Engine Block Structure Design Considerations 3

[mloew]

Engine Experts,

What are the first order engine block structure design considerations? I came up with the following:

Cylinder internal pressure
Head joint loads (fasteners)
Thermal distortion of bores
Heat rejection requirements from cylinder bores
Powertrain mounting considerations

I would like to assume that there are no accessory loads on the block.

Are there any acoustic, manufacturing, lubrication, etc. load cases that I am missing?

I have been thinking about the optimal configuration (structural design and semi-materials considerations) of an engine block. In order to develop a parameterized analysis model, I would like to make sure I have all the important first order effects captured.

I welcome your input. Best regards,

Matthew Ian Loew

 

[GregLocock]

Block stiffness in torsion and bending is pretty crucial for radiated noise.

Impedance of mounting points for accessories.

manufacturability, obviously.

low weight

small package (ie get the FEAD compact) Cheers

Greg Locock

 

[mloew]

Thanks Greg!

I was considering not having an accessory drive. I think the engine structure would be with fewer compromises (not to mention easier to design) without bolt-on accessories. I was thinking about an auxiliary power unit to manage the accessories.

Low weight would be a product of the optimized design, it is not a load case. The suggestions of considering tensional and beaming stiffness and the reason for it is a good one.

What are the manufacturability considerations?

Also, what is state-of-the-art for power train mounting? Neutral Torque Axis was the buzzword a few years ago, if I remember correctly. Can the mounting details be considered in a first order analysis, other than the stiffness (impedance) of the mounts on the block? Best regards,

Matthew Ian Loew

 

[GregLocock]

Manufacturability - How to get the sand back out of the waterjackets is one big issue! Also, more generally, how to design it so it will cast properly.

Torque axis mounting is a very old concept, and I'm not sure that it really does much for you. Front wheel drives often use an approximation to it, but RWDs don't, since the front mount would be on the front X member, lousy for crash and package, and hard to get stiff enough.

For a RWD you have two alternatives - 3 mount or 4 mount. Every car I see uses three mount, now.

For this you put the front mounts more or less symmetrically across the c of g of the entire engine and transmission, and then use a third mount on the back of the gearbox.

4 mount has a big advantage if you can get it right - you could eliminate the gearbox cross member as a noise path. I'd be inclined to stick two mounts on the firewall by the rails, but further in, and two in the front somewhere. We need to keep the roll mode low, I don't know if that applies to everyone, so we want to minimise the lateral span of the mounts. SInce we are usually thinking about AWD as well it makes sense to maximise the longitudinal spacing.




Cheers

Greg Locock

 

[Azmio]

Mloew
Here are more considerations that may help you:-1) Output both ps/liter and torque/liter
2) Max engine revolution
3) Bellhousing bolt pattern
4) Target cost
5) Supplier
6) Structural consideration; head to block, crank to block and block to bellhousing (powertrain bending is the real challenge nowadays)
7) Fuel sulfur content
8) Warm up time
9) Recyclability
10 ) weight target
11) Bore surface selection
12) Dimension because this will determine bore to bore distance, interbore cooling, etc.
13) Displacement
14) Vehicle application
15) Balancer shaft attachment
16) Gasket and sealing strategy
17) Cooling strategy
18) Manufacturing
19) Lubrication

Once you're cleared with the above items, you can go on with the conceptual stage in determining whether you want to choose open deck, v engine vs i engine, bedplate, etc.

 

[1969grad]

Mloew.

Wow, what an undertaking. Don't forget that when you get all of the factors listed that you have to assign importance (weight) factors to each feature which is just as difficult. Good luck!

 

[Azmio]

1969grad,
Well, I am now designing the 4th production cylinder block. Of course, the cost, manufacturability, supplier choice and performance requirements weight above the rest. Once we take care of these three, the rest of the points above would be the guidance for me to finish up the design. To some the points above can be a major headache, but if we use them as our guide, the design task would be an easy one.

