This video demonstrates the assembly and maintenance of Rolls-Royce Merlin engine ancillaries, highlighting the mechanical face seal technology used in coolant pumps (with carbon seals on bronze faces and spring-loaded capsules), the dual fuel pump redundancy system for preventing engine flooding during negative G operations, and the Constant Speed Unit (CSU) that uses centrifugal weights to automatically adjust propeller pitch for optimal RPM across varying altitudes. The video also covers the repair of cast aluminum cylinder covers using tin bashing techniques, which exploits the magnesium-containing material's ductility to restore dented components without cracking.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
Rolls-Royce Merlin MkXX rebuild, Episode 17: Ancillaries, plus tin bashing cast alloy!
Added:Hello and welcome back to this latest episode of building a Rolls-Royce Merlin. This time I'm going to assemble some of the ancillaries and put them back onto the engine. Starting with this coolant pump. So, starting with the main body here. If a coolant pump is fitted to an engine and not just driven by a belt, there are two seals on it. There's one here which prevents the coolant from inside escaping. And then you also have to have an oil seal on the engine drive which is down inside here to prevent oil escaping from the engine. Then you generally have an air space in between the two.
So this pump runs on plain bearings and you have a gland packing inside here with this nut which is left hand threaded due to the rotation. I'm just going to put this on a small amount and uh it's got that ratchet. Lovely sort of design.
Sounds great, doesn't it? So, I've lightly greased this shaft, which is stainless steel.
Feed that in through there. Now, that material there, they call it morganite. It's basically a carbon material, and it runs against this face on the back of the impeller.
And that's the coolant seal. And then the inlet casing also has a carbon bush inside it which this thing will rest on giving you the correct clearance between the blades and the housing.
And this stuff works really well in coolant or water.
So we're going to insert the shaft into there.
There we go. That's fully back.
You can see the splined shaft sticking out of there. So then we would tighten this up to a point where the gland packing is just starting to nip the shaft a small amount and that seals it.
It also has a grease cup which you can gradually screw in. Goes into the side of there.
And these are obviously maintenance issues which we'll come back to later.
So then once the inlet casing goes on, carbon bush is pushing against here and there's obviously a very small amount of inflat on it.
So the drive housing goes on like that and it also forms the bearing housing for the lower vertical shaft in the engine wheel case. And the pump is driven by a very long quill shaft starting up at the top of the lower vertical drive shaft.
And this device here is to prevent too much oil coming down to the bottom. So this inlet casing also incorporates the drain valve for draining all the coolant out the engine and also a secondary part of the cooling system which is there to heat the supercharger intake elbow to prevent the buildup of ice.
Don't want that going into the supercharger blades do we? And then depending on the installation there are various different intake connections onto the pump which are unique to various different aircraft of course. So the pump outlets feed up into the engine cylinder banks.
You have a rubber seal in there which will be replacing.
Then compression ring there. Then compression nut that screws onto there like that. On this particular engine, you have the same fitting on the way into the engine as well. It can be very difficult to fit these pipes. And another thing is they have to be absolutely correct in terms of their positioning. Um, by that I mean the bends in the pipes because if they're out at all or if they've been knocked or whatever, then they just don't line up and um they are very very difficult to fit. So later on and here is one of the outlet pipes from a later engine. There's the gland nut there on the other end over to this which is called an AIMO coupling and just made it a bit easier to fit. So this was introduced on the two-piece cylinder blocks. You can kind of see they call these ram horn pipes as well. So I said that these outlets feed up into the coolant inlet on the cylinder banks.
That is the case on every installation except the mosquito. So on the nozzie they have a short coupling like this.
Now Rolls-Royce called this a reverse flow cooling system which does cause a bit of confusion because it's not strictly reverse flow. The coolant flows in the same direction through the engine. The difference is you come out of the coolant pump into the radiator, back from the radiator and up into the engine. And there'll be a radiator on each wing of the aircraft, I think. Pretty sure about that. Uh all the other installations they go from the pump up into the cylinder banks out of the top of the engine through the header tank which is the same on the Mosquito back down through the radiator and then from the radiator back into the intake housing here. So on the Mosquito you just go straight from the engine header tank back into the inlet. No radiator obviously there. So that's what they call a reverse flow cooling system.
So, I mentioned earlier on about this gland packing being a maintenance issue because every so often you have to tighten it. Now, there were plenty of automotive engines around at the time that still used this system, uh, including some with a grease setup on them as well, but they went over to this system, which Rolls-Royce called a packless gland, but actually nowadays we call this a mechanical face seal. So on the coolant side, this this is mounted in the middle of the pump here. On the coolant side, you have your carbon seal there running on a bronze face like that. Now, this is a sealed capsule, which also acts as a spring. So, it puts pressure onto there.
And then on the other side you have a ground steel face running onto another bronze face there which will be lubricated with oil.
