steam turbine and generator

[Paul HorthParticipant @paulhorth66944]

I am thinking about having a go at a small steam turbine and generator. I have in mind a machine which can be mounted on my 2 inch traction engine (the source of steam) and which would be reasonably to scale. A turbine wheel of about 1.5 inch diameter and a generator about the same diameter and about 2 inch long. Enough power to light a few little bulbs, say 1 – 5 watts.

Now, such a small turbine rotor will have to run at a very high speed – 50000 to 100000 rpm. Leaving to one side the question of how to produce the blades or passages in such a wheel (which i am still thinking about), my  next question is where can I find a small electric motor which can be run as a generator at this kind of speed? It would need to be direct-coupled not geared down.

Most small motors that i can locate seem to be specified for 6000-10000 rpm which I think will not be high enough.

 

If anyone has any suggestions or experience of making a small steam turbine, then i would be most interested to hear from them.

 

Paul

[Paul HorthParticipant @paulhorth66944]

[Alan Worland 1Participant @alanworland1]

Hi Paul, I reckon the motor armature would burst long before it reached your goal rpm! I have never had any experience of what you are trying to achieve but I would suggest it may be possible to rotate one (or more) of those super powrfull neo something magnets within a winding?

 

[John HaineParticipant @johnhaine32865]

Another issue is that generators are intrinsically AC machines, even though they may have a commutator, so you have to consider core losses.  At 60,000 rpm a dynamo would be working at 1000 Hz where ordinary laminated motor cores would be rather lossy.  Best would be either ferrite or even (assuming neodymium magnets) no cores at all?  Also use a diode rectifier rather than attempting a commutator. 

 

You could also think about a Tesla turbine which uses plain discs rather than blades – much easier to make.  Very exaggerated claims are made for them (as for everything to do with Tesla [oh dear, I can feel the flames already]) but worth looking at.  Goole will lead you to more sites about Tesla than you want.  Wikipedia has a reasonable page –

here .

 

 

[DoddyParticipant @doddy]

Nice set of plans on John Tom site of a turbine by Christopher Vasconcelos

http://www.john-tom.com/MyPlans/SteamPlans3/TurboGenertor/TurboGeneratorAll.PDF

Will your steam plant keep up with steam demand? turbines eat steam

[TerrydParticipant @terryd72465]

Hi Paul,

 

Just a thought. Why such high speeds on a model?  We used to make simple demo generating plants using simple 12v DC motors and small filament bulbs, using Mamod steam plants as the power source. 

 

May I suggest that nowadays it should be possible to use high power leds as a light source (disguised as lamps if necessary).  Use them is series to match your output voltage.  A cheap rechargeable drill is a good source of small dc motors which are quite robust and capable of a good current output when used as a dc generator.

 

Regards

 

Terry

[Paul HorthParticipant @paulhorth66944]

Thank you all for your replies, all within an hour, amazing. These are helping me to think about the practicality of this problem.

To reply to some points raised:

Alan ,and J ohn – I was hoping to avoid the task of designing and building my own generator, as I consider both designing and building an electrical machine to be well beyond my skills. If I cannot make use of a commercially available machine, then I had best stop now I think.  I have read a book on the Tesla turbine but I can’t understand its principle – it seems to work on friction between the discs, and i can’t apply my thermodynamics and fluid flow to the concept.

David – I haven’t yet gone to the site yet, but will do. Your point about steam demand is absolutely right. I want to limit the flow to about 3 lb/h, as the traction engine boiler can make up to about 12 lb/h. This is a small flow and one problem is to make the turbine passages small enough so that the steam velocity is still up at about 300 m/s in the wheel. I would use a jet of about 0.05 inch diameter.

Terry – maybe my speed doesn’t need to be so high, but even 20000 rpm would be high for a typical motor. The turbine efficiency drops off rapidly at lower speeds and so I wouldn’t get the power – but I don’t really know?

 

Thanks again,

 

Paul

[Jeff DaymanParticipant @jeffdayman43397]

Hi Paul,

 

I would suggest building your turbine and test run it with a dummy load using a small drum and a brake band. This will give you an idea of what power and speed the turbine will deliver. Then you can plan your transmission and generator combo.

 

You mentioned you could not use gearing – may I ask why? many turbine applications use gears to reduce output speed. A worm and wheel or a planetary gearset would give a high ratio in one stage and would work fine for a low power turbine. Jensen model turbines from USA have a worm and wheel first stage and a belted second stage I believe. Try Googling Jensen turbine.

 

Maybe a scrap electric cake mixer or electric screwdriver could be a source for worm or planetary gearing – these appliances have them.

