Aluminium Pistons

[NickM]

Good Afternoon Gents,
Could anyone enlighten me on the pros and cons of using aluminium pistons in a steam loco? They seem to work exceptionally well in internal combustion engines, any reason for not using them in a steam engine?
I look forward to your replies.
NickM

[millman]

Only reason I can see not to use them is the expansion when hot. Aluminium will expand more than cast iron and I believe more than gunmetal so if made a good sliding fit when cold they will seize out when hot.

[oldnorton]

Aluminium pistons are essential in a modern (post 1920's) internal combustion engine. They have to be light enough to achieve the higher RPM than we need for a reciprocating steam engine. Cast iron does a better job in cast iron bores, because it has matched thermal expansion, and will not suffer the electrolytic corrosion of aluminium in contact with wet iron. Yes you could make aluminium work in a cast iron steam engine bore, but why use it if it affords no benefit?

[uuu]

I knew a chap put aluminium pistons, with cast iron rings into gunmetal cylinders in his Speedy. We expected the worst, but it did seem to work OK.

Wilf

[andyhigham]

I have a Villiers 2 stroke stationary engine. These were available with cast iron or aluminium pistons. The aluminium piston requires an fuel/oil ratio of 16:1, the cast iron piston required 32:1.
Cast Iron is an exceptional bearing material, and works well cast iron on cast iron unlike most other materials

[cfmrc]

I had a titanium piston in my single cylinder 4” agricultural traction engine, with cast iron clupet rings. It reduced vibration - there was no balancing on the crank. Road running was therefore rather better.

Tim

[penybontshed]

You beat me to it Mr M!

[jasonb]

Popular on traction engines for the reason mentioned above of reducing the recipricating mass, as most of the full size pistons were hollow iron castings it make sense to use a lighter material if just having a solid piston. I use them all the time on my engines.

IC engines will get as hot or hotter than a steam engine's so expansion is not an issue unless you make things too tight to start with.

[92220]

It would be a problem with aluminium pistons and cast iron cylinders, whether a steam engine of IC engine. Whether it works or not is down to the clearance between the piston and the cylinder. On IC engines, the seal is not made with the piston, it is made by the piston rings, which are often cast iron. There are some IC engines that are aluminium and fitted with cast iron cylinder liners, Like the Hillman Imp, and there have been IC engines that had aluminium cylinders and pistons. These had some sort of plating on the inside of the cylinder, for the rings to seal against and not gall the cylinder surface. I don't think they used this system often though. I think it was experimental for motor racing. Just as an example of aluminium cylinders and pistons, look at the Hillman Talbot Imp. This was taken from the Imp design spec:-

Engine
The standard 875 cc block is an aluminium casting with ribbed liners cast in position. Unlike many blocks, the Imp block is not split at the main bearing centre line, but is of the deep skirt configuration, where the sump flange of the block drops well below the crankshaft centreline. The block has three main bearings and the caps, which are themselves aluminium, are secured to the block hi high tensile steel bolts. Early blocks suffered from lack of stiffness, and were not ideal for tuning. Later blocks, from 1965 on, were much stiffer, and could be used for competition.
There are differences to the Mark 1 and the Mark 2 engine blocks. On Mk1 (or curly edge) blocks there is room for 'movement' due to the top deck being a little flimsy. This can lead to gasket failure.
On later blocks (and heads) the castings were thicker and generally the quality was better. For tuning, use the Mk2 engine. The largest size valve that can safely be fitted into the Mk1 head are those fitted standard to the Mk2.
Retooling for the straight edge block must have been very expensive, and Rootes would not have spend money like that without reason. The Mk1 block assembly is too thin in all the wrong places to convert to 1000cc.
Pistons
Pistons are cast in LO-EX aluminium alloy and have two compression rings and one oil scraper ring. The piston runs in a nominal bore size of 2.6770 in. dia and the piston-to-bore clearance, measured at the skirt is, depending on tolerances, between 0.0017 in and 0.0010 in. These measurements are taken at right angles to the gudgeon pin hole.
Cylinder Liners
Centrifugally cast in iron.

Bob.

[jem]

Out board motors ic had aluminum pistons and cylinders, the cylinders being chrom plated, others had cast iron liners.

Jem

[andyhigham]

My Race bike, and my BSA both have aluminium cylinders, the bores are plated with a "Nikasil" coating
Having the cylinder and piston made from the same material allows closer piston to bore fit.
Piston In my race bike, 90mm dia
00501-GM-piston by Sigma Projects, on Flickr

[mikessme]

I have dural pistons with cast iron rings in a Modelworks Black 5 and they work fine. The previous issues I always encountered with stuck cast iron pistons in cast iron cylinders after resting over the winter despite removing as much condensate as possible after the last run of the year and putting some oil in have completely disappeared. A few pushes up and down the steaming bay to expel remaining condensate after it has cooled is now all that is necessary.

