Performance Cranks Vs Std. Ones

[Mandic 0]

Hi,

 

I'd like to put theory and practice together a bit in this post.

There was a debate quite some time ago whether reducing crank weight or even fit 8v crank to Mi16 engine to minimize rotational mass has any downsides.

Now lets skip that part about total weight and focus on design differences between XU9 8v and 16V cranks.

 

The biggest difference (bar mass and material) lies in counterweights. Mi16 has 8, 2 per piston/rod assembly, where 8v only has one

 

So, because engine has 4 pistons they can not be positioned in the same location. Therefore they produce torsional/rocking couple as will any two forces acting on an object if they do not act on it in the same location Torsional couple explained here.

 

 

A single counterweight web can cancel out some of the rocking couple between piston pairs but leaves a smaller rocking couple within the confines of each bore, where a double counterweight web removes the rocking couple within the confines of that bore but can still never exactly cancel all the rocking couple between opposing piston pairs.

 

As we see single counterweighted cranks can not cancel out those forces so they will tend to bend/flex the crank.

 

Going from this Mi16 crank is more suited for high rpm as it will flex less than 8v crank.

 

 

So far so good. But what if we take crank which is made out of steel rather than being cast and with 4 counterweights? Material wise everything shoud be ok as steel is able to cope with forces involved easily. But what about flexing?

 

Surely it will flex more than Mi16 crank, but are these flexing oscillation big enought to cause any engine damage, will uneaven gap cause premature

bearing wear, etc.

 

So why did Peugeot use 4 webbed steel crank in T16 engine? Why are BTCC steel cranks also 4 webbed? Where's the catch?

 

Theory states one thing and yes, all of the above is true, fact of the matter is that 8 conterweighted crank WILL flex less (ok, material has influence as well), but how big are those oscillation anyway? Can one put 4 webbed crank in and be sure that by doing so will not shorten engine lifes-pan?

 

 

Thanks!

 

Regards

 

Ziga

[James_R 3]

stroke's also got to be a factor, but there's people using 4 web cranks to 8krpm all day.Bit like "harmonic balancers"8v's don't even use them and they're only 4 web, but Mi's s16, 6's etc.. all do and they're 8web.Think unless you're trying to spin it very fast it doesn't matter much.

[DrSarty 90 1 Cars]

unless you're trying to spin it very fast it doesn't matter much.

 

So how fast is fast? I've read about many people (now) taking TU's to 10k RPM, but surely engine capacity/no. of cylinders has a natural effect on practical rev range.

 

Sure, there's a trend with modern, hi-tech V8s to rev pretty high, such as the M3 & RS4 lumps doing over 8k. BUT, are these - albeit impressive - high RPMs really necessary?

 

What I'm saying is, if 8k is achievable from this lighter, 4 web crank (like the BTCC & T16 cars use), then surely that's enough, and anyone doing a full engine, performance rebuild could...should go this route?

[James_R 3]

V8 configuration are naturally more harmonised.

 

There's probably some benifit to "normal" engines to be full counter weighted that race engine's don't have issues with, like being revv'd below 3-4k that a road engine will live at.

 

when I say high I'm taking 8k area,s have to be running pretty big cams to see peak power there.

[brianthemagical 1]

Usually if an engine revs to a certain rpm, especially if it has been thought out at all, then it will still produce useable power. on the t16 cranks, more webs is more weigh, so if there not needed then they will be omitted, as the main reason for inclusion was reducing flex, then maybe the steel was strong enough to eliminate excessive flex. one of the main factors deciding max engine speed is piston speed, and the engine is then designed around it.

[Owain1602 0]

The steel used is much less brittle than the cast due to the finer crystaline grain structure so it has less chance of cracking or snapping under the forces caused by the coupling motion.

 

Crankshaft mounting, bearing material, engine weight and mounting are also factors in determining crankshaft design and materials.

 

You can workout the coupling forces acting on the crank and the plane in which they will be acting by formulae, I can give you some formulas if you would like but it requires you to know quite a lot of variables before you start. You can then use this expression to compute a graph or table of pistons acceleration against crank angle. Or you could work out piston velocity against crank angle then differentiate the expression to obtain acceleration over time.

 

I have just started to design a crank for a 2stroke engine that will rev to maybe 40,000rpm if everything is ok with induction and inlet timing. I draw this on a CAD package and can analyze the stresses and crank deformation througout the structure, the constraints I use will determine the results of which forces I want to analyze.

 

Your basic theory is reasonably sound but as with everything there are variables that dont allow basic formulas to be exactly applied to real world situations.

 

This is a really deep subject if you want to get in to it and understand, this knowledge can also be applied to any other problems, not just automotive.

 

Let me know if you want to know some more.

[carbers205 1]

This topic has suddenly become of great interest to me!

Im building a bit of a hybrid engine and am planning on putting an XUD (88mm) crank in my XU10 bottom end. Ive just noticed today the purpose of this thread in terms of web numbers as both XU10 86mm cranks have 8 webs and the XUD only has 4.

