303. Engine Performance

I’ve been puzzling over the reasons for the quite remarkable increase in performance in the engine since the refurbishment.

Last year the best we ever got on the ground was 1050rpm. Normally you would expect an increase in the air as the load on the propeller is reduced, but that wasn’t the case; it was exactly the same however fast we were going.

This year the maximum speed on the ground is 1150rpm, increasing to 1200 in the air, which is clearly a big increase in power, and behaviour more like a normal aero engine.

In fact power is a function of the cube of the rpm, so an increase of 100rpm on the ground means we’re getting an astonishing 30% more power from the engine. It means that last year we were flying with the same power output that Grandad had from the 80hp Gnome, before he managed to switch to the le Rhone.

So what’s changed?

We’ve changed from pure castor oil to Castorene R40.

We’ve replaced all of the cylinder liners. A few had been scored by the flight with the oil turned off, but most were astonishingly clean, and obviously damage from the last flight is irrelevant to this discussion. They were all more or less on the outer limits of diameter, however. And three of them had been replaced at some point with ductile iron, instead of cast iron.

We’ve replaced the piston rings. The original ones (fitted by TVAL as part of their restoration) were leaking a little near the slit, and so one would suppose that we were losing a little compression, but in truth the compression always felt excellent whenever we had to depress the exhaust valves for priming. It was very hard to do on those cylinders that were on the compression stroke.

We’ve replaced the steel pistons with aluminium ones. I was concerned that because the thermal expansion of aluminium is double that of steel, we’d need increased clearances.

I knew that the cylinder bores were tapered inwards at the top – presumably so that it would end up as a parallel bore at operating temperature. Knowing the nominal bore and the taper, I could come up with a reasonable estimate of the design temperature of the cylinder head (and therefore, of the piston). I was amazed to find that it was only 80degC. On reflection, however, this would seem  to match our own experience; after a ground run, you can touch the heads for a short time without burning your skin.

When I did the math for the aluminium piston and compared its nominal diameter with the measured diameter of the old steel one, the additional clearance was about four times what was needed. Which was a relief, since as far as I know the only other airworthy le Rhone with aluminium pistons is in New Zealand.

But where is all this extra power coming from?

What about the new liners? Well, these are very accurately honed to the as-new diameter. They are also all of cast iron, which is picked for its ability to provide a ‘slippery’ surface because the intrinsic porosity retains oil. Could the ductile iron have been less good in this respect? When you think of the high friction forces in a rotary engine, with nine cylinders, each of 105mm diameter, 75mm long (that’s a contact area of nearly two and a half square feet), travelling at a maximum of around 20mph, any improvement in the coefficient of friction would have an important effect.

So what about the piston skirts? The original steel pistons were left with a slightly rough-turned finish, the idea being that the very slightly ribbed effect would help to retain oil. The aluminium ones have a more conventional finish, but with an increased clearance, even allowing for the extra expansion of the aluminium.

And while we’re thinking about friction losses, could the change in lubricant be having a beneficial effect as well?

One factor that seems to mitigate in favour of this argument is the fact that the engine is increasing in speed in the air; could it be that the friction losses on the cylinder walls were using up so much power that the poor old thing couldn’t speed up, even if the propeller loads reduced in the air?

I don’t know, and I’d love to investigate the information from the other airworthy le Rhones. Perhaps it’s just a spot of Shuttleworth magic.

But boy, are we chuffed with the result!

Advertisements

296. Little Things

While 1264 has been here at home, I’ve taken the opportunity to sort out a couple of things.

The socks (or are they gloves?) that protect the propeller in transit have done sterling service, but needed replacement. An offcut from the local carpet shop proved exactly the right size, and is now doing the job.

And I’m proud of myself for having come up with a way of filling the pulsometer which is supposed to tell us that the oil pump is working correctly.

Word from all the experts was that it has proved impossible, but I’ve come up with a way that takes only a couple of minutes, and doesn’t spill a drop of oil anywhere…

The pulsometer is the glass dome on the right of the picture. It’s connected to with a small bore oil line to a tee in the output from the oil pump, and the level should pulse in time with the strokes of the pump.

The difficulty is that the line is closed – it comes to an end in the glass dome, so it’s difficult to see to persuade the oil into it.

The solution? Undo the two bolts securing it to the panel and let it hang upside down, with the isolating cock open and the knurled nut holding the glass in place loosened.

Then disconnect the oil line into the engine and connect it using a piece of rubber hose to the primer pump (an oil can modified to allow us to prime the cylinders with petrol).

Fill the oil can with castor oil and pump until the pulsometer bulb is more than half full, then shut the cock, tighten the knurled nut, and re-attach the pulsometer to the panel. Simples!

It will be interesting to see how it works in practice. The manual says it’s possible to use it to check the speed of the engine by counting the number of pulses against a stopwatch, though I doubt anyone ever did this in flight, particularly as the pump is turning at 1.8 times crankshaft speed, so you need to be pretty good at mental arithmetic as well!