So this is stationary this housing and the whole assembly rotates and it is keyed together in the middle by these splines. And this is the principle used by pretty much every modern coolant pump. This is a similar component that was devised by Packard for their Merlin which is rubber all the way around but also spring loaded. So these type of mechanical face seals meant that instead of having the shaft incorporated with the impeller it had to be separate.
So it was then splined into there like that and they also went over to a ball race on the back of it here and the spline coupling going up into the engine there. So this bronze impeller is reasonably advanced, not just for the time, but even by modern standards. It's far from primitive. You can see there's some intermediate veins here which are smaller than the main ones. So the purpose of that is there are trade-off between increasing the flow of the pump and choking the inlet or coming in here by putting more large veins in there. So it increases the flow through that part of it.
Too large a gap between the main blades by missing these out can cause flow separation and vortices at the blade tips making the pump less efficient. And then all these holes that have been drilled in it combined with these slots in the back here are for pressure equalization to prevent excessive thrust load on the bearings. Particularly because this is working against a carbon thrust bearing. And this pump goes at 1.5 times crankshaft speed. So when the engine's cruising, it's doing 3 and a half to 4,000 RPM. So now we're going from the relative simplicity of the coolant pump back to the normal Rolls-Royce complexity of things. So this is the fuel pump. It's a double pump. And this is the drive housing which goes straight onto the wheel case.
Again, this has a gland seal on either side in the same way as the coolant pump. So you have to stop engine oil going down or through into the pump from this end and you have to stop fuel from going back into the engine and in the void in between the two you have a drain which goes to atmosphere just in case there is any leakage. So the shaft just is a plain bush just pushes through there with a thrust washer on the end of it like that.
And then this helical gear is fitted on with a woodruff key onto there to tap it end of the shaft. So we then have another gear. Now this has suffered from water ingress. It's one of the few parts on the engine that actually did. So there's corrosion on this gear, but it'll run fine. But there's an awful lot of stuff in this fuel pump that would, for example, never go back in the air. And this is a w this is a bronze component which is white metal on the outside that pushes onto there and they turn like that and there's a bit of resistance in the gland seal inside there like there should be of course.
So the purpose of having a double fuel pump is that these gear pumps would be susceptible to any large contamination going through there.
Now it is filtered but what they're thinking was as with a lot of things on aircraft that you build redundancy into it so you've got two of everything if you possibly can.
So if something goes in between these teeth or something happens to this half of the pump, these phasing gears here have these quill shafts splined into the ends of them like that.
So one of them drives the front pump, one of them drives the back pump.
meaning that if the uh if one of the fuel pumps does jam up then the other one will keep running. So this was this contributed to the problems they had with negative G operation of the engine because when the floats in the carburetor float chamber are forced fully open, you have two fuel pumps pushing fuel in at a rate of maximum power plus 20% times two, which is a lot of fuel. So once the engine floods, of course, it's very difficult to get it started again.
Again, you can see some of the corrosion pitting inside here caused by the watering grass. The housing contains these bronze bushes again which are white metal and the whole assembly including these gears here run in petrol. There's no oil inside here at all. Now this part of the pump here we have the relief valve. So that basically opens at 4 PSI and all it does is it just couples the inlet and outlet together when the relief valve opens. So at all times, no matter how much power the engine's producing, particularly because of these two fuel pumps, the fuel is always circulating around like this. And then it's just bringing in as much as it needs to from the fuel tank.
It's actually marked on there from tank to carb and then back into the engine.
Then on the outlet connection, you have the main fuel line going to the carburetor there. And you have these two takeoff points. One of which is used for the oil dilution. So in very cold weather, what they used to do before they shut the engine down, you had this connection with a solenoid valve going back into the oil tank. And you would actually dilute the oil with a certain amount of petrol. And it had to be the correct amount so that when you started the engine up the next morning, the oil was just that bit thinner because monograde oils of sort of aeros shell 100 grade which is an SAP50 are very very thick in cold weather if it's like minus 20° or something like that. They don't circulate very well and you can get very high oil pressures as well. But if you dilute the oil with too much fuel then it won't boil off when you start the engine. So it damages the quality of the oil after that.
And here is a fuel pressure switch which went to the pilot. So this relief valve is measuring the pressure difference across the pump by pressure acting on that valve and the holes around here behind it. So later versions also went over to what they call a balanced relief valve which is a diaphragm operated puppet valve. And instead of sensing the pressure across a pump, it looks at the inlet pressure going into the air intake of the engine. So it's actually measuring the atmospheric pressure.
And this was also sometimes used in conjunction with this so-called AML valve because it was made by the AML carburetor company even though they didn't actually make a carburetor for the Merlin. And they call this a pressure reducing valve. It's controlled by a much larger diaphragm. And again, this goes back to the air intake via this unit here.