Now firstly what is the XUD crank actually made of? Im hoping its steel.

So am I right in now concluding that in this scenario with the engine revving to no more than 6500rpm that having only 4 webs may in fact be of benefit due to lower rotational mass and that any flex will not be sufficient to induce premature wear?

And its it possible to lighten and balance the crank any further?

Is knife-edging of a crank carried out on the webs? It would seem the logical place in which to do it. If so Im guessing this is the method of lightening in this case.

 

Andy

[Sandy 191]

I go on reliability. The least reliable crank in the range from what I've seen, has been the GTi6 one (8 weight Iron). We've used the 4 weight steel diesel cranks to over 8000 rpm without any problems and the 4 weight Iron TU cranks beyond that.

 

I like your post Owain, illustrating an ever important point, that the science/maths are of limited validity on some issues, because of the variables involved. That's where experimenting and observing can often provide better answers.

[jimmer 0]

So an XUD crank in my XU10J4RS rally engine would be a good idea? What cars have an XUD crank, and is a straight swop?

[Owain1602 0]

Yes Sandy, but I wans't stating that science/maths are limited in some areas but what I was saying is that the level of knowledge needed to account for every little detail when analyzing/designing a component is beyond 99.9% of people.

Normal theroy can get a ball-park figure but to get the actual figure it takes so much more, its like the old saying 90% of the work takes 10% of the effort.

 

As fior your oppinion on experimenting is best sometime, yes I totally agree, if the funds dont permit sufficient research and development.

 

In the end almost every single thing can be calculated and analyzed with maths, I dont like the oppinion that numbers dont come in to things, this is just because the person doesnt understand some basic maths in most cases.

 

As for XUD cranks, I have no idea, never seen one, not to mention analyze one. What you must remember when using different cranks is that a crank has been designed around everything else such as piston weight, rod weight/length, block weight, bearing materials. If anything changes then its not quite right.

For example if you use the XUD crank, the counter-weights have been accurated designed to counter-balance the weight of diesel pistons and rods, no if you used petrol ones, its not exactly balanced. Anything out of balance can cause nasty harmonics and if these frequencies occur at the natural resonant frequency of the crank then a lot of undue stress can be created on the crank.

 

You see, everything can be analyzed deeply but what is your requirement of the crank is the question.

 

As for the post about lightening a crank, I really would not do it, unless it can be dynamically balanced with the rods and pistons. But if you remove material from the weight that counteracts the piston and rod, it only goes to say that your rod and piston would need to be proportinaly lighter. Your coupling forces will certainly be out of balance: piston 1 up, piston 2 down will create a clockwise couple on the crank, what counteracts this is the counter-weights are in the opposite cycle so they would create an anticlockwise couple, this is what balances the couple out. So if you removed weight from the counter-weights, its not right.

[DrSarty 90 1 Cars]

Owain, your posts and knowledge are fascinating. However, and whilst I agree you can calculate and pontificate all day long, cliches become cliches for a reason, such as 'tried and tested' and the 'proof's in the pudding'.

 

I feel the calculations will only ever give you a starting point. The better the calcs - the nearer the goal you will be, but you still need to convert the goal.

 

We should find the specs of two of the Pug cranks discussed here for you, 4 and 8 counterweight, together with their corresponding rod and piston masses just to see if {a} Peugeot DID calculate it correctly OR {b} (which I think is what Sandy's alluding too) whether there is room to play with with these components via machining or swapping to take advantage of some of the inherant unpredictabilities of life.

 

Edited February 24, 2008 by DrSarty

[carbers205 1]

Ah, I was hoping you might have a look in on this one Rich!

Are you sure you would want the results of these calcs as having re-read your build thread I noticed you're down to 4 counterweight too!

Anyway it is quite obvious that QEP know their stuff so if going from 8 web to 4 were to cause a problem they would have flagged it up at the start.

It seems now that providing the components are balanced and matched properly to each other that problems should be minimal, though lightening of the crank is probably best left alone.

 

Andy

[Sandy 191]

I'm perfectly able to do the maths, I've always been in the top few people throughout for maths. My point is, there comes a point when modeling with numbers isn't practically possible because there's so much going on, chaos essentially. Ultimately, theoretically calculable of course, but realistically sometimes not all the variables can be simultaneously accounted for in a meaningful way. This point were the established theory diverges from the reality is always progressing, but it still exists in engines, which is why no two engines are the same and some people build better race engines than others etc. Knowing when to trust the calculator and when to trust experience is the critical thing IMO.

[DrSarty 90 1 Cars]

Awesome Sandy!

 

I've heard that diesel cranks like mine are much stronger, down to material more than anything. And don't forget, TD engines produce masses of torque - likely even more than my 2.2 will - so should stay the course. Here's hoping eh!

