By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by Allendale Jan 24th

Steam engine from scrap metal

I had a block of brass but no mill, no casting, and want to design my own steam engine

All Topics | Latest Posts

Search for:  in Thread Title in  
Evan Lewis17/11/2020 12:44:07
23 forum posts
10 photos

Hi everyone,

I have already made the cylinder block for a two cylinder, double acting steam engine. I have just read through "Steam Engine Number One" and found that and many other posts useful. And I have watched just about everything I can find on YouTube and read Wikipedia. I am doing this whole project with a lathe and decided to use piston valves which came into fashion about 1900 instead of the slide D valves used before that.

I am not an engineer because I failed to get into engineering school and became a doctor instead sad. My father taught engineering at Te Puke High School in New Zealand and I was in his class for 4 years doing engineering theory, workshop practice and technical drawing in the early 1960's.

When Dad died recently he left me a lot of tools and a lathe that he bought from the school: A 1955 Boxford Model A. I now have that set up in my car garage. I do have a propane-oxygen welding torch. It is adequate but there is no spare space for more equipment. I do not have a compressor but think I will need one to test the steam engine. I use an old propane tank flushed out and filled with compressed air at the petrol station (NOT recommended). I have made about 32 YouTube videos about how to use and overhaul the lathe. (under the name evan-e-cent or evanecent : see Playlists on steam engines and using an engineers lathe).

My first project was to build a Hero steam engine (designed by the Greeks 2000 years ago) and calculate its power output and efficiency documented at (there are some numerical errors in the text shown on the video.)

On the same web site I wrote software for working out what gears to use in the gear train to cut any thread, metric or imperial, and to calculate feed rates (see

I built a model of a gold-mining quartz stamper battery and that worked out very well, but is not on YouTube yet. (see my wife's web site for a video of a small battery). The plan is to possibly use the steam engine to drive this battery.

Cylinder block for a 2 cylinder steam engine made in a lathe.


Edited By Evan Lewis on 17/11/2020 12:54:15

Evan Lewis17/11/2020 12:45:15
23 forum posts
10 photos

I went to a scrap metal yard in TN, USA and they had a solid round brass bar 75mm diameter by 2 meters long in the scrap! I bought about two feet of it (600mm). Also a rectangular bar 50mm x 38mm and I bought about 600mm of that too. It was only $2.80 per pound. I cut 150mm off each and brought it in my hand luggage back to NZ in January. What to do with it! I used up the round bar making involute spiral cams for the drop hammers in the stamper battery and produced huge amounts of swarf!

My Mac is so full of stuff that I didn't have room to install CAD software so I just drew rough sketches freehand with a blunt pencil on the knee. Too shocking to publish!

Well I decided to build this steam engine using the rectangular block. I drilled 3 holes across the center so that I could bolt it to the crompound slide. Put a quarter inch steel plate underneath to center it and used a dial gauge to line it up. I tightened two of the gib screws to clamp the compound and cross slides. This meant that I could bore all the holes on the same plane. I made a boring tool holder for the 4 jaw chuck and used it to bore two main cylinders 22mm diameter with a planned stroke of 36mm and piston length of 10mm.

I bored two cylinders for the valve pistons which ended up 20mm diameter after making the mistake of boring before drilling the port holes which left terrible burs in the cylinders. So I had to bore them again. I wrote in my notes "drill before boring", but with weeks passing between stages I forgot about that!

I drilled steam galleries parallel to these cylinders placed between each main cylinder and valve cylinder, but not passing right through, and drilled similar holes on the other side part way through leaving metal blocking the mid-portion.

The plan is to make heads that duct the steam from these galleys into the main cylinder heads. That way the piston can reach the end of the cylinder without getting fouled up with a port.

I have a good drill press and used that to drill vertical holes from the top of the block - one near each end of the valve cylinder to act as steam inlet ports, and a single hole at the center as a steam exhaust port. This should be bigger as the steam expands so I drilled right through the block so that steam can exit from both the bottom and top of the block. Again I could have drilled holes side by side to act like a slotted port, but I am concerned that all these ports may damage the piston. I was going to make the pistons from bronze but perhaps I should use stainless.