 

A few years ago ME ran a series by Peter Southworth on his Turbo locomotive which worked well and was powered by a Stumpf style turbine. Would be well worth the read. His biggest issue overall was adquate steam supply, and I believe he ended up redesigning the loco as a pseudo Garratt type with two power bogies and a much bigger than original boiler to run the turbine.

 

Re Tesla turbines – they can be made to produce good power to weight ratio but multiple disks and very close disk to casing clearances are critical, based on my own experiments. In principle it is a simple device but the devil is most certainly in the details. I never got one to produce the expected power. Unfortunately there are many forum-dwelling myth-makers who embellish Mr.Tesla’s achievements, and reality often goes out the window. However Mr Stumpf’s turbine wheel as done by Mr Southworth was both sound in principle and reasonably easy to make, with Woodruff cutters and a dividing head as I recall.

 

JD

[John OlsenParticipant @johnolsen79199]

I have never played with this sort of thing myself, but have seen this sort of thing tried. Locally the full size locomotives often carried a Pyle-National turbo generator set, which is an American device that sits on the side of the boiler and generates enough power to run the locomotive headlight, plus a few lights in the cab. In full size this is very successful. In model sizes they can be made to work OK but tend to be a bit hungry on steam. The ones I have seen used a small DC motor direct driven off the turbine, and I think the mismatch between the speed the turbine ought to be doing and the speed the motor can survive is one reason for the low efficiency. A small AC machine with a solid state rectifier would be a better bet I think.

 

So far as I can recall, the tubine wheels have been the type  that is a drum shape, with small scallops made by an end mill to make a bucket shape…I think that makes it an impulse wheel? I suspect like all turbines, they would be happier exhausting into a vacuum, but that is not so easy to arrange on a wheeled vehicle.

 

You could dig out the articles about Professor Chaddocks experiments, he was aiming at something a little larger but it would still be good reading.

 

regards

John

[Richard ParsonsParticipant @richardparsons61721]

 

Small turbines ‘Oi-vay’!. You would get more power out of a small oscillating steam engine that out of a small turbine. I have heard a lot of reasons for the low efficiency of these little brutes. But I am still not convinced.

In the mid to late ‘80 I had a lot of correspondence with the late Prof Chaddock on this subject. A few years ago I tried to get help with a letter to the editor Model Engineer.

The problems you will have are

A.   The bearings – they tend to ‘fry’ I tried all sorts even old battery electrodes!

B.   The gearing – most wheels seem to need the attention of a good dentist after a few minutes.

C.   Balance – Prof Chaddock wrote to me and said that he had to get the centre of rotation within 0.000002” the centre of mass. He did this with two razor edges some ‘bluetak’ and a small scraper.

D.   Low power output – it is abysmal I used to test run mine using a magnet and a coil connected to an old oscilloscope. I could hear the speed fall off when I connected them together.

E.   The turbine wheels I tried: DeLavals, Stumphs, and once as an absolute masochist a three stage Curtis   Stumpf wheels are the easiest to make with a little cone shaped end mill.  DeLavals you can make in two stages. Stage 1 gash your wheel. Stage 2 make a small hollow thin walled saw edged cutter to cut the ‘inner’ side of the blades.  I did not worry about the outer side of the blade. Sometimes I just gashed the wheel, shaping them made little difference. 

F.    Test them on a stout bunker. I used to disintegrate 75 to 85% of them.

G.   Have 50 times more steam available than you think you will need. Then treble that amount. Small Turbines are GREEDY. At low speeds they drink it and at high speeds they consume it as if it was going out of fashion!

Tessla turbines: I made one. To get it to run the turbine was made a push fit into its casing at normal temperatures.  As things warmed up it would expand and it would turn. I only once got it to run at nearly ‘full chat’ and the plates distorted at about 90,000 rpm!.

 I used a flash steam boiler made of 3/8” diameter about 18 foot long mild steel tube heated by 3 large propane burners I borrowed from a roofer. I found I could light a cigarette of the turbine input pipe.

I spent about 6 years on the project and got nowhere!

Finally remember that the speed of sound (in air) increases by about 0.6 M/S per degree C. So be careful when folk talk about steam flow being supersonic it often is not.

[V8EngParticipant @v8eng]

Not sure if there might be some advice available from the Gas Turbine builders association.

They have experience of the speeds involved, but never tried anything in this line myself.

I do not know if there any is relevance, but have included a link to their website.

 

http://www.gtba.co.uk/

[Donald MitchellParticipant @donaldmitchell68891]

Hi Paul.

 

A steam turbine such as you might require has already been successfully designed and made by Raymond McMahon, from Stranraer, Scotland.

 

The turbine was described in ME sometime around 1 year ago and will be further described during the build of Raymonds Darjeeling B Class locomitive currently being serialised in ME.

 

I will try and find the relevant ME issue number this evening and report later. 

