also hope to work out the amount of torque required to spin it, with this I should be able to work out its efficiency?…

All in all I am hoping this should make an interesting little experiment.
As my knowledge of these is quite limited I would welcome any comments or advice anyone here has to offer.

So far I have started to make the disks, there will be six in my initial design.
They have a diameter of 80mm and the “effective” area is 2mm thick.

The engine I intend to use for spinning the compressor is a high performance 1.5hp nitro engine capable of going up to about 35kRPM, this with a 1:3 gear ratio gives me about 105kRPM (in theory), that should be more than enough to perform some interesting tests, or it could all just blow up in my face, only time will tell.


Here are a few pics of the disks so far, more to follow when I have more time.


Jonathan
Attachments
3.jpg
Almost finished, just awaiting air slots and a hole for the shaft
2.jpg
All six after about 45mins work
1.jpg
After roughing out and finishing the first opp
I don’t see problems, only opportunities.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: Tesla Compressor

Post by larry cottrill » Mon Nov 24, 2008 4:26 pm

fastnova wrote:The engine I intend to use for spinning the compressor is a high performance 1.5hp nitro engine capable of going up to about 35kRPM, this with a 1:3 gear ratio gives me about 105kRPM (in theory), that should be more than enough to perform some interesting tests, or it could all just blow up in my face, only time will tell.
Jonathan -

The disks appear to be basically flat. For more efficient use of material and possibly better performance, you could design them as "disks of uniform strength" (unless you have reason to believe that the non-flat surfaces would degrade their "Tesla turbine" properties for some reason). This gives a lighter disk that can handle higher RPM, given the same material. Invented by none other than the great Dr. Carl Gustaf Patrick deLaval (1845-1913).

When you spin these up, remember that "Nature always sides with the hidden flaw" ;-)

L Cottrill

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Mon Nov 24, 2008 9:43 pm

Larry, I am not quite sure what you mean by “uniform strength”
I think I should explain the design a little more as I left out a bit of info last time.

They are not quite as flat as it would appear; they have a recess in the middle where I intend to put some air slots and to save weight.
You can kind of see this in the first pic, but the other two pics are of the disks after roughing out so don’t have the recess in yet, it would be easier to see if I didn’t always find photographing even remotely shiny metal a real pain in the ass, it just seems to reflect back at almost every angle I try, probably just my crap camera?

My reasoning of having the recess in the middle is that I don’t want the slots to create a kind of turbine affect altering ultimately what I am trying to achieve. Also the surface speed drops as you approach the centre, therefore the centre of the disks can only provide drag to the air if it is already moving closer to the surface speed of the outer parts of the disks? Although this just seems logical to me I could well be wrong so if I am please put me right.

I haven’t read any books or even done a single calculation in the design so far, I am just taking a logical and “what feels/looks right” kind of approach to the design stages.
Probably a somewhat daft thing to do, but common sense is all to often overlooked in my opinion, I believe with the right application of a little common I shouldn’t be to far away. Again put me right if this is just wishful thinking.

This is only a first prototype, I intend to produce several different sets of disks, using many different materials and testing them all to see what gives me the best results.

As for wasting material, I’m not really worried about that as I am using bar ends and odd bits from work that are just lying around or waiting for the scrap man to turn up. We got so much metal in stock at the moment we could loose a few tons (literally) and no one would have a clue!

I would put up a few pics of the factory I work in and some of our toys but I don’t want to come across as condescending (wow big word) although I may well be a little too late? That is unless anyone would like me to put up a few pics, in which case I would be only too happy to.

Back to the point, below is a drawing of what the disks look like, I am a little worried about the strength and how they will hold up to the forces generated by 100k+ RPM especially as the recessed part in the middle is only 1mm thick, but we will see.

Jonathan
Attachments
Rotor Blank3.JPG
I don’t see problems, only opportunities.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: Tesla Compressor

Post by larry cottrill » Tue Nov 25, 2008 3:27 pm

fastnova wrote:Larry, I am not quite sure what you mean by “uniform strength”
It is kind of a misnomer; it is also called the "disk of constant strength", which is no better. What it really is, is a disk whose faces are a very shallow "sombrero" shape (but not turned up at the edge). It is actually a disk of constant unit stress during rotation; i.e. its rotation-induced stresses are never concentrated at any particular radius. There is, of course, a way of calculating it properly, but it is a calculus exercise of which I am unworthy.

The disk you show is OK, of course -- up to a point! Its main problem is that there will be a very high concentration of stress at the radius where it undergoes the transition to the thinner inner "web". At some RPM, it will blow itself apart, with the outer mass flying off in pieces, leaving the inner web of the disk practically intact. The question is, at what RPM? I will leave the answer to this as an exercise for the reader - ha.

At the very least, you should make sure that the transitions from thick to thin are NOT sharp, but are machined to gradual, flowing curves along the radius. This will mitigate stress concentrations, but will not ultimately get around the basic failure mode.