Edited By Evan Lewis on 17/11/2020 12:58:33

Evan Lewis17/11/2020 12:45:32
23 forum posts
10 photos

The valve piston works exactly the same way as a D slide valve and the cross section of the piston assembly is exactly the same as the cross section of a D valve. It consists of two pistons, each 9mm long with only 5mm space between them. Steam galleries 6.5mm diameter were drilled from the ends of the cylinder block, straight across the valve cylinder and into the steam gallery described above. Then the holes in the end of the block were tapped and blocked off with 8mm stainless steel grub screws. That is why the 6.5 mm drill was chosen.

The biggest flaw with this design is that the steam galleries are too small. And the large mass of brass will take a long time to heat up resulting in condensation. A third problem is that I have read that you should not use brass with steam because it will corrode. I am not sure how quickly that occurs and wether it is really a problem for a model.

The diameter of the ducts is restricted by the limits of the valving system. I was surprised at how precisely this has to be made so that the inlet and exhaust valves do not open at the same time. Usually slots are used to increase the steam flow. At one stage in the design process I was going to have two sets of these galleries for each piston but forgot about that when I actually came to make it! See photo. Since the holes are 6.5mm and the valve pistons 9mm there is 2.5 mm total overlap giving 1.25 mm steam lap and 1.25mm exhaust lap but i have not been able to find any documentation on how to design the lap. It may be too much? When the main piston is at the mid-point of the cylinder the valve pistons occlude with inlet and exhaust ports.

I have not decided on rings for the pistons. I was just going to manage without rings but that may not be wise, so I think I will use O-rings but could use some tips and tricks to work out how deep to cut the grooves. Should I have rings on the valve pistons - they would have to pass over numerous port holes and might get torn up? Also the valve pistons would ideally be sealed along their whole length.

Also I have not worked out how to seal the glands that the con rods pass through. O-rings here too? Placed inside the bearing surfaces?

So now I plan to bolt flat plates on each end of the brass bar as a frame with a brass plate for the floor screwed onto the under side of the walls. These walls will support the crankshaft. Then I will place additional 3 walls between the cylinders and mount brass bars 5x20mm across the tops of the 5 walls to make a T shape 100 mm long for the cross heads to slide on.

The cross-heads will be about 10mm thick (is weight a concern?) with grooves milled down the side. I do not have a mill but have a milling attachment for the lathe. (Otherwise it could be made from 3 plates screwed together.) The connecting rod will screw into the crosshead and the "swing rod" which connects to the crank shaft will require a wrist joint.

I plan to have double eccentrics for each cylinder so that they can be connected to a C-shaped slot controller to provide forward and reverse and variable power. But in the first instance I will just use one of them to make it a bit simpler until I am sure it will work.

I do have some scrap bronze and all the bearings and cross heads will be made from that.

I also have some interesting scrap I got from a scrap dealer in NZ. This is 80mm diameter solid round bar about 500mm long which he gave me. It was attached to a plate 100mmx10mm and the same length and I was able to remove it. It appears to be very hard magnetic stainless steel but the Menz Shed has an electric hacksaw which will cut off disks.

At first I assumed I would be able to make the crankshaft out of a single block but with 36mm swing it is too big. I can just fit the 80mm bar into my chucks but cannot set an offset to turn the bearings for the big ends. So I have decided that I will have to cutoff and face 3 disks and drill holes to place shafts for the big ends. I do not know how they will be kept in place. Probably pins drilled across the holes. Any suggestions here would be helpful. I think that threading these joints would not be sufficiently accurate or rigid. Can it even be done?

Any advice that you can offer would be greatly appreciated!

Ady117/11/2020 12:52:47
4669 forum posts
713 photos

If you upload pictures into here you will get a better response because folk can see whats going on


Evan Lewis18/11/2020 04:21:32
23 forum posts
10 photos

Thanks for the tip GL. The brass block shown in the photo above is the only part that I have made so far. The rest is in my head and some rough sketches. Since you requested visual representation I have added my sketches here but they are VERY rough.

