Steam Engine Setup: Cylinder Volumes.

[Dr_GMJNParticipant @dr_gmjn]

All, I did ask this on my 10V thread, but I thought I’d ask here too.

I’m nearing the end of my build, and am making the piston, rod and cross head. The cross head obviously needs drilling for the connecting rod pin.

my question is: At mid throw of the crankshaft – ie the webs horizontal – should the piston be in a position in the cylinder that results in equal volumes above and below it? In which case it wouldn’t be in the centre of travel, but slightly higher in order to compensate for lost volume due to the piston rod in the lower chamber.

You may ask why not just follow the plan dimension for the cross head hole position and stop messing about. The truth is that since this is the first time I’ve done any model engineering, the dimensions of the individual parts aren’t spot-on, and the cumulative error may affect the volumes one way or another.

My plan was to calculate the vertical piston position for equal volumes, machine a spacer for the lower cylinder and temporarily assemble. Then with the crank webs horizontal and the connecting rod fitted, mark the cross head hole position.

Any thoughts welcome. Cheers.

[Dr_GMJNParticipant @dr_gmjn]

[Ramon WilsonParticipant @ramonwilson3]

Hi Dr G,

I would say you are perhaps over thinking this but as you say with the webs horizontal the piston will not be at mid stroke – which is correct.

Ignore the piston rod volume and keep to the drawing dimensional layout – your engine will work perfectly if the valve timing is right (which I'm certain it will be)

I see you also mention possibly lapping the brass/bronze piston – IMHO there will be no advantage to this. As long as the bore is smooth – lapped as you have done then a piston a thou down on bore diameter will be fine enough tolerance and an ideal running fit.

You're not far off that all important moment now – looking forwards to hearing you say it's running

Ramon

[Paul KempParticipant @paulkemp46892]

I think definitely a case of over thinking. Important bit is the top and bottom or front and back dead centres. There will always be slightly less power on the rod side of the piston because the piston has less area for the pressure in the cylinder to act on (less the area of the rod). At the point of admission the volume effect of the rod will be small, increasing as the piston moves but all the while the inlet port is open it is only the area effect that is important, volume will only come into play after the port closes and the steam is expanding (slightly less steam to expand compared to the other side of the piston) but you can't compensate for that! If I read your post right (and entirely possible I am missing something!) you are intending to set the stroke by drilling the cross head at mid point. If you do that how will will you gaurantee the piston does not hit the cover at one end or the other? Overall length of piston rod and con rod, (including the cross head) needs to be according to the total stroke and the length of the cylinder – best set at both ends.

Paul.

[John OlsenParticipant @johnolsen79199]

I would agree with the others that you are definitely over thinking this, especially for a small engine like the 10V. On the really big stuff they used to do a certain amount of fiddling with the gear to try to get equal indicator diagrams from each end, but this is not going to matter on a small engine. Steam engines typically have fairly long connecting rods in proportion to stroke anyway, so the effects of rod swing are small. It is more important to have clearance at each end of the cylinder at top and bottom centres. Ideally this clearance would be quite small, but you have to remember that wear in the bearings will tend to reduce the clearance, and it is not nice if the piston starts to hit either cover.

I built a 10V for my second engine many years ago, after making a simple oscillator for practice.

John

[JasonBModerator @jasonb]

Just aim for an equal gap between piston and top & bottom covers. I usually do this while the piston can still be removed from it's rod which makes it easy to adjust rod length before fitting and final piston skimming but you will just need to play with the shoulder position of the 5BA thread.

Edited By JasonB on 05/10/2020 07:02:47

[Dr_GMJNParticipant @dr_gmjn]

OK thanks guys that’s great.

[JasonBModerator @jasonb]

On a small engine like this the valve position will have far more effect on which side of the piston does the most work (assuming it ever gets put to work as most don't) as you can only adjust it's position by half turn increments of the nut so 0.011" at a time so chances of getting it exactly equal are slim. Therefore one side is likely to see steam earlier than the other.

On a larger engine you would tend to have nuts and lock nuts either side of the valve so get far finer adjustment of the valves position.

[Ramon WilsonParticipant @ramonwilson3]

Posted by JasonB on 05/10/2020 07:59:56:

On a larger engine you would tend to have nuts and lock nuts either side of the valve so get far finer adjustment of the valves position.

Exactly the same as having a 'plain' nut with a grubscrew in to set the valve infinitely and without having to disconnect the linkage to do so too.