The other type of fuel pump that was used on Merlin with SU carburetors was this Pesco F8. It's based on an American design but manufactured in the UK. It's a sliding vein pump. And then underneath that is this relief valve which is adjustable on this um pump by that square hole in the middle there. And it also had a balance to atmosphere connection on it. One thing I didn't mention on the gear pump, which is the same on this one, is that it's actually fed by main engine oil pressure rather than lower pressure. And they took a tapping from the oil pump because it's quite close to this. And that just lubricates the bearings in this drive housing here. This only has a single pumping element in it, unlike the Rolls-Royce dual gear pump. And both these pumps run at 0.6 times crankshaft speed, so about half speed. This one suffered from the elements a little bit.
It came out of one of the stilages containing hundreds of these things, hundreds of these gear pumps and all manner of other things at Park House Aviation back in the good old days. So these two units go on the dual drive or V drive on the front of the crank case.
This is a vacuum pump. So this runs the gyroscopic instruments and it is just a very simple sliding vein pump like that.
and it's lubricated by engine oil which is then separated on the exhaust line further down. And on the Merlin, the lubricating oil and return go through two of these small holes in the bottom. But the pump's quite universal. It's designed to have lubricant fed in these ports on here as well.
And it's an American designed Pesco pump, this one, but manufactured in the UK. So this is the rotal propeller constant speed unit and as I mentioned in the last video I ran this on my own ground running engine for quite a few years with the propeller off this Halifax as well and it works perfectly. Took it apart and it was absolutely clean inside.
Cleaned everything up anyway and it works fine. So there are two ports here that control the propeller. It's a double acting unit. Some of these are single acting.
So, it puts oil pressure out to one or other of these two ports. And then, because it's a multi-engineed aircraft, you've also got this input, which is a quick feathering pump input, where you have an electric pump feeding engine oil straight into there. And the pilot can either feather or unfather it then using this control. But what this control normally does is it sets an RPM. I don't know if this has got marks on it. No, some of them do and some don't. So basically it's a governor and it was designed by Woodward Corporation in America. Nearly all CSUs look like this.
Again, this was built in this country, but the basic design is used under license. So there's a cup in here with centrifugal weights inside it. There's a spindle through the middle of the rack and that operates a spool valve through the center. Look at all the oil coming out of this thing. It's great, isn't it?
And in addition, in the bottom of this thing, there's a an oil pressure pump as well. So, if I turn the input shaft, you can see there are two gears inside here, and the other one's rotating there. This is because the propeller requires more pressure than the normal engine oil pressure. So, it boosts up to about 600 PSI. And the electric feathering pump goes even higher than that. It's quite often well over 1,000 PSI on these things. So, one way of illustrating what this unit does is that the pilot will set a given RPM that they want to climb at. So, you're climbing at 2650 RPM. As the air gets thinner, the propeller wants to go faster. And this governor puts more pitch onto the blades to bring the RPM back down. That's just an example of how they work. I need a whole quite lengthy video to go through everything on it, but that that's the principle of the governor. low the pilot doesn't have to keep making corrections to the propeller pitch. And the reason for having a variable pitch propeller in the first place is to match the density of the air as you climb. And in particular as aircraft got higher and higher and particularly with photo reconnaissance.
That's a very wide range. So you can consider it as being a little bit like a a variable speed transmission really. So this is the 90° hydraulic pump drive that goes under the lower crank case.
The input shaft from the engine has a long quill shaft running into it. Goes into there. There's a roller bearing in the bottom of there. These see tiny little rollers on it and they actually go into that race there. So that goes down into the top.
Then the output shaft has its own bearing and housing which then goes down inside there.
And then into this you have this coupling shaft which engages down there and onto the hydraulic pump itself. And there are various different types of pump manufactured for various different aircraft and some of them are interchangeable as well.
Interesting design feature inside here.
I've never overhauled one of these things, but if you look, if you can see, there are four back inside there, there are four bolts.
There are two at the front there and two right much nearer the back with tab washers on them. So, he must have had to have had fairly small fingers to be able to put that bearing housing in there.
You can see the top of the uh four bolts there in there. So this is a a Lockheed 7cylinder radial hydraulic pump and in the UK there was Lockheed and then automotive products and they made car braking systems but they are in fact all named after Loheed brothers who developed a hydraulic braking system for cars and also went on to manufacture aircraft as well. is all the same name, but this was a British company making this pump and it was in use for quite a long time. It came in early in the war and it was certainly used on the Hawkilly 125 aircraft much much later.
So presumably quite a successful pump.
This one's in really good nick actually, I have to say. And not very heavy really for a hydraulic pump either. This is a universal flange and a universal shaft with the six splines. And a lot of equipment that's fitted to these aircraft uses that standard. So it makes them very interchangeable. So theoretically you could put all sorts of different bits of equipment onto here.