This is a photo showing the cylinder boring process with the block of brass (covered in engineers ink) mounted using 3 bolts on the compound slide.

Incidentally you can see the DC treadmill motor I just installed with variable speed control 2.5 HP at 4700 RPM motor speed.

A photos of the boring bar holder in the 4 jaw chuck for boring cylinders

This 3D sketch shows the cylinder block, two main pistons near the center and two valve pistons towards the sides. Steam ducts are shown for the piston on the right. Intake and exhaust galleries are not shown here.

Rough sketch of planned engine design #1

Sketch #2 includes the cross head and crankshaft. It top left I sketched two types of cross-head slider and plan to use the design to the right. Probably made of bronze.

Rough Sketch

The following is an image I took from an animation in Wikipedia and this has been the basis of my design. I altered the slide valve to make it look more like a pair of pistons. The gap in shaft shouldn't be there! I said the valve pistons would be 180 degrees out of phase with the main piston, but this is adjustable and an approximation.

Rough Sketch #3

This is the end view showing the "walls" between all 4 cylinders with a brass bar on the top to act as slider surfaces for the cross-head. The 4 cross heads will be about 25mm square and 10mm thick with grooves cut down the edges to run on the brass slider.

Rough Sketch #4 - End View of the cylinder block with slide rails

Plan view showing the walls and the brass sliders will be mounted on top and screwed to the walls. I have reduced the thickness of the walls to 5mm aluminum bar.

Rough Sketch #5 Plan view

Evan Lewis18/11/2020 05:55:31
23 forum posts
10 photos

I just drew another sketch of the steam ducts and galleries which I have already drilled in the cylinder head shown above.

I have also realized that if I obtain a 50mm diameter steel bar I can easily achieve the 18 mm offset I need with my 5" 5-jaw chuck. So i will be able to make the crank shaft from a single piece.

Sketch #6: The steam ducts and galleries that have already been drilled into the cylinder block.

Turbine Guy18/11/2020 19:51:38
387 forum posts
219 photos

Hi Evan,

I can't find any way to view your photos or sketches. If you can make the photos and sketches into JPG files and add them in your album. You can add them to a post by clicking on the camera icon and selecting the file you want to show. They will appear like the following example.Solid Model of Chiltern

Neil Wyatt18/11/2020 21:42:30
18722 forum posts
729 photos
80 articles

Odd... the photos and sketches are album images and I can see them OK.

My observation is the valve needs to be 90 degrees (plus say 10 to 20 degrees advance) out of phase with the piston, not 180 degrees.


Michael Gilligan18/11/2020 22:36:08
18700 forum posts
912 photos
Posted by Neil Wyatt on 18/11/2020 21:42:30:

Odd... the photos and sketches are album images and I can see them OK.



Odder ... I can see the images but, according to the sidebar, Evan doesn’t have an Album on this site.


Ady118/11/2020 22:47:27
4669 forum posts
713 photos

He's uploaded them to his own site and linked from here that's all

spooky magic

Edited By Ady1 on 18/11/2020 22:53:08

Evan Lewis18/11/2020 23:35:40
23 forum posts
10 photos


VERY rough sketches.

Steam galleries:






Side walls and internal "Walls" which support brass sliders (20x5mm brass bars) centered on the cylinder axes:






Showing the walls and slider bars




Rough 3D Sketch

Possible cross-head sliders shown at top left. 



3D Sketch - an early version




PHOTO of the brass block covered with engineers blue, mounted with 3 bolts on the compound slide of the Boxford A lathe. This uses a purpose built boring bar holder in the 4-jaw chuck. The position of the boring tool is fairly easily and accurately adjusted with a dial gauge.

Note: You can the motor I installed recently. It is a 180 volt DC motor out of a treadmill with a Chinese pulse width modulator HQ-SXPWM-X speed controller.  It is very nice to use!





