Kings Langley garage meet

[Keith Clements]

Thanks Philip for detecting the mistake and probable reason for it. I have corrected the articles you found were incorrect.

But for those who may need help understanding the issue......
First you need to define how you number the cylinders.
Then you need to see in which order the cylinders are on their compression stroke.
Then you need to see where the rotor arm lines up with the post in the distributor cap to fire the cylinder at the end of its compression stroke.

In the case of the Jup and Jav, cylinder number one is on the left when sitting in the car and at the front of the engine. They are then numbered from front to back so that number 4 is at the back on the right. The brass plate on the bulkhead is specific on this with a small picture and the firing order 1, 4, 2, 3

Where the leads attach to the distributor depends on if the drive to it has been correctly installed and if you have a plug in or screw in cap.
My Jup has a plug in cap with no 1 at the back and no 4 at the left when looking from the front of the car, then 2 and 3 moving anti clockwise.
The reason I raised this is because those used to other vehicles could wrongly assume cylinder numbering, firing order and rotation of rotor.

That is why I interpolated the Jowett firing order to the VW cylinder numbering. Yes you then have to take care because the rotor is rotating the opposite way.

If you read my post again you will see I have corrected. I was trying to show any difference in the couples being generated.

Here are some locations in JT that I have found were incorrect!!! Anybody anymore?
viewtopic.php?p=23545#p23545
https://jowett.org/jowettnet/dt/tech/jt ... lletin.htm which points to
https://jowett.org/jowettnet/dt/tech/jti/jav.pdf This has a rewrite of the Bill Boddy Article !! but is elsewhere correct.

These are correct
https://jowett.org/jowettnet/dt/tech/mo ... er/mt4.gif

[Keith Clements]

An original flywheel weighs 11.20 kg whereas Mike's aluminium one weighs only 4.18 kg.

But Mike was unsure that it might not have issues so we decided to lighten another one which I did yesterday and got it down to 8.36 kg.
I had checked out the dimensions of Mike's and noticed the crank to clutch face distance to be 1.231mm less than the original. This was measured first with a dial micrometer and then with the more accurate DRO on the mill. Whether this causes issues with clutch operation, we will see. I also found the three oil release holes had not be drilled so I put them in.

Today Harry has taken the lot for balancing.

The following pictures show the allen countersunk bolts that hold the clutch face and the starter ring to the aluminium.

There are also two pictures of a possible donor that was rejected because of a worn starter ring.

[PJGD]

Keith,
Is there a reason [other than perhaps you forgot about it] that you did not copy the JCL drawing of their lightened flywheel (that I assume they used on their competition engines). You have the drawing in the collection of JCL drawings on this site. It is dated 31 March 1952, so I assume that it was used on the '52 Le Mans cars.

If you have measurements of the flywheel that you modified, I may be able to work out a comparison of the total weight and the inertia between the two designs.

[Srenner]

Mmmmmm, an original flywheel at 2.5 lbs??? Miracles at Christmas!

[Srenner]

Interesting thing on flywheels, as there appears to be two different versions. My dad had the original factory blueprints on lightening them (now with JCC) and I used to stare at them for hours as a kid. I sent a copy to Tilton when I had my aluminum ones made and of course, they did it their way. 30 years on and no issues.

[Keith Clements]

Sorry decimal point error! Should be 11.20 kg not 1120 gm. Now corrected.

I was erring on the conservative side on the meat behind the clutch face. I will be lightening another this week to remove more metal. download/file.php?id=7736&mode=view

Scott, maybe aly requires a different shape. I am not convinced that Jowett had the knowledge or tools to design the optimum. Have you pic or can you draw what Tipton designed?

Philip, Having seen a flywheel disintegrate at Santa Pod drag strip in front of me I would not want it to happen. Dropping the clutch at 7500 rpm on a Jowett wheel with all that rotating mass puts a a lot of torsion and centrifugal force on it. The Jowett design has too many sharp curves for my liking so I prefer more radial strength . The oil capture recess is bad IMHO.

[Srenner]

No good pics of the Tilton. It is standard period stuff from them with a replaceable friction surface held in place by countersunk #10 screws with jet nuts. The nuts can be seen in the picture. The height of the flywheel remains stock as I did not want a competition clutch assembly. The mounting boss for the crank is the same and it has a generous radius (bowl shape in profile), so no sharp edges. IIRC, the ID of the recess for the mounting bolts matches the ID of the overhang of the original, gaining wall thickness in that area. This flywheel weighs 11 lbs.
Tilton:

Alum flywheels were the rage for quite some time, but wear/worrying on the mounting holes and registers has lead many to return to steel.

Next up, a reference photo: MGB current practice with AP racing clutch. The steel flywheel is only as thick as the ring gear (roughly .5") at the outer edge and is undercut on the backside. Outside to inside, it concavely arcs by the middle of the lightening holes to .200" thick, then ramps up to the register area.

Two other light Jupiter units done by persons unknown. The first has simply been skimmed across the back, leaving a substantial amount of material around the center. This one is 13 lbs.

Surprisingly, this one is 14 lbs in spite of the work it took to drill all those holes and relieve the back.