But this was usually reserved for undercarriage pumps or hydraulic pumps anyway. But for example, the air compressor that I showed in the last one of the last videos has the same diameter flange on it as that. I don't know if that's a British or American standard.
but I'm not so sure. Okay, so I'm cleaning up the cylinder covers or cam covers, rocker covers, whatever you want to call them. And you'll be able to see on this one, it's not very obvious looking at the light, but it's got some damage to it cuz the engine landed upside down. So, there's this dent across there. But if I put this rule on there, you'll see it's quite a large dent. And then along that side there, there's an area which is pushed in again by quite a large amount.
And then there's a really big indentation across the top of it. Now again, doesn't really look much, but it is.
Um, and also it's got a quite um serious crease in it just across here.
There you can see it better from there.
Well, if you look at the coloration on here, these dark patches, these covers are actually made of a different material and a different heat treatment to the rest of the engine.
They've got some magnesium in them and they're softer as well. So, sometimes, and it doesn't always work completely, you can actually panel beat these things back out. I know any other piece of cast aluminium on these engines. You can't if it's been knocked out of shape if it hasn't cracked at that point. If you start knocking it back again, it will it'll just break off. So, it's quite fortunate with these covers. However, I'm suspecting that when I straighten out particularly this bad one across here, I think that's probably going to crack and it's going to need welding.
But anyway, we'll give it a go.
So, I can see the dent much more clearly from the inside now. I've cleaned the oil off this bit. Let's give it a color.
So, there we go. It's not perfect, but if you can see on there, but it's pretty much perfect. Very, very close. Right now, we've got this one on the side and here. So, initially I thought this was a crack in the cover, but it's not actually. It's a casting floor, believe it or not. So, the dent in the side, it's hard to see on this image, but it's around there, and it's quite substantial. It's much bigger than the other ones.
Yeah, pretty happy with that. And now the big crease. And you can actually see now I've cleaned the inside. There's a crack at that end. It's a much bigger crack there. And that's where the really bad crease was. the line you can see across it here. It's just uh one of the um casting molds.
So, this is a much bigger ask. It's obviously going to need welding anyway, but I'm going to start knocking it down in the middle and just see what happens.
If there's some damage left on this thing, that's great because it actually shows part of the history of the engine and what happened to it.
So, that's opened up these cracks on the outside, which I can fill.
I'm going to knock that one out a bit from the inside, but other than that, I'm happy with that. And so, there you go. Tin bashing cast aluminium. So, that biggest piece of damage that I repaired on this valve cover ran across there.
And you can see if you look where I've welded it, there's some very tiny blow holes in it there caused by oil which soaks into the material.
Another thing which might happen with this because these are quite thin and they obviously expand quite a lot as the engine heats up. I've got a similar repair on my ground runner to this as well where that was damaged and it cracks occasionally and has to be rewelded. Never told me reweld. You can end up with a cleaner weld because you're welding in between the new material that you've put in. I mean I'm inclining now that I've got it to this stage because it was quite badly damaged to leave it because it's part of the history of the engine. Similarly on this one, you can see a couple of marks. One there, one there. Right, so this is where we're up to now. Really starting to look like a complete engine actually.
All these components have just been fitted. We've got the hydraulic pump there with a 90° gearbox underneath the sump.
Lower crankase as they call it. There's the coolant pump. So that coolant pump outlet, the pipe that I was talking about before runs up around there into the inlet on the cylinder bank there.
It's quite tight around here. It's quite interesting actually because it goes around in between there and that lateral bolt and even that can cause problems with clearance.
So I've got those to do. We don't fit these pipes until the engine's actually been fitted because they run down underneath here. And when the engine's being lowered down onto the engine bearer or just even onto this trolley here, the pipes can get damaged basically by the engine mounting. So I always fit them afterwards anyway.
Starter motors back on. That's the hand turning gear, which I don't know if I mentioned that before. It's got a ratio of 15 to1. It's really for turning the engine, but in some instances I do talk about starting the engine from there.
So, here's the vacuum pump on the front on the V drive.
And then the CSU over there, which is all plumbed in up into the two feeds on top of the engine here and here, which go down through into the propeller shaft through rotating seals. And then the oil feed tubes tubes will come out through there into the prop hub.
And on this side, got the fuel pump fitted. Obviously, there's the other coolant pump outlet there.
This pipe here is the high pressure oil feed to that coolant pump drive that I was talking about in the last episode.
and working away towards the back of the engine.
Obviously got the supercharger inlet there, which is where the carburetor is going to go. And that is really the only major piece that's missing from the engine at the moment. So, I hope you enjoyed this video. And if you did, please tune in next time where I'm going to be taking the carburetor to pieces, cleaning it up, testing it, putting it back onto the engine, and I think it's pretty much finished at that point. It's going to go onto the trailer that I run the ground running engine on and be tested ultimately. That's quite a few months down the line really, but I think probably four to six weeks time should have the engine totally completed and we can get ready to talking about running it. Yay.
Related Videos