A real disk of uniform strength would have almost parallel faces at the edge; then, as you go in toward the center, the faces would slope more and more apart, but then as you get quite near the center, they would turn back to become parallel again, much thicker than at the edge. I'm sorry that I can't take the time to sketch the section out for you. My description that each face is a "shallow sombrero" is as close a picture as I can think of. If you have any knowledge of telescope optics, it is something like the contour of the front face of the famous Schmidt corrector plate designed by Bernhard Schmidt in about 1930.

Good luck! Don't stand too close ...

L Cottrill

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Thu Dec 04, 2008 10:15 pm

Hummm, could be interesting, would also increase the surface area available, which could be handy.
I did find some old chunks of 3” Titanium the other day so if it fails in a shower of shrapnel I do have something a little more exotic available, but for now am going to stick to ally.
I am a little worried about the disks having a little growth spurt at high RPM’s they could come into contact with the casing, stop a little suddenly and mess up my engine. I may incorporate a deliberately weakened point on the drive so if the disks to seize it will just shear off at some point leaving the drive engine un-harmed.

By the way does anyone know how to make an airflow meter, I thought about using one off a car, but having one and getting a reading off one are very different things, especially as I haven’t a clue how they work!
I don’t see problems, only opportunities.

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Tue Jan 13, 2009 9:23 pm

Its been a while since I posted any progress on my little project so decided it was about time, its been a bit busy at work trying to catch up with downtime over Christmas but have managed to get a few bits done.

The disks are now finished, the hole for the shaft came out a little bigger than I had hoped so I wont be able to use 6mm ground stock 303, I will have to grind down some ¼” but at least I can use 316.

I have also started making the outer casing, I just have to put it back on the lathe and face it to length, then it will be going on the mill for the finishing touches. The only suitable metal I could find that wouldn’t need drilling was HE30, a pain to rough out, but you can get a great turned finish.

Jonathan
Attachments
13012009451.jpg
13012009450.jpg
13012009449.jpg
I don’t see problems, only opportunities.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: Tesla Compressor

Post by larry cottrill » Wed Jan 14, 2009 12:19 pm

fastnova -

Man, what beautiful stuff! I hope this works every bit as planned.

Nice work, indeed!

L Cottrill

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Thu Jan 15, 2009 10:42 pm

Just a quick drawing to show what the finished thing should look like, obviously there are quite a few details missing, like all of the milling on the outer casing, it’s getting late and I cant be bothered, but you get the idea.
You might note I am slightly over engineering the housing; this is deliberate to provide protection should the disks fail at high speed, I figure it’s better to be safe than blind!
When I get more time I will add more detail.

Jonathan
Attachments
Assy3D 1.JPG
I don’t see problems, only opportunities.

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Mon Feb 02, 2009 9:31 pm

This is the latest edition to my little project for those of you who are interested…
Its not finished yet, but most of the work is done.
Attachments
01022009464.jpg
01022009461.jpg
I don’t see problems, only opportunities.

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Thu Feb 12, 2009 10:08 pm

Ok, I think I have made enough progress in the last week to justify an update, so here it is.

I finished all the tuning on the first of the casing ends and made a start on the second one.
Most of the milling is done on the outer casing, still needs a PCD of tapped holes putting on each end, but that will be easier to do at the same time as some of the other bits to save on set-up time.
Also started on the shaft, I ground it down from a piece of 1/4 316 and screw cut a left hand thread on one end, I had to go with a BSW thread in the end as they were the only left hand taps I could find near a suitable size, also made a handful of left hand nuts to suit. In hindsight I probably should have left the grinding till last in case I mark it, but never mind.

I should be ale to get quite a bit more done this weekend, will maybe try to get a few pics of things in action.

Jonathan
Attachments
12022009478.jpg
12022009475.jpg
12022009473.jpg
I don’t see problems, only opportunities.

Irvine.J
Posts: 1063
Joined: Mon Jun 05, 2006 4:28 pm
Antipspambot question: 0
Location: Brisbane, Australia
Contact:

Re: Tesla Compressor

Post by Irvine.J » Thu Feb 12, 2009 11:39 pm

I love this fastnova, looking forward to this for a long time.
Rock on.
James- Image KEEPING IT REAL SINCE 1982
http://pulseairdefence.com
[url=callto://project42labs]Image[/url]

Viv
Posts: 2158
Joined: Sat Oct 04, 2003 2:35 pm
Antipspambot question: 125
Location: Normandy, France, Wales, Europe
Contact:

Re: Tesla Compressor

Post by Viv » Fri Feb 13, 2009 12:04 am

Hi Jonathan

Nice build and machining work, looks like its coming to together very well, one question as I am to lazy to get too far in to boundary layer turbines is this, should the working area of the discs be polished smooth and shiny so they are low friction or alternatively bead blasted to increase their surface area and friction?