[Keith Clements]

Thanks Scott. After our marathon WhatsApp conversation last night trying to sort out why your pictures did not come out properly.
Clearly there are various ways to get a lighter flywheel but there should only be one which meets the design brief.
I also recalled this morning the time when John Blanckley in his Javelin on the Marathon in Slovakia managed to shear the flywheel bolts when the overdrive dropped out on full chat going up a hill. This emphasises the torque that can be put through from the clutch to the crank and so meat around that oil capture recess is probably quite important. The critical stress line is between the flywheel bolt hole and the clutch friction plate face. This passes very close to that oil capture recess creating bending moments and stresses in that area. One should be able to work it out by first estimating what possible force went through that area in the Blanckley situation above where the road wheels were suddenly driving the engine 15% faster than they were a second before. This was probably considerably more than the 100 bhp (or perhaps more correctly 100 lb/ft) you might expect. You should also add into that the centrifigal force of rotation at possibly 8000 rpm. Could aluminium stand that?
See viewtopic.php?p=45394&hilit=torque+curve#p45394
download/file.php?id=22562&mode=view

I also realise that some heat soak is required under some circumstances with repeated clutch changes, so perhaps meat behind the clutch face is also necessary. One area that perhaps can be reduced is the annulus where the clutch bolts to the flywheel. There is torque being transmitted momentarily through the 6 holding set screws, but even so it might be possible to replace these with HT nuts and bolts and then remove metal to reduce thickness to say width of starter ring.
I also think that the starter ring needs to register and thus the flywheel needs to be just a bit thicker to accomplish this at the very outside. Sure lightening holes can be added between clutch set screws and starter ring, and even possibly extended between these bolt holes to up to the clutch friction face. This might even improve cooling the clutch.
It looks like 5.9 kg or 13 lbs is a conservative goal for the steel flywheel, so I will need to find a bit more meat to remove from my previous attempt.

[PJGD]

As a further contribution to the discussion above about crankshafts and particularly 3-main bearing examples, I would recommend this link: http://www.n56ml.com/corvair/crankshaft/

It discusses crankshaft breakages in the Chevrolet Corvair engine, an approx 2.5 litre 6-cylinder air cooled HO engine of the 1960's, when modified for use in light aircraft where of course reliability is key. It is quite a long article, and although general aviation is not directly applicable to our situation, there are a number of take-aways which are relevant. For instance, the Corvair crank is a 4-bearing design with flying webs between crank pins; note that they have had to add a fifth bearing to take the propeller loads.

Specifically, the author stresses the benefit of nitriding (which of course we already know about) although we should probably look into the pros & cons of gas nitriding vs ion nitriding: https://www.ahtcorp.com/articles/blog/i ... -and-cons/. But the main take-away is the importance of achieving a smooth blend of the 0.1" [2.5 mm] fillet radius at the conjunction of bearing and web side wall. Again, we are well aware of the need for the 0.1" radius, but the need to closely inspect the fillets especially after a re-grind is critical. Note also the recommendation to stress relieve, then nitride, then shot peen the crank to get a significant increase in fatigue strength.

[Keith Clements]

Harry Naerger brought the bits bits back from balancing and Don Bell joined us to observe (but ended up helping to change the diff in Amy's Jupiter.
The balancers report showed Mike's aluminium flywheel was 2760 mgm out of balance and the flywheel I had lightened was 9.6mg out. Consequently a lot more metal had been removed from the already light aly flywheel and the steel one only a couple of small drillings.

So the crank was put in and then I remembered we should put a balance pipe in. But no, there was no pipe work in the tappet chamber and the heads had an aly plug inserted in the balance pipe hole. So we have to make an external balance pipe between two inlet stacks that will have to be fabricated. So have to wait for Harry to source some carbs.

But we will need an oil feed to the rockers. So I made up a pipe and found some unions to make up one. But the other side had had the base of the stud pushed through , so there was nowhere to seat the flanged pipe. I will attempt some welding later after we discussed various alternatives. I know some have an external feed to the rocker shaft but I have no experience of this so will stick with what has been good for me over many years rallying and racing. The feed holes in both studs were blocked, one with araldite (perhaps to fix the brocken casing) and the other with solid oily muck. Note to self always check those holes.

One of the stud holes still had the copper flange of the pipe in it , so a tool was made up ( a sawn off nail) to remove it.

[Keith Clements]

Today we will look at adapting this that I unearthed from my cellar where it has been for 30 years.

I would like the experts to say why an external balance pipe is desirable especially for racing and what are its design considerations with regard to torque and power, rev peaks and standing wave calculations etc.

[PJGD]

Well, I am no expert in carburettor systems; my expertise is with diesel injection systems. Nevertheless, it is clear that from an originality and aesthetics POV that the internal balance pipe is much preferred. What the effect of diameter, length, and location is on mixture strength and distribution across the speed and load range, I have no idea.

