But despite all the soul-searching over the car, we’ve managed to make significant progress on the matter of the Scout’s apparent diarrhoea.
On completion of the last flying day, we checked the oil consumption and confirmed that it was correct; 7 litres in an hour and a half matches the 5lt/hr it’s supposed to be.
We’d also discounted the theory that it was leaking out of the back of the engine, since Gene DeMarco very correctly pointed out that there is a partial vacuum in the engine which sucks the mixture into the crankcase!
So the engine was performing exactly as it should; it’s simply a question of where the oil goes after it’s come out of the exhaust.
And during our day at Shuttleworth, we uncovered what may be the deciding factor. When you remove the cowling and slowly turn the engine to see where the exhaust valve opens, it becomes apparent that it does so before it gets to the slot in the bottom of the cowling. This means that the oil, instead of being ejected from the aircraft by that initial blast of gasses when the valve opens, are being largely trapped inside the cowling.
And because there are wire edges round the rear of the cowling and the slot, the oil can then form a puddle in the bottom of the cowling, which gets picked up by the whirling engine and distributed clockwise (from the pilot’s viewpoint) – mainly to the port side of the aircraft.
The photograph below may show faint witness marks where the exhaust valve has been opening adjacent to the slot in the cowling, and you can see the wire reinforcement around the slot.
Of course our fancy lip seal should then stop it working its way through to the fuselage side, and while we can’t say definitively why this should not be so, it’s possible to speculate that the rotation of air inside the cowling during flight is sufficient to lift the corners of the seal and get them flapping, thereby letting the oil through. If anybody would like to volunteer to sit on the undercarriage axle in flight with a mirror and a torch to verify this guess, please apply for the job immediately!
The obvious solution, of course, is to open up the bottom of the cowling as much as possible to let the oil blast straight out from the exhaust valve, but we need to try and validate this against the historical records. We checked the cowling against the drawing – I wondered if perhaps the opening should have been slightly off-centre – but there’s no indication that’s the case.
We don’t have any pictures of 1264 which clearly show the lower half of the cowling, so we started to look at ones produced at a similar time.
The first picture we noted was this one, taken at the factory, of 1607, which was factory serial number 456 – about 33 before 1264 – and it’s clear we aren’t the first to suffer the problem!
But then we noticed this one, taken at Imbros, – as far as I can tell either 1259 or 1263 – and as you can see, they’ve cut the lower cowling right away.
This one – also taken at Imbros – is an unidentified one, but appears to have a cutaway lower cowl. It’s possible it’s even 1264, since the gun is on the starboard side and there’s no cutout in the centre section trailing edge.
And here’s another Imbros machine – 1262 – which also appears to have a cutout.
But this one – either 1263 or possibly 1264 – has no cutaway, and there are plenty of later machines which don’t have the cutaway before they modified the whole cowling, including the mounting method, much later on for the Type D.
So – yer pays yer money and yer takes yer choice.
So the current plan is this. We don’t want to disturb the existing slot, since the wire reinforcement adds significantly to the stiffness of the whole thing. We don’t want to add openings either side of the existing slot – partly for the same reason, and partly because they would prejudice the possible later modification of the existing cowling to a cutaway shape.
But we think it should be possible to add a second, unreinforced slot behind the first, and probably a bit wider, which would help to drain the oil away as soon as possible and stop a puddle forming, and – having discussed it with Steve Moon who made it, this is what we’ll do before the next set of test flights. There is ample photographic evidence of field modifications being made to Scout cowlings, so we feel justified in making these changes from a historical point of view, and then we can evaluate the results before deciding whether to give the cowling back to Steve for modify to a cutaway design – which is also pretty well justified on historic grounds.
We still haven’t come to a definite conclusion about the misfiring that occurred during Dodge’s test flights, but I’m beginning to think that if – as seems likely – the air inside the cowling is a lot oilier than normal, it’s possible we’re getting arcing on the commutator ring which is mounted to the rear of the crankcase and us fully exposed to the air there. So we’re hoping the one modification will fix both problems, and in any case we’ll be trying the new plugs as recommended by Shuttleworth with which they seem to have had good results.
A couple of other changes we managed to address were to weigh up a light spring trim to the bloc-tube (throttle) control. For some weird reason it wants to close itself in its mid-range positions. There are no springs in the system, so we tried increasing the friction on the lever until you could barely move it, but it made no difference. There must be some sort of ratchet action going on with the vibration, and so we’re going to try fitting a very light spring tending to open the throttle to see if that will make it stay where it’s put.
And finally we drained out all the Castrol R ready to use pure castor oil instead, on the recommendation of TVAl.
On Sunday 20 September it will be on display at the Bicester Sunday Brunch Scramble, and we have set aside the following week for the remainder of the flight tests. Fingers crossed!