[carbers205 1]

Heres hoping indeed Rich! Lol.

Am I right in thinking your crank was just matched and balanced but NOT lightened?

I think half the problem with advice is people do things so differently, be it theoretical v. practical, or different methods of practical application and so on.

All I guess Im looking for is to whether people have run these setups without issues, whilst taking into account the theoretical advice of making sure all the components will work smoothly together!

Does all that make sense?!

[Owain1602 0]

Fully agree Sandy, thats why Mclaren/Honda/Cosworth whoever dont just design and build a crank and just throw it in an F1 engine, they know it should work, but will it work? Infact they will spend more time testing the part that all the other process put together.

I wasnt inferring that you weren't able to do the maths Sandy, but it seems to be that word of mouth gets passed on so much, you end up with people arguing about things that they have absolutely no idea about, just because Joe Bloggs told them it was good.

 

In the case of Peugeot 205 engines, regardless of spec, this type of analysis is simply not justified by the cost of things. Even if you have a £8000 XU engine, it would cost multiples of that sum to analysis an item such as a crankshaft.

 

I have just started to design a small 2 stroke engine, totally from scratch, every has to be designed and fully analyzed, then made by myself. This will take me some time bacause its something I can only do when I have some spare time as I have other commitments at present. An added complication I have when it comes to designing the crank is that a section of my crank will need to be hollow inside to allow for air induction as I have gone down that type of induction route.

 

In these instance I dont believe that using 4 weighted cranks instead of the 8 will matter too much, we are not talking about huge engine speeds plus the block is quite a sturdy item. Advice from people on here such as Sandy can also help people becuase they know what cranks to avoid and which ones rairly fail.

 

Oh yes, I was going to build a 1.6 16V XU a while back but never got round to it, I stripped a 1.6 8V down ready to measure the bits and pieces to see what components I would need and what sort of compression I would get with various bits but again time didnt allow me. What do you guys recon of this idea? Im sure some other people have done it in the past.

 

Thanks,Owain

[Sandy 191]

It's a topic that has come up recently a few times, in fact someone on here has built a 1.6 Mi16 IIRC. I'm building two 1.8 16v XUs at the moment, one race spec (but OE crank) and one medium budget hillclimb spec. They might give some pointers on what's possible. I'll been working with cam and exhaust geometry from John Read's XE engines (he's currently exceeding 145bhp and 95lbft per litre on his best engines!) and head work to match from Mark Shillaber. The problem is that the head, bore/stroke and rod/stroke ratio are all different to the XE engines, so it only provides a starting point. Induction side tuning and cam timing will have to refine the package and experimentation really is the only way to evaluate that!

[CaptainK 11]

The least reliable crank in the range from what I've seen, has been the GTi6 one (8 weight Iron).

Now you tell me.

[B1ack_Mi16 67]

About crankshafts, I've made a little simulation of a crankshaft I modelled in Catia with more or less correct XU engine measurements.

Have just attached some rods yet but wanna try to put on pistons later on to see how much flex there is in the crank at high revs.

 

This is just a small movie of it spinning at 9000rpm with the rods attached.

 

http://psaforum.dk/tmp/von_Mises_stress_and_deformation.avi

[Owain1602 0]

What solver did you use?

I genreally use UGS NX 5 for modelling and Nastran as an FEA solver.

[B1ack_Mi16 67]

I also use NX 5.0 and Nastran a lot As we speak really.

 

The simulation is done in Fedem (www.fedem.no) which imports nastran files.

 

Joints etc are modelled and connected using either RB2 or RB3 spiders to the surfaces that shall be connected to the different joints.

[DrSarty 90 1 Cars]

 

I'm currently favouring the '2008 Ask PeteT/Mattsav/Sandy/Miles.4.1' package myself.

Edited February 25, 2008 by DrSarty

[Owain1602 0]

Can anyone give me the following data, I will work out piston acceleration values and if I get some time ill start some basic FEA on a standard XU crank, ill do a 1600 crank.

 

Crank stroke and connecting rod centre to centre length is what Im after please?

 

I cant start on modelling the crank for a while as the cranks I have are at home and that is currently a long way away but I can start t put some calculations together to get some data out.

 

Thanks a lot,Owain

[Mandic 0]

Dont know for 1600 but for std. Mi16 values are cca: maximal force on piston is in range of 1500kg, maximal acceleration is about 3200m/s^2 and max piston velocity is 34.5m/s, average torque transmited from cylinder (if pressure is constant, havent done calculations if it is not) is 0.626 and maximum is 0.988. For S16 which has longer rod is 0.628 and 0.99, but again, at constatn pressure, we all know it far from constant.

 

Cheers

 

Ziga

 

P.S. will answer on PM, just am short on time atm

Edited February 27, 2008 by Mandic

[Owain1602 0]

I only need crank throw and conrod centre to centre length thanks

 

I have neverr done any measuring of the XU bottom end components or i would have the data.