PHOTO of the finished brass cylinder block. That is all I have made so far.



Yes it seems spooky and a bit weird for Newbies. I could not see how to make an album which was required for the camera icon above, but the icon next to it allowed me to link to photos on my web site and they showed up on my posts.

Anyway I finally found my way to albums and created one by uploading the same photos. For those having trouble seeing them I will display them again using the album!




Edited By Evan Lewis on 19/11/2020 00:11:25

Evan Lewis18/11/2020 23:50:49
23 forum posts
10 photos

Thanks Neil for the timing tip. I knew I had to find out the exact phase difference. It will be adjustable by moving the eccentric cam on the crankshaft which will be held in place by a grub screw. So 90 degrees plus 10-20 degrees advance will be the starting point.

I assume the advance is to cushion the change in direction as the piston approaches top dead center?

I have read about steam lap and exhaust lap, but not sure what effects they have. Not sure about the role of "lead" either. Do the following numbers seem reasonable?

The ports are round, 6.5mm diameter (originally planned to be 8mm) and the valve pistons 9mm long by 20mm diameter. If these pistons are placed so that they are centered over the two inlet ports when the valve piston assembly is in the mid-position, this will give symmetrical laps of (9-6.5)/2 = 1.25mm. ie steam lap=1.25 and exhaust lap=1.25 with stroke length 24mm.

I have not made the piston assemblies yet. If I put O-rings on the valve pistons it might affect this behavior if the pistons are not a tight fit. Should I use O-rings. If not the design will be subject to decreasing efficiency with wear.

I plan to have two eccentrics for forward and reverse operation with a sliding control added later.


Edited By Evan Lewis on 18/11/2020 23:55:13

Evan Lewis18/11/2020 23:57:49
23 forum posts
10 photos

I have the option of making the pistons from bronze or stainless steel. They will of course be running in the brass cylinders. Which material should I use?

I also have 1/4 inch brass rods for con-rods but I see model kit sets often use stainless. I am thinking of turning down to 6mm so that I can use a 6mm reamer for the con-rod bearings.

PatJ19/11/2020 03:56:42
132 forum posts
104 photos


I checked out some of your online videos.

Pretty interesting material, as is your engine build.

And if I am understanding you correctly, you are in TN, in the USA, and if so, then we must meet up and perhaps cast a few engine parts in iron in the backyard.

Pat J

JasonB19/11/2020 07:21:14
21294 forum posts
2416 photos
1 articles

To me your valves and passages are massive compared to the cylinder diameter, if running on steam it's not going to work too well and drag from those O rings will be high. I have done 24mm dia cylinder engines with 6mm dia piston valves and 3mm passages, may be worth sleeving down those big valve holes.

Evan Lewis19/11/2020 10:41:16
23 forum posts
10 photos

Thanks for your advice Jason. I am surprised because I thought these ports and ducts were too small and didn't imagine it would be much of a problem if they were too big. What happens with the large size? Does it have difficulty maintaining pressure with the large volume.

You mentioned specifically running on steam and that suggests to me the problem may be caused by the steam cooling down, and the huge mass of the cylinder block wouldn't help with that. Yes the large passages would slow the steam passage. But the pressure seen in the passages would be the same as the pressure produced by the boiler if the passages are large enough (?)

If it is a volume issue I suppose I should insert sleeves in as many areas as possible? I can't change the ports themselves as that would require re-boring the valve cylinders yet again.

The valve cylinder size was supposed to be half the final result of 20mm. I had some manufacturing problems. I did not figure out why, but on one occasion the cylinder turned out larger at one end -cone shaped. The boring bar must have moved. I bored the cylinders and THEN drilled the ports leaving serious burs inside so I had to bore it twice more. Again I did not think that the passages being too big would be a problem.

Yes friction may be an issue, especially with O-rings. I may leave the valve pistons plain and even the main pistons too -depending on what advice I get from you and other experts. Of course I can always take it apart and add O-rings later.

Thanks again.