AO

 

[GraviMan]

One thing that never seems to be mentioned is service life and maintenance. OK this doesn't directly affect an FEA activity, but should be looked at in the design stage. I would imagine FORD do FMECA (and other paper intensive) studies on in service failures of engines - Q.E.D. use of chains for timing. Rover (pre BMW) was even considering a one-piece Head/Block to improve cooling and avoid gasket failure. Good for high compression and low mass too. A favourite in off highway apps is mounting point failures for ancilliaries - external anciliary motor might not be such a bad idea.

Might pay to look at vehicle dynamics and mounting point accelerations. I recall several incidents of engines suffering mechanical failure due to severe off-road conditions.

Is your app Diesel, Petrol or Turbine? What config (V8 etc)? Different designs have different in service problems - Ford (Europe) favourite seems to be oil leaks after 100'000 miles. Some K series jam VVT cam gear. Take your pick...

Mart

 

[Azmio]

Graviman,

If you look at the K series, the design is not robust at all. Imagine, the block uses long bolts from the head until the plate below the bedplate. The bolt is subjected to elongation and the head+block+bedplate+plate structure also experience some stress relaxation over time that may effect the overall height. The liner is pressed fit and not supported uniformly around the liner. It is a recipe for leakages at the head gasket.

As for the mounting points, the transverse engines that I have do not have any mounting bracket bolts attached to the block. The head and front cover are sufficient enough to take up the load.

As for the longitudinal engine application, I am worried more about exhaust heat going to the engine mount rubber.

Generally, for mounting points there will be engineers from the vehicle side who will determine the exact location of the engine mounts. They have better ideas because the mounting points depend a lot on the engine+transmission. My group handles mostly the engine side and not the whole powertrain or vehicle engineering.

 

[patprimmer]

Matthew

Oil filter location and orientation can be significant re service and drain back, and what blows through the bypass on start up.

Also cam position and cam follower bores if it is a push rod motor, and maybe timing chain tensioner if it is OHC.

Side loads from rod angularity.

Greg covered engine mounts. I would like to add that if the 4 mounts are added at the corners, they are supported by a bulkhead, whereas, when 3 are used, 2 are normally onto the flat, unsupported side of the water jacket.

Although you talk of ancillaries on a separate motor, the starter must be mounted near the ring gear, on either the block or bell housing, and I think it generates enough torque to warrant consideration.

Whether the block structure, water jackets and bores need to be the same material, and therefore can be one piece, or if optimal materials can be used in different areas, thereby creating more pieces and assembly processes.

The thermal expansion characteristics of the block vs other components like head and pistons.

Will the block end at the centreline of the crank or go well below the crank centreline. This adds weight, but increases stiffness around the main bearing caps and helps reduce nouse.

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[mloew]

Thank you all for your comments in this resurrected thread. This was never actually a formal project, but one of the exercises I use to illustrate my desire for performing top-down design of major systems. I am actually doing this at a full vehicle level (chassis, structure, body, hybrid powertrain, material handling systems, etc.) and always start with the First-Order Engineering Fundamentals. Member Greg Locock initiated a similar topic on FOE considerations on suspensions systems in the thread Important suspension parameters (thread800-66001).

On the engine side, I still feel that if the FOE principles are considered from a "blank computer screen" that significant improvements in the packaging and mass efficiency of ICE can be realized. Some of my thoughts are starting be to proven out in newly developed engines, but I still do not have a full grasp of all the design drivers. I am not an engine expert. Just a dedicated proponent of Top-Down Design and First Order Engineering.


Best regards,

Matthew Ian Loew
"Luck is the residue of design."
Branch Rickey

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

I think (hope) we'll see more accessories electrically driven in the future, which should simplify the FEAD design.

If push comes to shove I'd like to see electric waterpumps and oilpumps even - it is astonishing how much power can be wasted in these things due to poor design.

pat's right to bring up the 'depth' of the structure around the crank. The Lexus V8 is immensely stiff in that area partly due to the use of cross bolted mains, and a bearing bridge of some sort (can't remember - it's been a long time)

There was a fad a few years back for casting the engine bellhousing and trans case in one piece, since the design of the bolted connections always compromises the bending stiffness in particular. I can only think of one manufacturer who got it into production, everyone else probably got cold feet.

