V-Twin 100cc Design & Build

[Craig Booth 1Participant @craigbooth1]

Hi all,

and so on to my next project. This is going to be a big challenge but have always wanted to design my own engine, so have made a start using some spare time this summer holiday.

The engine type and size is based around a new RC plane I am also designing. The engine specs so far are:

Type: V-Twin gas

Capacity: 100cc (2 x 50cc)

Stroke: 3.9cm

Bore: 4.0cm

Compression Ratio: 8.5

Ignition: Spark

My starting point was to draw the basic geometry in Autocad to check the key dimensions. Hopefully the time spent on this will reduce the problems I have later.

[Craig Booth 1Participant @craigbooth1]

[EmgeeParticipant @emgee]

Did you consider less than 90 deg Vee to keep width less ?

Emgee

[Craig Booth 1Participant @craigbooth1]

next step was to draw everything in Autodesk Inventor. I am learner to this software but it is proving relatively easy to use and extremely useful in checking clearances, parts fit, and also checking the moving parts. Each part right down to the bolts are drawn as individual components and then "assembled" using physical contraints. Any subsequent changes to the components is automatically reflected in the assembled model. A big plus also is the it is extremely easy to output working drawings (and again these automatically update if any changes are made in the model).

Below is rendered version from the assembled model, plus a series of screen grabs.

[Craig Booth 1Participant @craigbooth1]

Posted by Emgee on 09/07/2019 22:19:50:

Did you consider less than 90 deg Vee to keep width less ?

Emgee

Hi Emgee, yes I did, but was struggling to get the cam shaft housing to work, plus the 90deg actually works well with the plane I am designing. The cylinder heads will intentionally project out of the cowling as I want them on show to suit the style of plane. (I will try to get a drawing of the plane uploaded soon)

[Tim StevensParticipant @timstevens64731]

Might the valve guides be on the short side? – resulting in valves not seating firmly, perhaps.
And is the plate through which the valves work a separate part as drawn (but not illustrated in colour) – (increasing the problem, perhaps)?

Regards, Tim

[HopperParticipant @hopper]

Big advantage of a 90 degree V-twin is perfect primary balance. Start closing the V up and they start vibrating. Vis: 90-deg Ducati motorbikes vs 45-deg Harley-Davidson.

Looks like a nice engine in the drawings there. Should look pretty cool in the plane with those exposed rocker arms sticking out in the breeze on full view.

Edited By Hopper on 10/07/2019 01:31:12

[Craig Booth 1Participant @craigbooth1]

Posted by Tim Stevens on 09/07/2019 23:09:27:

Might the valve guides be on the short side? – resulting in valves not seating firmly, perhaps.
And is the plate through which the valves work a separate part as drawn (but not illustrated in colour) – (increasing the problem, perhaps)?

Regards, Tim

Hi Tim, thanks for the comments. I've shown a closer cut away below. The valve guide in contact with the valve is 13mm in relation to a 4mm diameter valve. The sleeve (colphos) also passes through the rocker arm bracket (aluminium), but the valve does no come into contact with it.

Let me know what you think. In my head 13mm feels about right to keep the valve aligned, but I don't have any reference to go by so happy to listen to experience.

While we are on valves, I am going to use a split collet/collar to retain the spring. My last engine used a circlip which makes me nervous about failing.

Cheers Craig

[Craig Booth 1Participant @craigbooth1]

a few more details:

Main body and head: Aluminium HE30

Piston and conrod: Aluminium HE15

Piston Sleeve: Cast Iron

Piston Rings: Cast Iron

Crank: EN16

Crank support: roller bearings: nose/main/rear

Conrod: needle bearings big end/colphos bushing small end

Valves: Stainless

The camshaft will be mounted within the main crankshaft body on needle bearings, driven by a toothed belt and pulleys at the rear. This is the main area I am still working on in terms of how I achieve the connection to the crankshaft.