 

Donald Mitchell
Castle Douglas
Bonnie Scotland

[Ian S CParticipant @iansc]

There has been an artical in ME about a Turbo-generator to opperate the head light on a locomotive, I’ll see if I can find it over the next day or so. Ian S C

[Paul HorthParticipant @paulhorth66944]

Thanks again for the latest replies, I am benefiting from this education. I don’t really have the motivation for a lengthy period of experimentation and development, so I am grateful to learn from those who have already put in this effort.

 

What I had in mind was something that looks like the US Pyle-National loco turbine, with the turbine and generator direct coupled and in line. That’s why I wanted to avoid reduction gearing. It may be unavoidable that the practical speed limit means that the efficiency and power output are very low with this arrangement.

 

Donald – I remember the article by Ray McMahon, I have it somewhere. I think he used an ex cycle dynamo? I will check.

 

Jeff – I remember the Southworth turbine. That was an ambitious project and was for direct drive to the loco (no generator) and had the full steam output of a large boiler. it was impressive that he got this to work, but i was after something smaller, which might not be practical.

 

Richard Parsons – destiny in that name! I appreciate your sharing of many years of experimentation, enough to put me off, I would say. I do understand all the points that you make. I was aspiring to make an impeller rather closer to the conventional impulse wheel than the Stumpf, but I do accept that the Stumpf will work, although the efficiency would be poor. I am too lazy to attempt muli-stage turbines. Anyway, first I want to establish what the electrical end should be, this is really what prompted me to write in to this forum.

 

Some years ago the ME had an article about a Swiss gentleman who had built a turbo-electric loco in Gauge 1. He had fitted a steam turbine and generator into a Gauge 1 loco  which drove motors in the train and was radio controlled. A truly amazing achievement. I think his turbine had chevron-shaped passages milled into the periphery, and drove a special motor of Swiss manufacture, but he did not include details such as the rpm. I don’t know if such motors are available here in the UK.

 

For bearings I had thought that those used by the IC engine fraternity would be OK…???

 

Provisional conclusions – a Stumpf impeller connected to a small high speed DC motor at about 15000 – 20000 rpm might yield a tiny dribble of power? and might fall apart soon? I wonder if it is worth pursuing this?

 

Thanks again

Paul

[Jeff DaymanParticipant @jeffdayman43397]

Richard, your Tesla turbine experiences are along the same lines as mine were. Not successful at all, unfortunately.

 

I had forgotten about the turbogenerator article by Ray McMahon, I agree, reading that would be a very good start for the project.

 

Get the best special high speed bearings in the smallest possible OD you can for it. The bigger the OD the more they seem to want to explode at high speed.

 

JD

[Richard ParsonsParticipant @richardparsons61721]

 

Paul – it may have been genetics or just stupidity – think the latter. From memory the Swiss gentleman uses a DeLaval type rotor. My DeLavals were slab cut turbines rotors were gut with a slitting saw set to give a 60° (or there about) angle of attack. I also tried by drilling and reamed a series of holes near the rim of the rotor at about 70°. It worked, and is how I built the Curtis.

There is something you could try as you are talking about 2” (50 mm) diam rotor. Make your rotor from non-magnetic stainless and insert a rare earth magnet and a pickup coil in the case. Keep your bearings well outside the hot area. The coil(s) can be rectified to give DC. You might just get enough power to run a bright LED.

Any way good luck and remember at the sizes we are working, I was assured by some aerodynamicists, that our blades are below their Radar so aerodynamic shapes were not all that important. 

Dick Parsons

[KWILParticipant @kwil]

The various SAR 5″ Narrow gauge locomotives (seen at  MEEX) had steam turbine generators working when seen on track at Guildford.

[Les Jones 1Participant @lesjones1]

Hi Paul,

               Some of the brushless  motors used on model aircraft rotate at very high speeds.

I think they are really three phase motors with a rotating magnet. One of these and a three phase bridge rectifier might meet your requirements up to about 50000 rpm but I do not think they would survive at 100000 rpm.

Les.

[Keith Wardill 1Participant @keithwardill1]

Hi, Paul,

 

Richard Parsons comments are very relevant – you need to sort out the turbine before even thinking about the generator. I have the original series of articles by Prof. Chaddock to which Richard refers, I think. (published in 1951 – sorry, Richard) – these detail all his experiences with all the aspects of turbine design, including how he cut the rotor, gearing, lubrication and (flash) steam generation. Despite the improvements in material technology since then, it makes for interesting reading. I could copy and e-mail them to you if you are interested.

 

Keith

[Paul HorthParticipant @paulhorth66944]

Thanks again, gents, for all your comments.

 

KWIL – I am a member at Guildford and I have seen the small turbine that you describe mounted on a loco. That is the kind of size and shape that I was thinking about, however I didn’t find out what was inside.

