Jav Jup Crankcase alloy

[Forumadmin]

With an impending repair needed on my crankcases I thought I would research the required filler rod specification.
The Jowett specified alloy used in aircraft during the war from which the blocks were supposed to have been made was DTD 133B which was replaced by BS L51.
MSRR equivalent is 9500/43 http://www.aircraftmaterials.com/data/weld/range.html

The actual compostion of my blocks is unknown, knowing the metal situation at the end of the war, as they would have used anything available.

But anyway....have faith

Specification for BS L51
Chemical composition of alloys to aluminium and two common filler rods .
BS L51……………………………ER5356………………..ER4043
Cu 0.8/2.0 ………….Cu 0.10 %................ Cu 0.30 %
Mg 0.05/0.20……….Mg 4.5-5.5 %........... Mg 0.05 %
Si 1.5/2.8…..….......Si 0.25 % …..............Si 4.5-6.0 %
Fe 0.8/1.4 …………..Fe 0.40 %................Fe 0.80 %
Mn 0.1 ………………………Mn 0.05-0.20 % ...… Mn 0.05 %
Ni 0.8/1.7
Zn 0.1 …………………..Zn 0.10 %.................Zn 0.10 %
Pb 0.05
Sn 0.05
Ti 0.25 …………………Ti 0.06-0.20 % …..........Ti 0.20 %
other elements
None..................Cr 0.05-0.20 % ............None


Condition Precipitation treated only

Mechanical properties
UTS 160 N/mm2
0.2%PS 125 N/mm2
Elongation 2%
Comparable Specs MSRR8041/RR50 (UK), A-S2U (France)
Retains mechanical properties up to 200°C.

Ref: http://www.tritechgroup.co.uk/downloads/Aluminium.PDF

Well that is a blow as I have loads of ER5356 and lots of other unidentified aly rods.

So now trying http://www.vbcgroup.com/products/welding/aluminium to see if they can supply.
ER4043 might do but getting the correct filler would reduce possibility of cracking.

[Jack]

I think this would go on the list of "things to pay a professional to do" given the stress on the block and need to get it 100%. Plus the potential work if it fails would be the thing in the back of my mind.

Worth knowing so that owners can advise said professionals though.

Jack.

[Chris Spencer]

"things to pay a professional to do"

Each to their opinion but I work with a completely different school of thought and practice - I always take an approach that if I can learn from what I am doing, pass on the experience to others, make just as good job of it has engaging a professional to do it in the first place (or better) and get the satisfaction of 'I did that' then there is no contest - if you add factors such has finding the said professional, transporting the item to the said professional then maybe having to wait several weeks before it's ready to collect then the reason behind having a go yourself becomes amplified - if I can financially break even I would still have a go myself - If I can save money then even better - then again there are things that I am simply not tooled up for i.e. engine machining and in cases like this I would not hesitate in employing someone that has access to right knowledge & equipment.

[Keith Clements]

Showing the corrosion on the webs.

Showing the cracks around some studs.

On the milling machine after welding with ER4043 4% silicon filler.

A test was done on an old water pump to check on flow and cleaning.

TIG was set to AC with 70% positive balance to try to clean out any remaining oxide after degreasing, sandblasting and stainless steel wire brushing. A 3.4mm electrode tip was used to cope with the increased temperature in the tip due to the positive bias. A pass without filler was used as part of the cleaning process to try to get a decent oxide free pool. 2.4mm filler was used. The foot pedal was used to control current which peaked at 150A. The block was not pre-heated as (I learnt) this can ruin the strength of the block. The block was not allowed to cool quickly to reduce cracking. Short passes were done to keep temperature of block below 350F, checked with a heat gun.

To prevent sag of the web it was supported. I used a thick piece if steel. Next time I will try a bed of sand.

Obviously learnt a lot first time around and it did take a long time getting the right tools and settings together particularly on the milling machine. The block needed to be clamped in four corners to get it level, but amazingly it was parallel once this was done. I checked with feeler gauge on mill tool, after a rough check with dial micrometer. The 'old motto' measure at least twice!

I did try to remove the stud but it would not budge even after the heat of welding. I thought it best to work around it!

I think I managed to get a reasonable weld but with such old contaminated metal there will be some porosity or bubbles and some grain boundaries and cracks. There is still some corrosion further inside but this is difficult to get to. I did patch up corrosion where the cylinder liner seal was getting narrow. Machining this piston liner face proved to be a challenge, particularly as some of the studs were not removed. It required removing the depth gauge on the milling machine so the head would move low enough. As did machining the face that locates the top of the liner, as it was so close to the stud and required a long large reamer and a selection of Morse taper converters.

[Keith Clements]

The various hallmarks on the case.

[Keith Clements]

There are a few stages to this process.
The first is to thoroughly clean, compressed air blow out and then apply the special cleaner supplied with the pack of three aerosols. Paper towel was used to assist this process.
The blocks were then warmed on the wood burner to 20C as they had been around 7C.
The red indicator fluid was then sprayed onto all surfaces. Whilst this was finding its way into the cracks I went in search of UV lamps (finding a SAD lamp and two fly killers that contained UV lights). The reason is the indicator shows up better in daylight. I need not have bothered as the sun started to shine onto the work place!

After 20 minutes the indicator was cleaned off using paper towel with the aerosol cleaner applied. Then the third aerosol containing chalk was applied and left for 10 minutes.

Although the indicator did show up in many places, this was often caused by poor cleaning before applying the chalk. Some of the porous bits that showed up were not thought to be serious enough for rework but one stud hole that had been welded with the stud in place was not good. So the stud was cut off with a grinder as it had refused all attempts to be unscrewed. The surface was faced with a mill and centered before drilling out the core of the stud. A stud extractor was tried part way through this process but that did not work either. So the next size drill was put through and the remains of the threads removed with a start tap and then finish tap.