I recall Gerald Palmer saying that during development that they tried several types of carburettor including Amal, but got the best performance from Zenith so went with that. Your external balance pipe with SU carburettors is interesting and would be fun to try, but my guess is that a relatively large balance pipe would be required with SU's, particularly at low engine speed otherwise the suck-suck-nothing-nothing from each intake port on the cylinder head would allow the SU piston and needle to jump around to much. This would have the effect of making the first cylinder to suck be weaker in mixture strength than nominal, and the second one to be stronger. This happens already but would be aggravated with the SU's, although I suppose that a thicker oil in the damper might be the solution.

The somewhat longer inlet tract length between butterfly and inlet valve with your SU set-up might change the torque curve shape and peak torque speed slightly.

[Keith Clements]

I have no intention of using the SU carbs apart from possibly giving them a test. The system looks terrible so probably is.
There are quite a few people who have used external pipes and will now try to find out their thoughts.

I would think that different carbs are affected by the change in velocity of the air in different ways. SUs I would have thought have a built in averaging inertia based on the viscosity of the damping oil and the mass of the piston. DelLortos would be more instantaneous in their response but this might not deliver the desired charge to the cylinder. As I discovered on the Javelin with standard air box when starvation occurred at 70mph.

[Srenner]

The balance tube has the greatest impact at idle and just off idle carb openings. Somewhere on the net are some motorcycle groups covering modifications to racing old twin carb British bikes.

While I am unaware of how the diameter of the tube plays out, I can say the first thing done on "log" style manifolds on a racing motor is to block off the balance tube openings. The goal is always to smooth the airflow to keep things as laminar as possible. A typical SU log has a rather large passage, perhaps 1/3 or so of the intake runner, intersecting it at right angles and creating turbulence. A very specific epoxy is used inside the SU log. I will ask.

We had a (an?) MGB head flowed by bench specialist that had SU and DCOE Weber set ups. In both cases, the manifold crossover was plugged.

The opening of the balance passage in the Jowett head is smack in the middle of the parting line of the siamesed ports. This has to be the worst possible location in my thinking, as the gasses must pulse past the ridge and the hole. Also, should the ridge be smooth or sharp?

And in a phone conversation, Keith pointed out the sonic/pulse waves will have effects.

I'm with you on the SU set up as pictured. In racing BMC engines with SU carbs, the first thing is to NOT put oil in the damper unit. This gives an immediate response of the piston/needle to the butterfly movement. As most of the time will be WOT, not cruising, stability of movement of the piston/needle is not paramount. I often ran ATF just to have a little bit of stable idle with minimal resistance.

So, my question is if the balance tube is really necessary. I'm with Phillip on the intake runner length mods shifting peak power and torque. I guess if we had rolling road or dyno time available, we could sort all this out!

I suppose one approach is to look at the Ganes motor/carb stack as a guide.

[Keith Clements]

I had suggested to Harry that we try without ( mainly because at the time I had not remembered I had an external one)
Perhaps we could do and just fit standard carbs to start with.
Then try with the external pipe.
More in this link

In general, short intake manifolds produce more horsepower at high RPM and taller intake manifolds produce more torque at lower RPM. The power peak between "shorter" and "taller" intake manifolds is between 300 RPM to 500 RPM which is not significant for most applications. The OEM intake manifolds used with Weber are 68mm and the tall intake manifolds used on the 906, 911R, 914/6GT and 2.3 liter 911ST race engines are all 98 mm tall.

So a longer distance between valve and carb throat aids torque and shorter aids power, but not by much.

The main reason for taller inlets was because of 'blow back or reversion' when both inlet and exhaust are open.

Intake manifolds used for performance applications require careful selection or modification to match the carburetor bore diameter and intake port configuration in the cylinder heads.

The assumption that a race engine is tuned at wide open throttle (WOT) may cause significant running problems. Tuning for power at a specific narrow rev range may also give problems. The actual specification of this race engine and the resultant parameters to feed into design will take some time to gather. The valve sizes and their seat angles will need to be measured. The cam profile, open and close angles and valve lift need to be measured. We need to measure and average the head inlet and exhaust throats. Or we just get some fancy flow measuring kit.

https://www.cartechbooks.com/blogs/tech ... and-design
https://www.speed-talk.com/forum/viewtopic.php?t=729
https://bmwmotorcycletech.info/InExTuning.htm A good explanation on the inlet tract.

The flow of the balance tube will affect the fuel curve in the lower 1/4 throttle range quite a lot.
If you install fittings with an ID of ~3/8th inch and install a flow control valve in the line you can use it to help fine tune the throttle response in that range. It works rather well.

So my conclusion is that the balance pipe is there to aid balancing the carbs at lower throttle. Most Zeniths are difficult to balance anyway so the balance pipe is worth having. The balance pipe may affect the jetting of the carbs as there is some sharing of charge. Jowett chose jets for a specific compromise. With worn jets on old carbs and different requirements, the jetting is what to concentrate on.

Meanwhile I welded up the stud hole and redrilled for the oil feed pipe. There was a week of delay because the gas valve failed on the TIG welder and a new one is taking a couple of months to arrive from China. But Harry acted as the gas valve whilst I welded.

Chris came around to use the press for his Range Rover Wheel bearing.