 

Keith – that’s a generous offer, yes please, I would be very interested to read the Chaddock series if it’s not too much trouble, I’ll gladly pay the cost.

 

Dick – I was considering the concept of drilling holes in the periphery of a rotor to form an impulse wheel as you describe, thanks for confirming that this does work. I take it that you form the passages from both sides so that they meet in the middle to form a reversed path from inlet to outlet side. I would have to lash up a dividing fixture on the vertical slide mounted at an angle to the lathe axis.

Fitting magnets into the disc is a clever idea. If I understand you, the poles would face axially so that the coil would be mounted on the face of the casing, and there should be two magnets on a diameter either side of the central shaft, to keep the rotor balanced? That at least eliminates the coupling to the electrical rotor. I’d have to learn how to make the coils.

 

Paul

 

[Paul HorthParticipant @paulhorth66944]

A little more information on this subject:

 

Ages ago, I gave away my 1980s Model Engineer mags, but not before I had made copies of the articles I wanted to keep. There were a couple of small steam turbines described, which might be of interest to others ( if you are not already ahead of me).

 

Terry Timms, A Steam turbogenerator,(mounted on Sweet Pea), 17 January 1986.

A. Sherwood, Tinkerbell, a 5 in. gauge switcher (with turboalternator), 2 January 1987.

 

Both of these used similar rotors, about 1 inch dia with Stumpf type cavities. They drove alnico magnets with external stators with coils wound, and could light up a couple of small bulbs. the jet sizes were small – 0.033 for Timms, 0.018 for Sherwood. So these devices worked on a small steam flow.

I was aware of these designs, but I had hoped to find a way to avoid winding the coils by using an existing electrical rotor. However I take some encouragement from their success and I might have to follow one or other of these designs.

 

Paul

[VersabossParticipant @versaboss]

Posted by Richard Parsons on 13/01/2011 07:39:36:
 

C.   Balance – Prof Chaddock wrote to me and said that he had to get the centre of rotation within 0.000002” the centre of mass. He did this with two razor edges some ‘bluetak’ and a small scraper.

 Hmm, had to misuse the internet again….

If my calculator and old brain are not completely off, the stated distance above is around 100 atom diameters (of the larger kinds, but his turbine was not built from hydrogen I think).

Quite impressive with a bit of bluetak. Unfortunately we can’t ask Mr. Chaddock how he measured it.

I formally apologize If I should be wrong…

Greetings, Hansrudolf

[Richard ParsonsParticipant @richardparsons61721]

 

Keith I sort of knew about Prof Chaddock’s work. I had a long chat at MEX when it was at Wembley. This lead to a long correspondence between, us mainly about nozzles. I had an idea that the low output could have been due to shockwaves developing in the throats of the nozzles.

In the end (with some help from some Hitile designers –well the Rapier missiles did not often miss). After lots of calculations on my PC I found the steam flow was almost certainly sub sonic.  So it was ‘Hay Ho’ and back to the drawing board.

Paul are you talking about ‘pea bulbs’ –which do not care if it is AC or DC-.   

[Richard ParsonsParticipant @richardparsons61721]

 

Hansrudolph sorry I missed your posting

How it was done (I used the same technique). You spun the rotor on a pair of knife edged (mine were Razor blades). You found the heavy side and put a smear of blutak on the opposite side of the wheel to make it heavy (er).  With a scalpel scrape some off the heavy side of the rotor.  If the blutak was heavy then you scrape a little of the blutak.  You spin the rotor after each process. Every now and again you scrub the rotor clean,dry it and start again. Repeat the process as required until the rotor’s stopping point is truly random.  Prof Chaddock, who was far more skilled than I, claimed to be able to get centre of rotation of his rotors within 0.000002” of the true centre of balance of the rotor. I got mine so there was no detectable vibration at 5000 rpm when spun with air. The guys where I worked were very helpful –they made gyros- and helped with the vibration tests.

[Stephen RowleyParticipant @stephenrowley13300]

Hi

I made this small
turbine some years ago. With 25psi it will produce 10000 rpm.

I had a hell of a job
trying to balance it enough so I did not have to screw it to the
bench.

The turbine it’s self
is 1.3/4” in diameter and has 8 blades.

It was made as an
experiment to drive a locomotive throw a gear box which it did quite
well but the gear box was far to big as I needed to get 10000 rpm
down to 60 rpm.

I have just tried it
driving a small DC motor that I got from the local model shop for
£6.50. it has an out put of 12v @ 3 amps @ 1000 rpm.

I connected then
together with a bit of plastic tube and put my meter across the out
put of the motor and apart from the noise it went well and lit a car
head lamp bulb until the plastic tube broke. I will pick the bits of
the tub out of the roof later.

James.

[Author]

Posts