JasonB19/11/2020 12:01:56
21294 forum posts
2416 photos
1 articles

It's more to do with expansion, after the exhaust stroke those passages will be at atmospheric pressure so as the fresh incoming steam enters them it will expand and loose some energy before it gets to the face of the piston. Ideally passages should be of sufficient area to supply the required load and any dead space kept to a minimum including gaps between main piston and end covers.

Looking at your inlet, exhaust and passage sizes and comparing them with a similar bore Stuart 10 series engine they use 5/32" inlet pipework which would have approx 3mm bore and 3/16" exhaust with a bore of just under 4mm which is considerably less than the 8mm & 12mm OD of your engine.

Likewise the 18mm diameter of the crankshaft is probably a good 10mm more than it needs to be creating more friction in the bearings and the eccentrics needing to be big enough to fit such a large shaft will also have increased friction for the size of engine.

Piston rods (Conrods are what attach to them) are usually done in stainless and 4-5mm diameter would be well upto the job. I tend to go for graphite yarn to pack around the piston rod and valve rod, though O rings can be used. a single 2.4mm section x 19.4mm ID metric viton O ring in a groove 2.3mm deep and 2.6mm wide will seal the main pistons well and not produce much friction. Valves I would just lap into their holes.

It's what is in the boiling water in the boiler that can affect brass not the actual steam so engine parts will be OK

Turbine Guy19/11/2020 14:23:08
387 forum posts
219 photos

Hi Evan,

Thanks for adding the photos and drawings from an album. I can see them now. My computer is quite old and I may not have the apps needed to view them the way you originally posted them. I agree with Jason B that your valve diameter is quite a bit larger than optimum. The two steam engines I own with piston valves are a Chiltern and a Saito. The Chiltern has a 14mm piston diameter, a 7mm valve diameter, and relies on a close fit for sealing. The Saito has a 8.8mm piston diameter, a 3.8mm valve diameter, and relies on a close fit for sealing. Both of these engines have relatively low leakage from the valves.


Iain Downs19/11/2020 17:03:27
771 forum posts
696 photos

Welcome to the ranks of home brew engine bodgers!


(Steam Engine Number one. Some day it will be done. It may even run!)

Evan Lewis20/11/2020 02:31:55
23 forum posts
10 photos

Thank you Jason and Byron for these useful tips with such great detail. I should have checked with you before building the cylinder block. Now I am stuck with my ignorant decisions!

In conclusion:

  • The valve cylinder diameter should be about half of the main cylinder diameter.
  • The valve piston should be lapped in and no O-rings used.
  • The main piston could have O-rings as specified by Jason. No preference for stainless, brass or bronze?
  • The bearings for the piston rods could also have O-rings.
  • I plan to put a cap on the back cylinder heads to accommodate the piston rod as it protrudes. I think it will need to have an internal diameter larger than the rod, otherwise it will work like a syringe, creating pressure.

I plan to tap the biggest passage which runs parallel to the cylinders and fill it with threaded brass. Then re-drill the passage 3-4mm diameter, and closer to the main cylinder so that they will be covered properly by the head. My main concern is that using a small drill close to the cylinder may go off course and curve around into the cylinder wall. This is a particular risk if I am drilling through the edge of the threaded plug. I do have a small reamer. Any suggestions there?

Thanks again!

All Topics | Latest Posts

Please login to post a reply.

Magazine Locator

Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!

Find Model Engineer & Model Engineers' Workshop

Support Our Partners
Eccentric July 5 2018
rapid Direct
JD Metals
walker midge
Eccentric Engineering
Subscription Offer

Latest "For Sale" Ads
Latest "Wanted" Ads
Get In Touch!

Do you want to contact the Model Engineer and Model Engineers' Workshop team?

You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.

Click THIS LINK for full contact details.

For subscription issues please see THIS LINK.

Digital Back Issues

Social Media online

'Like' us on Facebook
Follow us on Facebook

Follow us on Twitter
 Twitter Logo

Pin us on Pinterest