Audi RS5 4.2 Tuning JDEngineering
JDEngineering
1K views•2013-11-14

DALI + KNX: 500 Lights Offline! BCU Code Lock & Short Address Fix!
EngineerIsmailTech
560 views•2026-04-13

Explaining Quality Control of Concrete
maherbader
4K views•2019-05-25

World Mining Production Peaks - Lead Antimony Arsenic Titanium & more
LucarioandDialga
1K views•2019-04-19

Tech Titans: LFP vs Sodium-Ion Battery | Which is the most effective energy storage solution?
Enfsolar
522 views•2025-11-06

Doing the Math: Analysis of Forces in a Truss Bridge
TeachEngineering
1K views•2025-06-06

Inside Midnight Fighter Jet Refuelling Secrets of Stealth Missions | WION Podcast
WION
3K views•2025-09-20

Flash Point, Fire Point & Auto Ignition Temperature
HSELessons
38K views•2019-08-28
Trending

MIC DROP: Smithsonian Director Called Out For Woke Propaganda
TheAmalaEkpunobi
37K views•2026-07-23

2.4 BILLION Records Got Leaked...
DeepHumor
15K views•2026-07-22

Americans Confused in Australia for 17 Minutes Straight
IWrocker
17K views•2026-07-23

The Cult of the Ultra-Wealthy with Ed Elson | Better Offline
BetterOfflinePod
16K views•2026-07-23