The DOHC 250 Velo Engine: Rebuilding the Crankshaft

The objective here is to build a crankshaft that has a 68mm stroke and with mainshafts that are appropriate for the Mk 1 cammy engine to accept the bevel drive on the timing side and a taper on the drive side for the crankshaft shock absorber assembly – standard “K” practice.

Although I have a spare "K" crankshaft, this is of no use as it has a 74mm stroke and the starting point is an MOV crankshaft (an eBay find) that I have stripped and cleaned and which has the required 68mm stroke.

A few words about disassembling one of these crankshafts:

The first step is to remove the big-end nuts and press out the big-end pin – quite straightforward.

Removing the mainshafts, however, has to be done with care. When they were assembled a ¼” BSF threaded section was positioned midway between the shaft/flywheel interface into which a truncated screw was inserted and was put there to prevent the shaft rotating in the flywheel (in addition to the interference fit). It can be clearly seen in the picture below.

This needs to be removed and I do this by drilling to a depth of ½” with a 7/32” diameter drill

so that after pressing out the mainshaft (from the outside in) only the remnants of the screw remain and there is no damage to the flywheel.

After degreasing and with the smaller parts in the tumbler overnight (except the mainshafts, which will be scrapped), the collection of bits looks like this.

The separate tasks to rebuild the crankshaft are:

-       Replace the big-end bearing, which is worn

-       Remove material from the flywheels to give the correct width to fit the available space between the bearings in the crankcase

-       Replace the mainshafts with those suitable for the “K” engine and with a central hole on the timing side to accept a quill

-       Assemble and check

Before disassembling the crankshaft, I made quite a few measurements. One of the most important of these is the internal distance between the flywheels, “B” in the picture below.

The second measurement that is of importance is the available distance between the bearings when assembled. This was done by making 2 new solid “bearings” in mild steel that are exactly the dimensions of the actual bearings (22mm x 50mm x 17mm),

dropping these into place in the heated crankcases

and measuring the distance between them when assembled using a telescopic gauge that can be easily inserted through the main hole for the cylinder barrel.

The MOV flywheels are wider than the "K" flywheels and both were reduced in width to fit the "K" crankcases.

I have retained an overhang (0.040”) on the outer rim width of the flywheels.

I was very lucky to find new mainshafts. I was talking to Nick Payton on the phone a few weeks ago (I have known Nick for many years – since my early days of Velo-ing) and he mentioned that he had a couple of KTT mainshafts. He sent me these in the next post

which are exactly what I needed. The section that fits into the flywheels is the same diameter and they have the correct dimensions to accept the bevel drive on the timing side and a taper for the shock absorber assembly on the drive side.

A bit of work is required to set them up. On the timing side, there is a drilling along the axis of the shaft that supplies oil to the big end and for which a radial 1/8” hole needs to be machined to line up with the corresponding hole in the flywheel. However, the axial drilling in the mainshaft would have needed to be extended and, in my opinion, would have come too close (or, worst scenario, broken through) into the hole that was originally used to grind the shaft between centres.

I therefore drilled the radial connecting hole at the end of the existing drilling and made a short slot (0.1” deep) to connect it to the hole in the flywheel.

The other end of the timing-side mainshaft was bored 5mm diameter to a depth of 0.75” in readiness for accepting a quill. 

Both shafts have a moon-shaped segment removed to allow the big-end cage to fit (see the 1^st and 3^rd pictures down from the top of this page). The diameter of the recess in the flywheel is 2” and to remove material of the correct size in the end of the mainshaft I used a 50mm diameter x 4mm thick slitting saw mounted on a mandrel.

This slitting saw is made of HSS and needs to be run slowly using small cuts and plenty of lubricant as the material is tough and the shaft is case hardened (Grinding would be better – or machined before heat treatment but I don’t have these as options. There are also alternative ways to machine this using either the milling machine with a smaller carbide cutter or the lathe but the way I used has a far simpler setup although HSS is not the best material for cutting through case hardening!). The machining operations on both shafts turned out OK.

The positioning of the cut is important to ensure that the oil feed hole (slot in this case) in the shaft aligns with the hole in the flywheel feeding the big end bearing on the timing side and, on the drive side, the cut is positioned in the correct place with respect to the exit hole for the timed breather.

On both mainshafts I found that the diameter of the portion that enters the flywheel was 0.002” larger than the mainshafts that I had removed from the MOV crankshaft. This would have given an interference fit of 0.004 and would end up in either broaching or distorting the flywheels when pressed in. The surface of the mainshafts is too hard to machine by any means other than grinding. As these mainshafts are new (with good centres) and would originally have been finished by grinding between centres it was possible to remove this small amount on a cylindrical grinder. I don’t have cylindrical grinding in my workshop (only a tool post grinder – which is not sufficient here) and Peter at Jayess Tools did a great job in finishing these accurately to size to give a 0.002” fit.

The hydraulic press registered around 5 tons for pressing in the mainshafts and using Loctite 638 as a lubricant.