Viv
"Sometimes the lies you tell are less frightening than the loneliness you might feel if you stopped telling them" Brock Clarke

Viv's blog

Monsieur le commentaire

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Fri Feb 13, 2009 9:55 pm

Well now, boundary layers and surface finishes, this is where things can get a little difficult
I have been meaning to post some information about this as there is so VERY little info regarding this in relation to tesla turbines available, that I have been able to track down that is anyway.

I suppose now is as good a time as any.

Lets start with the surface finish.
This is where most people get it wrong, friction is the biggest enemy of the tesla turbine, friction between air and the disks will cause cavitation, generate heat, turbulence and completely disrupt the laminar flow.
If you think about it, what is friction? It is two surfaces rubbing together, this is the last thing we want, what in fact we are looking for is adhesion. We want the air to stick to the disks and be accelerated or visa versa the disks to stick to the air and spin.

Therefore, the smoother the disks the better the adhesion and the higher the efficiency of the engine.
A thought for a later project would be to try and find a way to keep the air in contact with the disks for longer thus accelerating it further. I have a few ideas, but these are for another day.

Now thats dealt with lets move on to the mathematics. I am still trying to decrypt and understand the equations so until I have a good understanding of them I wont put them up as I feel it would be a little silly of me to post equations I haven’t fully grasped.

When I designed the test rig I couldn’t be bothered to track down the calculations required, partly because many of them are rather complex and partly in my arrogance (I am not afraid admit it) I thought all that was needed was a little common sense, how wrong was I! It turns out I got the gap between the disks completely and hugely wrong! With this disk arrangement the gap is 1.5mm, when in fact for the speeds I plan to run this at the gap should be closer to 0.1mm. Although if I decide to use water between the disks it could be ideal!

The gap size is quite an important part of the engine, although it will work with almost any size, the more accurate you can make it the more efficient the engine becomes.

The diameter of the disks, the RPM they will be spinning at, the temperature and the viscosity of whatever is between them determine the width of the gap.

If the RPM increases the gap must decrease
If you use a thicker substance (like water) the gap must increase
If the diameter goes up the gap must decrease
And if the temperature goes up the gap must increase

This of course is providing the other values remain constant.

This would suggest that if the turbine is going to be used over a range of speeds there will be a peek in the efficiency at a given RPM with it dropping off either side, how quickly it will drop of I am intrigued to work out.
Perhaps then another interesting thought for a MK2 would be a variable disk gap to maintain the peek efficiency over a broad range in the RPM thus increasing the overall efficiency of the engine.

With a little cleaver engineering this could be achieved using centrifugal force and a system to automatically adjust the spacing as the speed goes up and down, maybe some sort of governor, perhaps this is not the right name, but I know what I mean.

Again, I have a few ideas lined up for my next project.

I’m sorry I can’t be a little more specific with working out the disk geometry but as soon as I have a grip of the mathematics I will post them up and explain them as best I can.

Another problem I have discovered is that I wont be able to spin the disks quite as fast as I had intended. 100,000RPM was my original max, but have worked out that the tip speed of the disks would be approaching 940mph or nearly 420m/s at this RPM.
I may have to calm things down a little, some of you may realise that 420m/s is somewhat above the speed of sound. Although, if I am able to reach these kinds speeds it would certainly be interesting to see how the performance of the engine is affected, that is if it stays in one piece, the centrifugal loads acting on the disks at 100,000RPM are going to be quite substantial. :?

Jonathan

P,s
Thanks to those of you who managed to get to the bottom with out getting too board of my ramblings.
I don’t see problems, only opportunities.

Mike Everman
Posts: 5007
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

Re: Tesla Compressor

Post by Mike Everman » Sat Feb 14, 2009 7:02 am

You may be all right on the periferal speed. Figure the temperature rise of the compressor as gas is compressed out there, and the resulting local speed of sound. Good project. It's nice to have a good Tesla build log around here. Keep it up!
You've worked out the stresses on the wheel at that speed?
Mike Often wrong, never unsure.
__________________________

fastnova
Posts: 55
Joined: Sat Sep 06, 2008 8:14 pm
Antipspambot question: 0
Location: Sussex

Re: Tesla Compressor

Post by fastnova » Sat Feb 14, 2009 2:43 pm

Funnily enough I worked it out just this morning, my calculations show that the centrifugal force will be almost 1ton on EACH disk. I am a little doubtful about these results, so will have to go over the math again. I was expecting it to be high, but not that high!
I am working on a spreadsheet to do a whole bunch of calcs for me. I will make it available to all when it is finished.
It will work out:
Centrifugal force in a bunch of different units
Mass of an object from volume and specific density
Surface speed at a given radius
And maybe a few other things if I can think of any, if you have any suggestions get them to me quick and I will try to include them.

Ok so none of the above things are exactly difficult to work out, but I think it would be handy to have a spreadsheet that works them all out for you.

Jonathan
I don’t see problems, only opportunities.

Post Reply