Converting a 600cc bike engine to run on diesel

[Chairchild]

Nope, that's not a mistake, nor is it a "pipe dream"... this is a project one of my tutors at uni is going to start undertaking, hopefully in the next few weeks!

I can't remember the bike model that the engine came from, but it's 600cc, as it's from an old formula student car (610cc being the max capacity allowed). Next years car is going to be electric, but this tutor wants the following years car to be DIESEL


Current plans include a full-power run, including the intake restrictor, in the standard layout. After getting the baseline power reading, new pistons will be machined up (from a block of solid ally!) Diesel injectors will be made to fit where the sparkplugs reside, a timing belt pulley will be made to fit the end of the crank, and an adaptor plate made to hold a common-rail pump, including tensioner.

After that, it's just a case of making the pipes to connect everything up, and then wire up all the sensors to the "dSPACE" computer - basically a £15,000 ECU emulator, to program the firing sequence and fuel pressure


Pictures as and when progress starts

[Relentless Rob]

Is a 600cc bike engine going to have enough compression for dieseling to occur without a trigger (constant glow plug)?

If you were to raise the compression by skimming the head you may get valve bounce or worse detonation. Provided of course it isn't a CBR600 engine. They have gear driven cams and you can't skim the head without a lot of faffing about with the cam drive mechanism.

Configuration is the key factor here. A Enfield Bullet 406cc diesel single engine would be a good place to start. Taking profiles from the single and scaling it down to a Ducati M600 (don't you chuffing dare!) or a Skidoo parallel twin. Scaling down again to run an inline four. So each cylinder is basically acting as an individual engine. That way you won't need all the faffing about with management as each cylinder is running mechanically.

If you can convert something like a GS or CG 125 to run on diesel then you just repeat the process four times and adapt a teapot (GS600) engine or CB600.

Sounds like an interesting prospect.

[Rippthrough]

That's going to be a complete bitch, by the time you get the bottom end to take the combustion pressures and sort the airflow out, you'll find you have a crap charge for diesel use in the chambers and the block is now flexing after you've uprated all the bottom end anyway.

Ah well, good luck!

[Renrut]

Wasn't there a company making aftermarket diesel common rail ecus?

Surely that would shortcut the £15K dSpace ECU emulator which you probably wouldn't want to put on a formula student car anyway.

[DaveRoverMan]

Simtek came up with the 'Bodylogic' setup.
They had it running a TDV6 (Disco 3) engine in a P38 Range Rover.

[Chairchild]

Rob, the engine as standard chucks out over 90Bhp - in diesel configuration, it's going to be pushing maybe 50bhp? With maybe 50lb/ft of torque.

And it's going to get FAR more power on common-rail than mech (without all the smoke that I love however )

Standard CR is approx 13:1, and the "ideal" CR for a direct injection diesel is 15:1, but starting is very difficult, and it won't particularly idle too well either. Pistons will most likely have domed edged, with the centre bowl for the injection flame - it's gonna be awkward, as the head's a hemi layout

Head won't need to be skimmed, as we'll just make the pistons as required

[majic79]

I thought the normal CR for a diesel was around 19-22:1, with CR Direct injection stuff going at 26:1

Block stresses are going to be the deciding factor, and one of the reasons why many diesels were thick wall cast iron up until recently. Without high CR, the temperature rise won't be enough for the otto-cycle to work correctly, so it'll need a constant glow-plug

The only reason I can think of to do this is to get a smooth (ish) 4-cylinder diesel for endurance runs.

dSPACE sounds expensive for what it is - unless you're mapping common-rail stuff, in which case, it's getting well out of my mediocre knowledge of mapping and into the realms of black magic precision combustion!

[Relentless Rob]

600cc, domed pistons and a Hemi-esque head?

Is it a Bandit (Suzuki GSX600N) motor?

[mjslonergan]

Mmmmm Think I would just buy a Kubota... but I am lazy and un-adventurous...

[owelly]

Didn't the Military develop KLR650s with DERV engines, and prior to that, they fitted Lombardini engines to their scout bikes? IIRC they weighed the same as a fully kitted Unimog and made less than 30 hp. Top speed was @ 80mph if you could find a dumb enough squaddie to ride it. But they could run on aviation fuel too so they'd smell nice.
I like the idea of converting a petrol to derv but as said, the stresses involved when bang happens often negates any further blow, suck or squeeze.

[owelly]

Or are we talking two-stroke diesel?.....

[DaveRoverMan]

Like the old 3 Cylinder Commers?
Or are you thinking something more like the Deltic?

[ACE]

Compression will need to be at least 18:1 to get it to run properly.
Most motorbike engines are highly strung, highly stressed pieces of engineering. With low weight being one of the key design goals. I can't see any semi modernish bike engine being robust enough to take the strains envolved. At least with out major re-engineering.

Some questions I think you need to consider (there are probably lots more)
Are the head bolts strong enough?
Is the cylinder wall strong enough?
Is the head strong enough, and stiff enough not to distort and blow the head gasket every time it runs?
What head gasket will be strong enough?
Is the crankcase strong enough to prevent the head bolts simply pulling out?
Is the piston strong enough?
Will the rings handle the increased stresses?
Is the gudgeon pin strong enough?
Is the con rod strong enough?
Will the big end shells take the forces or break up?
Is the crank shaft strong enough?
Are the main bearings strong enough?
Is the crank case strong enough to take the forces on the main bearings?

Good luck!

[Chairchild]

The pistons will be made from billet aluminium, using standard pretty much everything else.

The hard bit will be raising the CR to achieve ignition without external help - strengthening bits to cope will come on a "whatever explodes first" basis

If the engine turns out to be utterly useless at coping, we've no quarrels with machining a new engine from billet.... but we'd rather just run it until it dies, and have some fun learning from the experience


There's been quite a few motorbike engines converted to run on diesel over the years - and I don't mean shoving in a kubota D600 engine

And the dSPACE is being used, because atm, it's just collecting dust, and they want to make use of it! The last project it was used for was never completed: A realistic driving simulator, which gave feedback via a modified electro-hydraulic power steering rack, and brakes. Graphics weren't going to be top-notch, as it was more the control+feedback side that it was meant to incorporate. Again, I'll try to get pics next time I'm in the workshop... it's kinda hidden by a bucketload of junk atm though