As I mentioned earlier, the mainshafts in the MOV crankshaft had the additional “belt and braces” fixing of a ¼” BSF threaded section positioned between the flywheel and the shaft. I had taken care not to encroach on the threaded portion when I drilled them out before removal as the plan was to replace them. As there is only “half a hole” after the new mainshaft had been pressed into place I couldn’t use a regular twist drill to reform the hole – it would wander off-centre, so I used a slot drill to reform the clearance hole

to be able to re-tap the thread back to the original size

before inserting a Loctited high-tensile grub screw.

The big-end had quite a bit of up-down movement before I had dismantled the crankshaft and would need attention. There was some visible wear on the ends of the pin on both sides where the bearing cage had been rubbing.

This phenomenon has long been recognized as an issue in caged roller big-end bearings (see, for example, Phil Irvings book Tuning for Speed, 3^rd edition, page 84). There are a number of mechanisms of how a soft material can erode a harder one (aluminium vs hard steel in this case) and I have no idea which of those mechanisms applies here – anyway, it happens. The damage it not all the way around the pin – it occurs either side of TDC and the loading on the assembly at this time of the engine cycle is quite complex – inertia loads resulting from the deceleration/acceleration of the piston, change in the direction of rotation of the connecting rod relative to the pin and cylinder pressure acting on the piston crown.

When I measured the central portion of the pin where the rollers are in contact, I found it was completely unworn and measured exactly 1.250”. Similarly, the bearing outer which is pressed into the connecting rod, is also unworn and measured 1.625”. Incidentally, when I measure these using a micrometer, I use a set of slip gauges

to check the calibration of the micrometer as absolute accuracy is needed here.

The really good news is that the rollers, which have nominal dimensions of 3/16” x 9/16”, were found to be 0.0015” undersize – which would have resulted in an additional 0.003” clearance in the bearing.

A new set up rollers

and the big-end is “as new” in terms of bearing fit. I can live with the wear that has occurred to the outer part of the pin and which is not too important for the functioning of the bearing; future wear on the pin extremities will, in any case, probably be reduced with new rollers fitted. Total cost of the rollers (including delivery) was 17 GBP compared to the cost of a new big-end assembly at 400 GBP so I’m more than happy with this outcome.

The final stage was standard crankshaft assembly: the crankpin was inserted into the timing side, ensuring that the oil holes were aligned, and the nut on the outside of the flywheel tightened. I put a small black mark with a felt tip pen on the pin to indicate its position - just in case it rotates when tightening the nut and putting the oil hole out of alignment. The bearing and connecting rod were then put on and the pin inserted into the other flywheel, pushed “home” and the nut screwed on. Adjustments were made (with a soft-headed mallet) before final tightening with plenty of checking for alignment on the crankshaft alignment jig.

I ended up with runout of 0.001” on the timing side and 0.0015” on the drive side which is not too bad for a crankshaft that is probably around 90 years old.

Although the crankshaft has been rebuilt and basic checks such as central positioning of the connecting rod etc. have been made there is still work to be done to complete the fitting of the crankshaft. The balance factor needs to be checked and the distance between the flywheels is around 1/8” less with the MOV big-end/connecting rod assembly compared to the “K” crankshaft which means the mainshafts are 1/16” further inboard which creates complications in fitting the shock-absorber on the drive-side. The location of the crankpin (closer to the centre resulting from the shorter stroke) also caused a problem by interfering with the main bearings which needed sorting out  ….more next time.

The DOHC 250 Velo Engine: The Resurrection Plan

 

It’s always good to have a plan. In practice, most plans usually don’t work out exactly as anticipated and need modifying along the way because of something that you hadn’t thought of but without a plan a project has no structure.

So, what is the plan? The main elements so far are:

1)  Check if the engine will actually fit in the Mk1 frame – because that’s what I have available

    2)  Check where the piston will be at TDC with a 68mm stroke and determine what modifications might be needed

    3)  Build the crankshaft with a 68mm stroke (to give 250cc) and with mainshafts the same as the 350cc engine to accept the bevel drive on the timing side and a taper on the drive side

    4)  Make offset/eccentric studs/adapters to move the centres of the holes for the cylinder barrel/head/cambox inwards

    5)  Modify the lubrication system to provide a positive oil feed to the cambox (the cambox has 2 entries for supplying oil and one drain pipe to scavenge the oil by gravity)

    6)  Check the cambox internals and set up the bevel drive shaft/couplings etc

    7)  Complete the engine build – piston/rings/valve timings/carburettor/magneto etc.

Although the project milestones can be summarized easily, most of these tasks are a substantial amount of work.

The first step was to check if the engine would fit in the frame. This was fairly simple as the chassis for the “cammy special” has already been built so it was just a matter of trying the engine for size.