291. Progress. Slow, but progress

Yesterday I spent all day making new guides for the pushrods.

You can see what they do here, complete with a graphic video.

The ones I made originally were intended for retrofitting to an already assembled engine, and weren’t ideal, so I’ve made some new ones which should be more secure and not need to be checked before each flight. The trouble is that while each one is fairly simple to make, nine of them take a long time! These need to be threaded over the pushrods before they are fitted, so it seemed a good opportunity to get the job done.

Then today Theo and I and Chill met up at Shuttleworth to get as much work done on the airframe as possible while we await the completion of the engine.

The engine itself has made a little progress since last week – our nice new pistons have been honed to fit the gudgeon pins.

These are aluminium pistons, copiers of those made by Thulin in Sweden when they manufactured them under licence, as opposed to the originals which are steel, as made for the American-built examples.

The valve gear has been fitted to each cylinder, so they just need screwing back into the crankcase. (Theoretically you can up the compression ratio by screwing them further in, but I doubt it would be a good idea. You’d need to take corresponding bits off the pushrods as well…)

I thought you might be interested in this picture of the crankshaft, taken from the front end. You can just see daylight coming through the hollow crankshaft. The carburettor goes on the far end where the daylight is.

And you can see the brass keeper for the main ballrace on which the crankcase revolves. I think it’s astonishing to think they could make such big ball bearings in 1908, and even more astonishing to think that it has survived everything we’ve thrown at it!

 

 

Once the pistons and rings are fitted, the cylinders have to be screwed in just the right amount and locked in place. then the complex cam rings and followers have to be fitted to the front of the crankshaft and the pushrods inserted. Finally the front cover plate has to go on. I suspect I’ve missed out some important stages from this list, but I’ve never reassembled a le Rhone myself.

We found that the propeller hub had picked up a little surface rust, so Shuttleworth kindly let me use their wirebrush and polishing mop to clean it up before refitting it to the propeller.

Also on the bench in Dave’s workshop is his current masterpiece – a new end for the feeder mechanism for the Spitfire cannon which will go inside the wing. You can see the original in the middle, and the new end on the right – a mirror image of the original. Dave’s machined it from solid. Amazing!

Meanwhile, Theo and Chill were retensioning the cabane rigging and refitting the refurbished tachometer and tachometer cable, complete with its new gearbox, which should make the tachometer read correctly for the first time.

 

The little gearbox should double the speed, with the result that the tacho should no read correctly. We hope!

Progress was slower than it otherwise might have been, owing to the glorious weather outside, and the sounds of all the Shuttleworth machines being run up and flown as part of practice week. Noisiest by far was Matt Pettit who was trying to sort out the problems with a Harvard, which involved running it at full power in very fine pitch about half a dozen times. As you may know, the Harvard is famous for the fact that under these circumstances the propeller tips go supersonic, and the resulting howl penetrated to the furthest corner of the workshop. making conversation impossible.

I also took the time to polish the copper inlet ducts, which will make the engine look as it it’s been refurbished once it’s reassembled.

Before (top) and after (bottom). Please tell me you can spot the difference!

Thankfully, as we had a final cup of tea before departure, Matt came in with his customary cherubic grin – Matt is just like a 6ft 3in schoolboy let loose in an airfield full of his favourite toys – to announce that he’d finally got to the bottom of the problem. The Harvard – a two-seat trainer – was never designed for fuel economy, and the ground running, most of which was at full power – had consumed 60 gallons of fuel!

We had a great day, but for us the tension is starting to rise.

The engineers were tinkering with the engine of the Avro 504K all day. Indeed the main reason that our engine isn’t assembled is that they’d been working on it all last week.

In a week’s time, we need to be packing the Scout up ready for her first static display of the season in Ludlow, and the engine must be installed and running.

The Shuttleworth knows this full well – I pointed it out to Chief Engineer Jean-Michel Munn today – but they only have limited resources and it’s going to be touch and go.

But we know that they will do everything they can to be ready, and we can only keep our fingers crossed…

The Shuttleworth Collection is just round the corner from the historic Cardington sheds, which housed the R100 and R101 airships in the 1930s, and today houses the amazing Airlander airship. And on my way home I spotted it outside, so stopped to take a couple of pictures.

And what does this remind you of?

272. Bits and Pieces

This is what happens if you leave your aircraft at the Shuttleworth Collection. The engine has now been reduced to every one of its component parts, and is spread all over Phil’s workshop.

AA tableful of engine bits. Connecting rods on the left (note the unusual big ends), the bloc-tube carburettor in the middle, and the commutator ring on the right.
A tableful of engine bits. Connecting rods on the left (note the unusual big ends), the bloc-tube carburettor in the middle, and the commutator ring on the right.

We had a quick peek today while we were collecting the airframe for its static display at the Bovington tank museum next weekend, and to have got so far in just a week is simply astonishing.

Everything is gleaming and clean, labelled and examined. Until we’ve heard Phil’s final report we aren’t sure what, if anything, needs to be replaced, but it’s clear that the vast bulk of it is absolutely fine, though we believe it will need new piston rings, the others having not bedded in very successfully.