Cheers Craig

[Niels AbildgaardParticipant @nielsabildgaard33719]

I think You need to reserve more space in crankcase for crank,counterweigth and conrod big-ends.

The crank shown from a leaf-blower engine balances all rotating and half reciprocatory forces.

To achive primary perfect massbalance i a 90 degree V-twin You shall balance all rotary and one reciprocatory totaly.

This means that Your crank will be two times as voluminous relative to piston as shown on picture.

Edited By Niels Abildgaard on 10/07/2019 08:30:57

[EmgeeParticipant @emgee]

Niels

Isn't the greater weight of the counter balance shown in your pic a result of the heavy fan unit, on the model engine it will only be a much lighter prop driver.

Craig. You could extend the valve guides into the port if further guide length is needed.

Emgee

[Niels AbildgaardParticipant @nielsabildgaard33719]

Posted by Emgee on 10/07/2019 10:01:57:

Niels

Isn't the greater weight of the counter balance shown in your pic a result of the heavy fan unit, on the model engine it will only be a much lighter prop driver.

Emgee

Nope.

The fan unit/flywheel with ignition magnets etc balances very nicely on the rotation centerline.

[John MCParticipant @johnmc39344]

Posted by Craig Booth 1 on 09/07/2019 22:21:06:

You need to sort out the crankshaft bearing arrangement, I assume they are ball races.

John

[Dave HalfordParticipant @davehalford22513]

Posted by Emgee on 10/07/2019 10:01:57:

Niels

Isn't the greater weight of the counter balance shown in your pic a result of the heavy fan unit, on the model engine it will only be a much lighter prop driver.

Emgee

prop driver plus prop.

I still think the valve guide is not supported firmly enough in the head.

The spring plate, being separate will have to be a clearance fit would it not?

That gives approx. 3mm of guide supported by the head, I would have thought that there would be a high chance of a crack forming at the base given the rotary movement of the rocker 'working' the top of the stem sideways over time.

But then I could be wrong

[Craig Booth 1Participant @craigbooth1]

Posted by John MC on 10/07/2019 11:41:51:

Posted by Craig Booth 1 on 09/07/2019 22:21:06:

You need to sort out the crankshaft bearing arrangement, I assume they are ball races.

John

Hi John, can you please expand more on the crankshaft arrangement. Yes, they are deep groove ball bearings. Main bearing open, nose bearing sealed only one side, what is not shown yet is the rear bearing and second crank web.

cheers Craig

[Craig Booth 1Participant @craigbooth1]

Posted by Dave Halford on 10/07/2019 11:43:50:

Posted by Emgee on 10/07/2019 10:01:57:

Niels

Isn't the greater weight of the counter balance shown in your pic a result of the heavy fan unit, on the model engine it will only be a much lighter prop driver.

Emgee

prop driver plus prop.

I still think the valve guide is not supported firmly enough in the head.

The spring plate, being separate will have to be a clearance fit would it not?

That gives approx. 3mm of guide supported by the head, I would have thought that there would be a high chance of a crack forming at the base given the rotary movement of the rocker 'working' the top of the stem sideways over time.

But then I could be wrong

Hi Emgee, yes I see your point, will look to address this.

thanks

[Steve King 5Participant @steveking5]

Really looking forward to seeing this engine buid progress. Good luck and keep us all posted.

[Tim StevensParticipant @timstevens64731]

My concern about valve guides is only partly to do with strength – and others have commented on that aspect. The other part of the problem relates to heat transfer. The exposed exhaust valve stem takes in heat from the exhaust gas, and the part in contact with the guide loses heat into the cylinder head. In your design there is too much exposed and not enough in contact.