It didn’t fit! The cambox would not fit under the lower (curved) frame rail; after removing extraneous bits and pieces (gearbox etc.), chopping off the extended bit of the frame that would support a hand gearchange lever

and removing the bolts connecting the engine plates to the frame the engine could be lowered into a position such that the top of the cambox cleared the lower frame rail and could be built or removed with the engine in-situ.

In the above picture, the steel rod that I have inserted through the cambox and cylinder head bolt holes is to ensure that they are aligned and the cambox valve pushers are resting on their respective valves with the cams set to “valves closed”. In practice, the engine height would be increased slightly with a valve clearance (say 0.020” plus a cylinder bases gasket (0.040”)) plus (as will be seen shortly) a spacer of around 0.2” under the barrel but this exercise is sufficient to determine that the engine will need to be lowered by around 0.75” at the front and 0.5” at the rear. The carburettor that I found among my collection is an AMAL RN and has the correct choke size. It fits OK although a remote float chamber will be needed – luckily, I have one.

One thing that became immediately apparent is that this engine will need a considerably longer K-49 shaft connecting the top and bottom bevels. The picture below shows a standard-length shaft held in position.

I have made a first estimate of where the piston would be relative to the top of the cylinder at TDC. I could calculate this but, as I have acquired an MOV crankshaft for this project

which has a 68mm stroke it is just as easy (and probably more reliable) to make up some dummy steel “bearings”

to be able to insert the crankshaft into the “K” crankcases.

It turned out that the MOV crankshaft is about 0.160” longer than the distance between the internal faces of the MC22 Bearings (22mm x 50mm x 17mm) that would normally be fitted and so I reduced the width of the “bearings” although on the timing side this didn’t help because the big-end nut fouled the “bearing” and this meant that I couldn’t completely close the crankcase halves.

Nevertheless, the assembly was sufficient to be able to put the piston in place on the con-rod (luckily the MOV small end is the same diameter as that of the DOHC piston) and slip on the barrel to see where the piston would be at TDC.

So far so good.

However, when I put the cylinder head in place the exhaust valve fouled the piston and so the barrel will need to be raised by putting a spacer beneath. I don’t know exactly how much yet – my guess is around 0.2” and I will need to make an accurate estimate based on piston/valve lift curve profile/compression ratio etc. in due course. As I mentioned, this will have a knock-on effect on overall engine height, length of the “pear-shape” section on the cylinder retaining studs etc. that I’ll need to take into account later in the project.

The lubrication system will need completely revising. In the picture below of the underside of the cambox, I have circled round the oil feeds that direct oil to the extremities – they squirt oil directly onto the cams - and also the drain pipe which would have scavenged the oil back into the engine (or, more likely, directly into the oil tank).

This suggests to me that this engine was designed to use a lubrication system akin to the Mk V KTT in which oil, under pressure and directly from the pump, is fed to a quill into the end of the timing-side mainshaft via internal passages and to jets in the cambox that direct oil onto the cams; this became standard practice on later engines.

The picture below of the (as received) KTT 581 engine shows the feed that comes from the pump and is connected to the cambox.

To do this, the lubrication system had been completely revised. The oil feed at the exit to the oil pump that on earlier engines pressurized the lower bevel chamber and then, by virtue of the pressure in the chamber, fed the oil up the drive to the upper bevel and via the slots in the shaft and large phosphor bronze bearing into the cambox had been blocked off. The internal volumes were not pressurized and oil was fed strategically to the crankshaft and camshaft as described above.

For the DOHC engine to operate on the same principle a number of modifications are required:

 1)  The small passage that connects the feed side of the pump to the lower chamber (indicated with an arrow on the picture below)

 

 

needs blocking off. (I have made this modification previously for the Velo oil pumps on both the AJcette and the AJS V-Twin – see here) by building up with braze and machining.

2)  The drive-side crankcase needs modifying with a union to take the oil directly from the feed side of the pump (as in the picture of KTT 581 above)

3)  The timing gear cover, K45/2, will need a quill and an oil feed to provide oil into the end of the crankshaft for the big-end

4)  The crankshaft will need to be modified at the end of the timing-side mainshaft to accept the inserted quill

5)  Oil feeds to the cambox (2) and to the upper bevel will be needed. A drain back to the oil tank will also need to be added.

Needless to say, the bevel drive shaft will need to be considerably longer but that will be one of the last parts of this project and when I see where the cambox is eventually positioned.

So, I have the outline of a plan and some of the implications that go with it.

In preparation, the MOV crankshaft has been stripped and cleaned, the valves removed from the cylinder head, the valve gear cleaned up in the tumbler and the main aluminium components (except the cambox) have been vapour blasted.

The first sub-project will be to rebuild the 68mm stroke MOV crankshaft with mainshafts appropriate for the “K” engine.

The DOHC 250: INDEX PAGE

 

The DOHC 250 Velo Engine

The DOHC 250 Velo Engine: The Resurrection Plan

The DOHC 250 Velo Engine: Rebuilding the Crankshaft

The DOHC 250 Velo Engine: The Lubrication System – Part 1