Cheers

Greg Locock

 

[GraviMan]

"This was never actually a formal project, but one of the exercises I use to illustrate my desire for performing top-down design of major systems."

Cool. I'm doing just the same with a 60 tonne truck - great fun!

"If you look at the K series, the design is not robust at all. Imagine, the block uses long bolts from the head until the plate below the bedplate. The bolt is subjected to elongation and the head+block+bedplate+plate structure also experience some stress relaxation over time that may effect the overall height. The liner is pressed fit and not supported uniformly around the liner. It is a recipe for leakages at the head gasket."

Agreed - it was designed more as a light/cheap motorsport engine than a high mileage unit. It was good in it's day, but is not a new design. I still rate the through bolt idea highly, but accept that the liner design needs attention - don't forget the engine was originally designed as a 1.4, not as 1.6 and 1.8. Pity there was never really a follow up...

Mart

 

[Kiwikid]

As an aside here's a little engine designed for the Ultralight aviation market. It was designed and built by some auto engineers in Christchurch NZ in a small workshop.

Areotwin in the US have taken up the International distribution rights for this engine.

It's a 2 cyl 4 stroke EFI with turbo and has some very interesting features.

http://www.aerotwinmotors.com/pages/aboutus.htm

Cheers
Kerry

 

[bazeleur]

Very interesting features indeed,

I like how they've integrated the cylinder heads and barrels. I see plenty of advantages in doing so, not least of which is dispensing with the troublesome head gasket, and replacing it with a regular gasket or sealant between the lower part of the cylinder barrels and the crankcase casting. Then of course there's the reduction in structural requirements and thus weight, because there aren't any head bolts/studs to worry about anymore. What I want to know though is, why don't we see this sort of solution implimented more often? Added casting complication perhaps?

 

[Azmio]

Bazeleur

Hmm.. I am from the production engine side, with the crankcase integrated to the head, there will be plenty of manufacturing complication:-
1) To assemble the valves to the head
2) To assemble the piston and upper conrod to the crankcase. I guess you need a separate upper and lower crankcase much like the bike engines, again you will have sealing problem to worry about.
3) It will be quite difficult to hone the cylinder bore.
4) It will be difficult to apply Nikasil or YRPS or Plasma spray to the bore if you are planning to have siamese bore
5) To enable the piston assembly to be assemble to the crankcase from the bottom, you will need to lengthen the whole engine
6) Hmm.. the casting will also be difficult because the sand cores require support from one and another
7) I wonder how we can properly control the coolant distribution from the block to the head. The gasket coolant aperture works well during warm up and during WOT operation. Unless we are happy with big apertures made from the sand cores, we may need to drill, but how?

Anyway, a solution to a problem may cause 10 more problems to occur. I am sure there will be many out there capable of providing solutions to the new problems.

 

[harrisj]

Another design issue is C of G. You might be steered towards a boxer or wide-angle V if low C of G is crucial - and then dry-sump it.

Then you would also have to consider maintainability - unless it's a pure competition job. Ease of maintenance is a big design driver for msot engines.

If the engine is for a hybrid, then there's no need to worry about starter, generator etc and all ancillaries (as Greg points out) could, and should, be electrically driven.

In a way, the biggest design driver is the application.

 

[bazeleur]

What about average wall thicknesses for aluminum block castings? I've seen/measured anywhere from 4mm to 6mm, with another 50% around bosses and stress concentrations...

 

[GraviMan]

It has to be said that this is where the K-series through bolt idea came into it's own. The block itself was only in compression, from strengthened head and bearing ladder. The liners (1.4 wet liners being far less of a compromise than 1.6 and 1.8 damp liners). Only powertrain modes and structural fatigue were considerations for block thickness. Can't remember thicknesses off hand, but it was light...

Mart

 

[GregLocock]

Yes. You should have seen the initial concept studies. Very whacky.

One idea was a vertical split in the block, so that the liners with pistons and conrods and crank already assembled could be dropped in by robot, then drop the other half on, tighten up a few screws, and hey presto, the engine is built. None of those got made.


Cheers

Greg Locock