Bristol Mercury revisited

[deltatango]

Hello Mike,
Only just read your last, I'm high (air) mileage but not too much wear as yet  .

I've been to the Melbourne GP once and decided then that, whilst the noise is wonderful, you get to see much more of the race on TV. It's also a lot cheaper - OK I did grow up in Yorkshire...
David

[Vixen]

With all the new imposed self isolation and distancing, I expect we will all be using the forum more and more.

I will start off with an update of progress on the second Bristol Mercury. I have reached the point of assembling the rocker gear on top of the cylinders





The front pushrod operates on the two front rocker arms, which in turn open the two inlet valves at the rear of the cylinder head. Similarly the rear pushrod operated the rear pair of rocker arms which open the two exhaust valves at the front of the cylinder head. The rocker gear control rod is between the two push rods.

As you can see, the rocker gear, which operates the four valves, is perched above the cylinder heads. The rear of the rocker gear is supported and free to rotate on two pillars located between the valve springs. At the front the rocker gear is attached to a control rod, which is tied to the crankcase, You can see the control rod between the two push rods. All the connections to the rocker gear are free to move. This was all part of Roy Feddon's (the designer) excessive attention to detail. Feddon calculated that the cylinders would expand in length when the engine was at working temperature, while the push rods would still be at outside air temperature and would not expand. The result would be a difference in tappet clearance between hot and cold. So Feddon devised this compensation linkage to minimise the tappet clearance difference. The geometry of the linkage ensured the tappet clearance 'error' was reduced by a factor of four.

This elaborate temperature compensation system was used on most of the Bristol poppet valve engines from the Jupiter to the Mercury and Pegasus. It was a cleaver approach but must have added a lot more to the cost of manufacture. Most other aero engine manufactures and virtually all auto manufactures use the simpler and less expensive approach of setting the tappet gaps wider when cold, so they become correct when working temperature id reached.




Below you can see the individual components of the rocker gear. The two sets of rockers are coaxially mounted on either side of the rocker box. THe full size engine had eight roller bearings in each rocker assembly. The model engine had to make do with eight phosphor bronze bushes




Production of eighteen sets (plus some spares) of rocker arms for the two engines was an endless, thankless task. It too a full year to fill these dog food tins with completed components for the two engines. I nearly gave up several times during that time.






Eventually, after all that work, I was able to build the nine rocker assemblies for the first engine.




Here you can see one of the rocker assemblies in position on top of a cylinder head. The rocker gear will be enclosed with a tight fitting cover to keep high altitude ice formation from interfering with the mechanism




Here are the rocker box covers in production. The top surface was profiled before the internal details were pocketed. The part finished covers were parted from the stock before the top front curved surface was filed and blended by hand




Here are a batch of nine rocker box covers, destined for the second engine.



All this is bringing the second engine near to completion. It was being prepared for this year's Model Engineer Exhibition at Doncaster in May. Unfortunately that's not going to happen. However, we still have the MEM forum. It's our worldwide virtual model engine showcase. So lets use that instead.

Stay Safe

Mike

[cnr6400]

 

Fascinating info on the valvetrain parts and design Mike. Great progress on the engine too! Thanks for sharing it.   

[nj111]

Phenomenal amount of complex work there Mike. It amazes me how you get so much done - compared to myself!  Please keep posting!

[Vixen]

Nick, Some of those photos go back 15 years or more. The two Mercury engines have been 'Work in Progress' for over 20 years. They stood collecting dust for several years before I came back to them. Hence the title 'Bristol Mercury revisited'.

Stay Safe

Mike

[awake]

The amount of work in this model is staggering - and the execution is absolutely brilliant! 20 years - now that's a lot of patience and dedication!

[Roger B]

Excellent    I do tend to think that the expansion compensation for the valve clearances was solving a problem that didn't really exist  Warm the engine up before putting it under load 

[Vixen]

Roger,

If you are going to solve a non-problem, then do it elegantly and expensively. The customers always fall for it.

Mike

[Admiral_dk]

If you are going to solve a non-problem, then do it elegantly and expensively. The customers always fall for it.

          Ain't that the truth 

Oh and I'm still lost for new words to hail your work Mike -> not many comments from me and strangely no questions either ....     

Per

[Vixen]

not many comments from me and strangely no questions either ....     

Per

Oh! I was waiting for some questions.

Mike

[kvom]

Just catching up after 3 weeks traveling and a week of jet lag afterwards.  Impressive work there as well as the patience required which is in some respects more worthy of praise.

[zeeprogrammer]

The amount of work in this model is staggering - and the execution is absolutely brilliant!

Well put. Every time I take a look, I have to reach down and bring my jaw up.

[petertha]

Magnificent!
- roughly what is the size of these square holes & how did you broach them?
- I'm trying to get my head around the pushrod contact geometry. That is quite the funky motion. Is the end of the pushrod like a cone & the underside of the rocker pairs ball-ish shape? Anyways, how did you machine the rocker underside ball-ish shapes consistently part to part?
- how are these metal to metal sliding surfaces lubricated?

You'll see my 5-cyl radial build one day soon, I promise. Its quite Neanderthal by comparison. But just know that your work is providing lots of inspiration over the rough patches to keep perservering.

[Vixen]

Hello Perertha,

Thanks for calling in. Forums like this one will become increasingly important as we all lock down and isolate.

The full size engine used multi splined couplings on the inlet rocker arms. So for the model I substituted a simpler square drive. The square hole is only 0.12" ( 3.1mm) across the flats. I formed these by drilling and some file work before bringing the hole to size with a homemade square broach. I made several of these brooches from hardened and tempered silver steel (drill rod?) as they did not last for long.

The push rods, each have a mushroom shaped end pad with a spherical end face, hard as flint. Each rocker arm had a hardened spherical insert which was free to rotate. The contact between the three spherical parts provided some rotation. For the model, I considered the spherical inserts to be impractical, instead I machined a circular pad (viewed from the front) on each rocker arm, followed by some hand filing. So I have cylindrical to spherical contact.

The lubrication of this area was minimal on the fullsize engines. They relied on hard felt pads in the cover plate soaked in oil. The pushrods and rocker arms rubbed against these oil pads. The maintenance schedule called for the oil pads to be soaked in hot engine oil every 40 hours. So, massive amounts of lubrication was not considered necessary. I have a real rocker assembly recovered from a crash site, Even after 60 years exposure to all weathers, the push rods and rocker arms are spotless. They knew what they were doing.

For your 5 cylinder radial, keep persevering.  Make one component at a time and you will get there.

Stay safe

MIke

[fumopuc]

Hi Mike,
every time I have a look at this thread, seeing all these parts and machining set ups I am getting more and more speechless.
There is a lot of space to increase my model engineering activities and this space seems to get bigger and bigger every time.