Clarkson’s Horizontal Compound

[John PurdyParticipant @johnpurdy78347]

Finally finished the Clarkson's horizontal mill engine. Ran each engine on air individually at about 250 RPM to bed them in, then connected them up as a compound. As I suspected it required considerably more pressure to get the same speed in the compound configuration. Individually, at 150 RPM, the HP required 4 PSI and the LP 3 1/2 PSI. But in compound it required 9 PSI. Also the lowest sustainable speed singly was just under 100 RPM but in compound it is 135 RPM.

The engine was built using a set of original Clarkson castings and drawings I acquired many years ago. I found some of the castings had no machining allowance at all. The worst was the main bearing pedestals on the bed plates. They are to finish 1 1/16" above the cyl. mounting pads but the rough castings were less than 7/8" above. This necessitated a complete re-design of the main bearings. The others were got around by adjusting some of the dimensions, with the consequent problem of then having to adjust others.

[John PurdyParticipant @johnpurdy78347]

[Jeff DaymanParticipant @jeffdayman43397]

Hi John, great job on the engine. The colour scheme is great too. Re pressure / compounding – could the pressure requirement be reduced by playing with valve timing on the secondary / compounded cylinder? Just a guess that it may be tune-able. Apologies if you have already optimized that.

[Phil PParticipant @philp]

Hi John

Nice job on the engine, I like the colour scheme.

Not a very common model these days, but I have a set of castings for one myself.

Unless the HP exhaust has got somewhere to go when it expands, it might be fighting itself in the LP side.

I found the drawings to be a bit lacking in some areas, so I have modelled it up in SolidWorks 3D CAD to make sure it will all go together for when I get round to building it.

I don't like the idea of four bearings inline on the crankshaft, I cannot say I have ever seen that in full size practice, so I may change that on my model if I can come up with something better.

Phil

Edited By Phil P on 28/08/2018 23:00:53

Edited By Phil P on 28/08/2018 23:03:09

[JasonBModerator @jasonb]

Very nicely done John, the light wood cleading goes well with the cream and makes a nice change to the usual darker combinations.

Phil, If the eccentrics were moved inboard of the bearings then you would have less overhang of the cranks so the outer bearings could be done away with. Maybe even swap the base castings over so the ctr lines of the two engines come closer together. Should be easy to play about with the parts in CAD.

[FMESParticipant @fmes]

Hi John, am I guessing you are running on air?

There is no expansion available using air so the pressure supplied to the second cylinder wil be lower, also you are relying on the exhaust from the first engine to be exactly in time with the inlet to the second.

If the two cylinders are the same bore then I would suggest that you run it as a 'twin' rather than a compound.

Regards

[duncan webster 1Participant @duncanwebster1]

Posted by FMES on 29/08/2018 07:51:55:

Hi John, am I guessing you are running on air?

There is no expansion available using air so the pressure supplied to the second cylinder wil be lower, also you are relying on the exhaust from the first engine to be exactly in time with the inlet to the second.

If the two cylinders are the same bore then I would suggest that you run it as a 'twin' rather than a compound.

Regards

There is some expansion running on air, but not as much as on steam. For instance if you start with 40psig steam (55psi absolute) and double it's volume you will have 27.5 psia, 12.5 psig. With air you get a pressure ratio of 2^1.4=2.64, so 21psia, 6 psig.

I agree if cylinders are same bore it will run a lot better as a twin, but I don't understand Jason's suggestio at all. If the OP has managed to get it running sweetly with 4 bearings in line, leave it alone

[JasonBModerator @jasonb]

Even with the compound cylinders of different diameters that the engine has it can be run as a double high and should allow you to get it back down to 3or4 psi running pressure and with booth cylinders working it should run at less than 100rpm. Just need to feed air into both valve chests from the compressor and let the exhausts blow to waste.

Duncan, that part of my reply was to Phil's post about a 4 bearing cranks, that is why I started the paragraph "Phil, ………."

[Mike BondarczukParticipant @mikebondarczuk27171]

Hi John,

What a lovely engine you have built and a great colour scheme to boot.

I managed to purchase a part complete kit for the same engine a few years ago and am trying to obtain the missing castings from Blackgates but having some trouble getting a reply to my enquiries.

The missing items are the steam chests, front and rear cylinder covers, trunk guides, crank discs, eccentrics and eccentric sheaves, steam chest covers and finally the drum for the wheels, and hopefully they are still available for purchase.

Did you encounter any specific casting issues or problems with the turning of any parts.

Mike

[Mick B1Participant @mickb1]

Very attractive piece of work. Phil's comment about 4 in-line crankshaft bearings prompted to me to search online, and indeed I didn't find any obvious examples at full size. Reducing it to 2 bearings doesn't look at all straightforward, not without risk in any case.

I built the little PM Research No. 7 twin some time back, and that has 4 in-line bearings – but I don't know what (if any) full-size engine it's based on.