Ramon

[Dr_GMJNParticipant @dr_gmjn]

Posted by JasonB on 05/10/2020 07:59:56:

On a small engine like this the valve position will have far more effect on which side of the piston does the most work (assuming it ever gets put to work as most don't) as you can only adjust it's position by half turn increments of the nut so 0.011" at a time so chances of getting it exactly equal are slim. Therefore one side is likely to see steam earlier than the other.

On a larger engine you would tend to have nuts and lock nuts either side of the valve so get far finer adjustment of the valves position.

Thanks Jason. I was also thinking about this. Despite the valve rod assembly being to the correct dimensions to the drawing, the upper spigot is only just located in the upper hole in the valve chest when at it's lowest position. I'm double checking all the dimensions of the standard, cylinder and caps, but one solution to get more engagement is to unscrew the rod from the forked end slightly, and make a brass collar to lock it in place (looks nicer than a nut). This will effectively also give me infinite adjustment on the valve slider position – unless I've thought it through incorrectly?

[Ramon WilsonParticipant @ramonwilson3]

Unscrewing the valve rod to give adjustment will as you say give 'infinite movement' but will also as you move it also move the valve relatively. The valve rod should be 'fixed' – any adjustment should come at the valve itself. As long as the rod engages the hole at the bottom of the stroke it should be sufficient. If in any doubt shorten the rod at the top and do away with the engagement – it will run just as well without.

Just trying to be helpful here Doc.

Ramon

[Dr_GMJNParticipant @dr_gmjn]

Posted by Ramon Wilson on 05/10/2020 09:45:14:

Unscrewing the valve rod to give adjustment will as you say give 'infinite movement' but will also as you move it also move the valve relatively. The valve rod should be 'fixed' – any adjustment should come at the valve itself. As long as the rod engages the hole at the bottom of the stroke it should be sufficient. If in any doubt shorten the rod at the top and do away with the engagement – it will run just as well without.

Just trying to be helpful here Doc.

Ramon

Thanks Ramon – I really hope I'm not giving the impression I'm not grateful for all the answers and input – I certainly am.

All I was thinking was that moving the rod up would engage the top slide further, and also give the option of infinite adjustment. Of course unscrewing the rod from the fork also unscrews the valve plate from the rod (both move upwards), but as you said, the plate can only be effectively be adjusted 1/2 turn at a time.

I think you previously suggested a grub-screw fixing for the plate, which makes perfect sense. The sole reason I didn't use it was becasue I wanted to prove to myself I could work to the drawing for this first build anyway!

Cheers!

[SillyOldDufferModerator @sillyoldduffer]

Interesting question. I went to my copy of Youngson's "Slide Valves and Valve Gearing", 4Ed, 1932, which goes deep into the detail, and I'd expect a mention if this was significant. The book is aimed at sea-going marine engineers studying for a Board of Trade Certificate, and although mainly theory there's a good dollop of practical as well. The book covers all the engines likely to be found on a ship; compounds down to simple single cylinder donkey engines.

GMJN's issue isn't mentioned. The book goes into friction, lap, lead, travel, cut-off, con-rod obliquity, power, cushioning, steam distribution, consumption, setting and other consideration, but not that one.

My guess is the effect is insignificant compared with everything else going on and is compensated within normal valve settings as other posters say.

I think the main effect would be the loss of piston area rather than loss of energy volume due to the piston rod taking up space inside the cylinder, particularly as the latter is tiny when the stroke starts and only reaches a maximum after the steam has finished working on the rear piston face.

Assuming a 300mm diameter piston with a 50mm diameter rod, the available rear piston area would be reduced by about 11% compared with the front. Sounds like it's worth fixing except the cure doesn't achieve much. Balance is achieved at the expense of the other side of the piston, so overall power output stays the same. Power output might be reduced, because oomph is stolen from the best half the piston to balance the other.

The piston rod must cause another power reducing problem. Inside the cylinder it absorbs heat that would otherwise be converted into useful work. Then the rod slides out of the cylinder into the open air, where it cools down. The naughty old piston rod pumps energy out of the cylinder – waste – and only one side is effected.

But it's just one of many reasons steam engines are inefficient. A good full size steam locomotive is only about 5% efficient and for reasons of scale small engines are worse. A really good one might get 2½%, most model engines rather less. When 95% of the energy input to an engine is wasted, it pays to concentrate on big problems rather than small ones. I suggest the Stuart 10 piston being unbalanced isn't worth fixing because the improvement has no effect on efficiency, or a useful reduction of vibration.