Another heat concern is the cooling of that area of head, which is exposed to a large area of exhaust port, and has no fins. You might think that modern engines don't seem to need fins here, but they are drenched in oil, which carries the heat away. Your system relies on no fins and next to no oil …

It may also be a problem with the whole exhaust port as a separate insert of bronze. This adds an extra joint where heat must be conducted, and this relies on real firm contact at all times. The port will heat up faster than the surrounding metal (although its coeff of expansion is going to be similar), so even if everything is tight on manufacture, the joint will soon cease to fit firmly everywhere. If you are reconsidering this area, I would suggest that the sides of the port are directly in the head aluminium, with a lid comprising the guide and a flange, in bronze. If the flange extended as a fin, that would be an advantage.

But then, what do I know?

Tim Stevens
[Silver medal, motorcycle engineering C&G]

Edited By Tim Stevens on 10/07/2019 12:32:49

[JasonBModerator @jasonb]

looks quite a similar setup to Malcom Stride's Bobcat/Jaguar designs and the Ocelot and Puma derivatives which seem to run OK with no fins whatsoever in the head and with inset valve cages. Though as you have a slightly larger engine may as well put some fins in as it woun't be too fiddly to do.

[John MCParticipant @johnmc39344]

Craig, the bearings need to be constrained correctly. Its a rotating shaft so the inner races need to be clamped to the shaft. As drawn, when the engine gets hot axial load from the expansion of the bearing housing (al/alloy?) will put extra power absorbing axial load on the bearings

One of the outer races needs to be fully constrained so an interference fit in it's housing and one side against an abutment in the housing and, in this application, a circlip (space/weight considerations) for the other side, that is to say the bearing cannot move axially and therefore provides the axial location for the crankshaft.

The other outer race needs to be able to move so a transition fit in it's housing but not against any abutment. This is so the bearing can move (a very small amount) as the engine gets hot. This will stop the bearings taking unnecessary axial loads.

The SKF catalogue has all the information on tolerances.

You could consider a fully constrained ball race for axial location at the drive (propellor) end of the crank, that would locate the shaft and a drawn cup needle roller (hard steel sleeve pressed on to shaft) at the other end, absolutely no risk of excess axial loading.

I see the incorrect mounting of rolling element bearings in the model engineering world all too often, has there never been an article in ME to show the right way?

John

[Jon LawesParticipant @jonlawes51698]

Posted by John MC on 10/07/2019 15:24:32:

I see the incorrect mounting of rolling element bearings in the model engineering world all too often, has there never been an article in ME to show the right way?

John

That's a great idea, I'm sure they would be very grateful.

[Dave HalfordParticipant @davehalford22513]

The much more substantial roller rockers in the Stride design would help a lot, though guides seem to be stronger than I thought they would be

[Craig Booth 1Participant @craigbooth1]

Cylinder head altered a little, increased height by 2mm which allows better sleeve support and also slightly larger fins.

[Craig Booth 1Participant @craigbooth1]

Posted by Niels Abildgaard on 10/07/2019 08:29:57:

I think You need to reserve more space in crankcase for crank,counterweigth and conrod big-ends.

The crank shown from a leaf-blower engine balances all rotating and half reciprocatory forces.

To achive primary perfect massbalance i a 90 degree V-twin You shall balance all rotary and one reciprocatory totaly.

This means that Your crank will be two times as voluminous relative to piston as shown on picture.

Hi Niels, yes balance will be via all rotary and 2 times half reciprocating mass.I'm pretty sure I have enough space to do this as there will also be the rear crank web. Of course it will not be based on volume as you note above but mass, with the steel of the crank web having a greater mass per volume than the lighter mass of the piston.

Another good think about Autodesk Inventor is that it calculates mass and centre of gravity of all parts based on material, so I will be able to use this rather than trial and error balancing.

Craig

[Craig Booth 1Participant @craigbooth1]

Some more work on the head to change the inlet/exhaust connection. I now have a plate connection which will allow me to easily make the manifolds that will be need to be bent to shape to suit the cowl. Manifold will be copper, silver soldered to flange. The lot can then be chromed.

[Author]

Posts