Now, with all traces of steel removed, the crack was ground out and rewelded. It was much easier this time without the contamination from the steel stud and any crud around the thread. Unfortunately the block had been removed to do this welding so the mill had to be realigned to drill the new stud hole.

After coersion from Peter Pfister I decided to put a helicoil in rather than just rethreading. The helicoil is about 40% stronger than a thread in the aluminium but it does require a larger hole, reducing the amount of metal between the stud hole and the water jacket which is very thin where it meets the liner.

All areas were checked for cracks and porous holes, so hopefully the blocks are now good enough.

The final act was milling out excess weld and fitting the liners. I think if I did this again I would remove all the studs. This block had been lying around for years and I do not know its history, so it is an experiment!

I am now checking the crankshaft and cam shaft alignment.

[PJGD]

Keith,
Please report in a similar fashion with photos on how you are going about checking the bore alignments.

Thanks

Philip

[Keith Clements]

I will have to use the bed of the milling machine as the reference surface which has shown that it was in agreement with the original machining of the block in that the head face was parallel to the face that joins the two halves of the block together. I would guess that the measurement was parallel over the faces to within 2 thou perhaps 1 thou allowing for feeler gauge and mill alignment errors. But I was not scientific about estimating the error. I will check with the parallel bars I inherited from my father and may be able to go to within .0005". Any tricks on how to do this as my experience is limited?

The mill also showed that the piston liner mating surface was parallel to these two surfaces. I did not check that the liner bores were at right angles to these surfaces but knowing perhaps the poor state of the Jowett Machines at the end of the war , I should have done.

Checking to see the two bore centre lines intersect and are at right angles to the crankshaft centre line is not going to be easy. Any ideas?

Yesterday I tapped and died and cleaned all the threads and reamed out all the dowel holes on the blocks. But discovered that one of the rear timing cover bolt holes was partially stripped. This will need to be helicoiled. There is little meat between it and the oil delivery hole and it needs to be drilled at an angle (90deg to the face) so will need careful setting on the mill.

[Alf Heseltine]

Hi Keith, maybe a check on the crankshaft first to make sure it turns when the crankcase is torqued up.
If the shaft is nipped will you consider line boring. Hope all goes well

Alf

[Keith Clements]

Alf, that was my next move having just re-threaded all the threads , reamed the dowel holes and cleaned out the oilways.





Note the plug of carbonised oil after the plug is removed




Note that all three of the top tie bolts have oil pressure in them as can be seen when cleaning them out with some aluminium filler rod. Also interesting when doing this down the rocker shaft feed, cam shaft and cam follower (tappet) feeds.

As an aside, to do this I made a block holder out of a part of a large piece of wood.
See this topic.

[Keith Clements]

Well this really did take longer than it should, but I treated as an exercise in measurement.

First job was to adjust the engineers spirit level which is potentially extremely accurate. Ok it may not be up to the latest electronic gizmo standards but I reckon it is good for 1 thou in 10 inches. I did this with my dad's ground parallel bars to obtain a flat surface and attempted to find the most stable surface I could which was the 2 inch thick workbench. I found if I leaned on this the level would detect the movement! It was adjusted with a pair of nuts at one end so it read the same when rotated through 180deg. Then a thou shim was added under one end of the parallel bar to level it and then the process was repeated.

The lathe and then the mill were levelled again. A 1mm shim at one end of a 4ft lathe was enough to move the level a graduation.


It was very important to clean all surfaces as the smallest piece of dirt would throw out the level.

With the mill bed now level in both directions the block was mounted and wedges put under until it was level on the oil housing face. The adjustment was checked with the dial gauge.

[Keith Clements]

So having this set up on the mill should I cut some O ring grooves to seal the oil galleries?

If so, any ideas on how to do this? Do I need a special tool?

[PJGD]

Keith,

I would say no, don't machine o-ring grooves in the crankcase. First, the material between the oil hole and the bolt hole is marginal, and second that suggests that you would be assembling the filter housing to the block metal-to-metal with no gasket. It can be done, but I would prefer the intermediate plate to hold the o-ring in place. It holds the filter housing up so that there is better alignment at the interface for the angled oil drillings for less flow restriction, and you can probably end up with a larger section o-ring with more crush hence greater tolerance for uneven or non-parallel faces.

Philip

[Keith Clements]

I agree Philip. I have used the aly plate method on three engines so far. Another method is to insert a sleeve inside the oilway but this requires aligning vertically which is not easy to acheive. This joint is probably the worst bit of Jowett design!

[Keith Clements]

Having spent an inordinate amout of time levelling the milling machine I then set about checking the set up of the block before centring the tapping drill for the helicoil. This I did first with a drill about the core size of the existing thread and then with the tapping drill in place I checked its cutting edges did not deflect when lowered slowly into the hole. Is there any other trick to do this?


Then the depth stop was set up using a depth vernier.



The tap was put into the mill chuck which was turned by hand to start the thread perfectly vertical.


Then the tapping was continued by hand with the chuck released.


The helicoil was then inserted by the special tool that drives the coil into a threaded cone to compress it radially.



So now the blocks are ready for cleaning. After the operation I checked the level of the mill and found it needed a 10thou shim under one corner. I also checked the lathe to the same accuracy and that needed a couple of shims under two corners. So then I cemented under both machines so that they do not move around. If there is any settling in future I can still insert some shims but it may take some jackin up of the bed rather than the base. Perhaps i should have put bolts in the floor and devised a method of having screw adjustment.