[andyhigham]

A thought on pistons.
Unlike an IC engine where the piston is restrained axially by the cylinder bore and any excessive clearance will lead to "piston slap" accelerated wear.
The steam engine piston is restrained axially by the crosshead and cylinder cover bearing. The piston clearance can be huge without ill effects as the piston never touches the bore.

[don9f]

On the BR Standards and other types, there is no actual bearing in the cylinder rear cover, just a clearance hole and then the gland that is free to “float” and follow the piston rod. The 9F I am familiar with for example had a bronze slipper at the bottom of the piston between the two rings, backed by springs and this (in theory) took the weight of the piston to help it run up and down centrally in the bore in a straight line. In practice, these things wore quite quickly and the pistons then probably made contact with the bottom of the cylinder liner. The allowable difference between the piston diameter and the 20 inch cylinder bore is nominally 3/16 inch.

Cheers Don

[lesstoneuk]

Aluminium expands about 21 to 24 microns per metre per degree c. On this basis, you can work out how much smaller the piston needs to be before binding will occur at the operating temperature of the cylinders. Steam at 90psi is 166°C. Taking room temperature as 20°C,thats a rise of 146°C. The increase in the circumference of a 25mm diameter piston would be 87.6 microns. That's about 28 microns on diameter.
Someone please check my workings as I've a habit of transposing number.
Obviously the temperature at the cylinders won't be 166°C due to wastage but you get the drift.
Oh gunmetal expands at 18 microns per metre and cast iron is 11 microns per metre.

[oldnorton]

Hi Les

I like it when anyone works out numbers, it does lure me in as I try to understand.

Now my favourite 'fact' for the thermal expansion of metals was when I had to determine what was going on with a stainless steel plunger reciprocating in a bronze bush on a racing motorcycle. If it was tight cold then what was the clearance when hot? My easy to remember figure was that the DIFFERENTIAL expansion between a steel (all steels are similar enough) and a bronze (all copper alloys are similar) was one thou per inch of diameter per 100 degC.

Of course, your numbers teased me to compare our figures. 1mm is 40 thou (appx.) so 10 microns is 0.4 thou, you have a differential of 18 - 11 = 7 microns per 166 degC, about 0.3 thou, per 25mm (about an inch). Whereas my numbers for bronze/iron are 1 whole thou per 100 degC per inch diameter.

We differ by a factor of four or five so we need a third person (or more) to check my maths from 15 years ago as well! although the trickiest thing is reading the tables of coefficients of expansion and, as you say, moving the decimal point lots of places.

Aluminium having a greater thermal expansion than bronze (your figures 28 versus 18, I cannot recall mine), I would have thought to allow at least 2 thou per inch of bore for an aluminium piston in cast iron. That would feel quite slack.

Norm

[lesstoneuk]

Jan 10, 2022 11:05:59 GMT oldnorton said:

Hi Les

I like it when anyone works out numbers, it does lure me in as I try to understand.

Now my favourite 'fact' for the thermal expansion of metals was when I had to determine what was going on with a stainless steel plunger reciprocating in a bronze bush on a racing motorcycle. If it was tight cold then what was the clearance when hot? My easy to remember figure was that the DIFFERENTIAL expansion between a steel (all steels are similar enough) and a bronze (all copper alloys are similar) was one thou per inch of diameter per 100 degC.

Of course, your numbers teased me to compare our figures. 1mm is 40 thou (appx.) so 10 microns is 0.4 thou, you have a differential of 18 - 11 = 7 microns per 166 degC, about 0.3 thou, per 25mm (about an inch). Whereas my numbers for bronze/iron are 1 whole thou per 100 degC per inch diameter.

We differ by a factor of four or five so we need a third person (or more) to check my maths from 15 years ago as well! although the trickiest thing is reading the tables of coefficients of expansion and, as you say, moving the decimal point lots of places.

Aluminium having a greater thermal expansion than bronze (your figures 28 versus 18, I cannot recall mine), I would have thought to allow at least 2 thou per inch of bore for an aluminium piston in cast iron. That would feel quite slack.

Norm

All my figures were from a simple Google search. My days of working out the coefficient of linear expansion for aluminium was over 50 years ago for 'O' level Physics. That was in the days of a gruelling 4 hour written exam. Ah... Those were the days... Lol