Vibration is another interesting problem in reciprocating engines, and maybe GMJN's issue was tackled formally when the Victorians attempted to build the high-speed engines needed to generate electricity. I don't know.

Dynamos are efficient only when driven at high rpm, and gearing up ordinary steam engines wasted lots of power in an unreliable gearbox. Many builders attempted high-speed engines fast enough to drive Dynamos directly and they all vibrated badly because big reciprocating steam engines are almost impossible to balance due to masses of heavy metal being rapidly accelerated up and down. As vibration is related to the square of the speed, doubling rpm quadruples vibration.

Early power stations were an outright public nuisance: how bad can be judged by London County Council's power station. It caused enough vibration to derange Greenwich Observatory's mercury horizon over a mile away. Nearby houses had plates clattering in on the dining table, pictures falling, and walls cracking. The Law was soon changed to limit the vibration a power station could cause.

Much effort was put into reducing vibration and might well have addressed GMJN's concern, but research ended abruptly when steam turbines provided a better answer. But perhaps there's something in the engineering literature circa 1900 that answers the question properly. I'm joining the dots and could be wrong.

Dave

[Ramon WilsonParticipant @ramonwilson3]

Not at all – just thought you might be missing a point.

Your principle of working to the drawing first off is without question the right approach but there are always areas where small improvements can be made over that shown – if desired. What you have done so far for a first engine is highly commendable and the results certainly convey that. I've seen your plastic modelling and your attention to the finished item certainly has been brought to this build and to the coverage of it.

The tip for using a grub screw came from 'KBC' – George – over on MEM. Previous to that I had done as you have – nut and threaded rod but as soon as I tried Georges sugestion the immediate ease of setting the valve became readilly apparent. It's a tip worth passing on , but as I often say to beginners in our plastic group (and yes we have had one or two!) 'Listen' to all but make your own mind up what you want to do and stick to it. Perhaps paradoxical in this case but fundementally the right approach in my mind.

Many many single and double 'Tens' have been built in their various forms over just as many years – it's a sound design and, if my 'experience' is anything to go by will give you an immense sense of satisfaction when it breaks into it's first run and sits there steadilly turning over. You'll watch it for hours

Regards – Ramon

[Dr_GMJNParticipant @dr_gmjn]

Thanks very much everyone. I’ve been measuring and setting up this evening. After drawing everything in Autocad to my measured dimensions, I’ve ended up with about +0.1mm different (too little) from the drawing on clearance height top and bottom. This is obviously a function of crank throw, but ty he crank was a bit tricky to make, so I can’t complain.

I’ve set the crank pin hole in the slider to give equal clearances (0.7mm) top and bottom. Hole position turned out spot-on to the drawing.

I guess all my inevitable machining errors combined to almost cancel each other out in this instance…

[duncan webster 1Participant @duncanwebster1]

A really neat method of adjusting valves is to have a different pitch thread at the clevis and the valve nut. Then rotating the rod moves the valve in a sort of vernier fashion. Once it's right tighten a locknut at the clevis. If you had say 40 tpi and 32 tpi one rev of the rod would be 1/32 – 1/40 = 0.00625".

Having said that I use the plain nut and grub screw as described by Ramon. Big marine engines were sometimes set with different events top and bottom to counteract the heavy moving parts. Not really neede at our scale

[JasonBModerator @jasonb]

The other simple way is to have a reduced diameter on the rod that slips into a notch cut into the valve nut. Then just use the thread where the rod joins the clevis and a lock nut to make final adjustments. Saves having to take the valve chest cover off to get at a grub screw

[Dr_GMJNParticipant @dr_gmjn]

Re. the valve setup itself – I saw somewhere that you should set the valve slider to move equally up and down, then set the piston to tdc and have the upper port begining to open a few degrees before.

Is that correct? If so, approximately how many degrees before tdc should the valve crack open at?

[JasonBModerator @jasonb]

Stuarts setting instructions

I usually go for 20-30 degrees when running on air so with piston at TDC the max throw of the eccentric wants to be at 4 or 8 o'clock depending on which direction you want it to run. Or if it is easier to get at the eccentric's grub screw when the max throw is at the top then do it that way with crank at 4 or 8 o'clock.

Edited By JasonB on 08/10/2020 13:14:28

[Dr_GMJNParticipant @dr_gmjn]

Thanks for the info.

[Author]

Posts