[JasonBModerator @jasonb]

Posted by Mike Bondarczuk on 29/08/2018 10:33:32:

The missing items are the steam chests, front and rear cylinder covers, trunk guides, crank discs, eccentrics and eccentric sheaves, steam chest covers and finally the drum for the wheels, and hopefully they are still available for purchase.

I don't see anything on that list that would really need a casting, few bits of CI bar and a bit of thick wall steel tube would give you all that is needed and probably cost less.

[Phil PParticipant @philp]

I am fairly sure the only reason the compound has four main bearings is the fact that is was also available as a single cylinder model, and then mirrored to make it into a twin or compound. Hence the need for two bearings on one bedplate.

I bought my castings from Blackgates a couple of years ago and everything was available then, the only one they had to have a new pattern made for was the main flywheel drum, but they did that at the time I bought my set.

Phil

[JasonBModerator @jasonb]

Phil

This would be how I would arrange things if going for just two bearings, all it needs is the beds swapping over, lop off one bearing base and probably extend the pad that the cylinder mounts on. This is a quick and dirty tracing but you should be able to play about with your CAD parts easy enough to try it out.

[Neil WyattModerator @neilwyatt]

Worth the effort of persevering with the castings John. I too like the colour, it shows off the metalwork really nicely.

Compounding really needs the cylinders to be of different sizes, and you might also benefit from a receiver between the two cylinders. Plus air is not good for compounding.

Neil

[JasonBModerator @jasonb]

The Clarkson Cross Compound has HP at 3/4" bore and LP at 1" bore both 1.5" stroke. It's the air that is the issue not so much cylinder sizes. It uses two of the same " high" cylinder castings from the 1" single but one is bored smaller.

[Ian HewsonParticipant @ianhewson99641]

Tangye twin engine uses four bearing crankshaft.

[Tim StevensParticipant @timstevens64731]

Four bearings on a crankshaft is not unusual – just count the bearings in a modern multi-cylinder car or motorcycle engine. My 1974 Ducati had two cylinders and a built up crank – with four main bearings to help keep the two sides in line. The only difficulty for a home made model is to get the bearings properly in line. One way to do this is to line bore all four at the same settings – having remembered to number and dowel (etc) all the bits so it goes back together in line every time.

Cheers, Tim

[John PurdyParticipant @johnpurdy78347]

To everyone, thanks for your complimentary comments. As far as compounding goes, as I suspected, and as most of you have pointed out, since the exhaust air has very little work energy in it as compared to steam, the HP cyl. is basically just motoring over the LP cyl. against friction with little assistance from the HP exhaust. Hence the higher pressure required.

As far as problems with the castings, besides the bearing pedestals mentioned earlier, the valve chest castings had insufficient machining allowance on both height and thickness so when just cleaned up to a flat surface they were smaller than the drawing dimensions in both height and thickness. As well the cast inlet boss was oval to the point that the 1/4" tapped hole would have broken through on the sides so were milled off and replaced with turned bosses threaded into the valve chests. The trunk guides were decidedly oval so after some work with files to get then close to circularity, they required very careful set up for boring to ensure that the bore didn't run out the side of the casting. As it turned out when bored to the stated 5/8" dia. the wall thickness was only about .045". It would have been easier to change that 5/8 bore to 9/16 which would have given a thicker wall and made setup that much easier.

One mistake I made when machining the cyls. was to use my usual practice of using the as cast radius of the flanges as the datum to define the cyl centres (always worked well before). This resulted in that when the cyl. was  finished, and the cavity in the valve chest was centred on the steam ports, the top of the valve chest sat proud of the top of the flat as cast surface of the cyl where the exhaust boss is, by about 1/16" with the result that there was no room for the top row of valve chest studs in the cyl. After some head scratching (and a few choice words!) the solution was to lower the valve chest to the point were the studs could be fitted and to offset the valve on the spindle so that it was centred on the ports (as can be seen in the photos the valve chest is still slightly proud of the cyl).   In hind sight( wonderful thing that) I should have used that as cast flat surface were the exhaust boss is as the datum to define the cyl .centres and filed back the flanges to suit. There was plenty of meat on the bottom of the casting for the cyl. centres to be lowered. As it was I machined off about 1/8" to bring the cyl. vert. centre to the 1 1/16" height required.

As Jason says, this engine was basically two of Clarkson's horiz. singles with and extra mirror image bedplate and a large casting for the rope drum. The ends of the rope drum are just the horiz. singles flywheels reduced in dia. somewhat. As the cyl castings weren't cored it could be made as a two cyl. high pressure of 1" bore or a compound with 3/4 and 1" bores. I chose the latter.

John

Edited By John Purdy on 29/08/2018 19:52:20

Edited By John Purdy on 29/08/2018 19:58:32

Edited By John Purdy on 29/08/2018 20:01:33

[Author]

Posts