Here you can see the core of the engine - the crankshaft, and the crankcase. On the top right is one of the cam rings for actuating the valves
Here you can see the core of the engine – the crankshaft, and the crankcase. On the top right is one of the cam rings for actuating the valves
Three cylinders, all cleaned up, with the valves lapped to ensure we get perfect compression.
Three cylinders, all cleaned up, with the valves lapped to ensure we get perfect compression.

And although we had believed that the engine has behaved impeccably throughout the year, and it will be an expensive exercise to get it airworthy again, it’s a real pleasure to see inside, and to understand more fully how it all works. We both hope to be able to get up to the Shuttleworth when Phil’s working on it to be able to actually see all the inner workings so that we can get an instinctive feeling for the best way to treat it.

268. The Nitty-Gritty

The Shuttleworth Collection have peered inside the engine and made some preliminary findings which are a bit salutary.

First, there appears to have been some overheating of the cylinder heads. We’d kept a close eye on things during the test phase last summer, but there was no indication that anything was getting too warm; no discolouration at least. However, the exhaust valve springs are significantly weakened, and that’s a sure sign of things getting too warm.

It’s also possible there may be a problem with one or two of the valve stems, but we won’t know for sure until it’s had a full strip down.

Rotary Engine internals

Next, the whole of the inside is covered with black goo. This, they are reasonably confident, is as a result of using castor oil instead of Castrol R40. Apparently their own engine stays spotlessly clean. Also, there appears some evidence of grit in the goo, which certainly won’t help the state of the internals. The engine has no air filter, so it’s very prone to dirt ingestion.

It also seems likely that the piston rings haven’t bedded in very well. This is likely to be the primary cause of the loss of compression, and it will take more investigation to see why this has happened. It might be the gumming up caused by the castor oil. Or it could be distortion of the cylinders caused by uneven heating. Or they might not have been too well manufactured in the first place. American-made rotaries weren’t famous for the quality of their build.

So it needs a complete strip down and rebuild, and we’ll need to see what, if anything, needs replacing.

That’s going to take a month or six weeks, and it happens that Phil has a quiet spell before the winter rush, so Jean Munn was keen to get the engine out of the airframe so that the work could start immediately.

This left us with no choice but to abandon the Shuttleworth and Duxford air shows. It’s a bitter pill to swallow, but probably rather easier than the bill we can expect from Shuttleworth shortly!

That’s the bad news. Is there some good?

Well, yes.

First, of course, 1264 has kept going throughout an astonishing summer. The engine has 11 hours on it – more than most other rotaries do in a generation these days. We made it to Greece and to France, and we’ve completed the filming schedule. That’s certainly something to be very grateful for.

Second, thanks to one of Theo’s contacts, it seems we might be able to fit another (non-running) le Rhone engine for the remaining static displays until ours is sorted.

And Third? Well, thirdly, we’ve learned a huge amount about operating a rotary engine. Some of the issues – such as the overheating – we still have to find a solution for; it’s tempting to make changes to the cowling to improve throughflow, but we must never forget that first and foremost this is an exact reproduction, so whatever we do must be, as far as possible, historically accurate. But oil and air filtration are things we can probably fix. And it’s possible that once rebuilt, we’ll be able to get that extra 100rpm which will make all the difference to the performance.

In the meantime, we need to start some serious saving; 160 hours’ skilled labour doesn’t come cheap…

233. Motorcycle links

It’s well documented that the rotary engine concept was developed from an original idea from the USA for a front-engined motorcycle. I’ve just come across this video which demonstrates – more or less – how the concept works, in a 1923 version.

There is no clutch, so presumably you have to warm the thing up on the stand and jump start it when you want to ride on it!

217. End of a Chapter

It’s only eight months ago I bought a Land Rover Discovery to tow the Scout trailer, after my lovely Octavia estate proved to be too light.

The Disco has a reputation as the best tow vehicle in the world.

Since then, it’s had to have replaced:

  • transfer box,
  • prop shaft,
  • two tyres,
  • rear brake shoes & discs
  • front brake shoes, discs & calipers,
  • one front wheel bearing,
  • suspension bushes,
  • suspension compressor,
  • two front suspension units complete,
  • drive shaft boot,
  • LP fuel pump,
  • towhitch and crossmember,
  • and the sunroof drip tray drains have had to be cleared.

But above all, I never felt happy towing the trailer at speeds above 45mph.

Its replacement, a Hilux which has a reputation for being a bit light at the back end, tows it on rails. I felt utterly confident towing at 52mph for a couple of hundred miles in all conditions.

And it’s all been unplanned expenditure. essentially it amounts to almost half of what I’ve put into the Scout, and – unlike the investment in the Scout – I’ll see very little of it back.

But today a nice man came and took the Disco away and I’ll never have to look at it again, and while the Hilux may be an expensive luxury, at least I have complete confidence in it.

Hooray!