Let's take a simple case, a 4 cylinder petrol engine. I have seen tuners balancing the crankshaft at 500rpm with dummy weights to simulate rods, pistons, rings, etc (even with 50% factors or whatever). Is that really needed?
I mean, in an engine there will be friction, heat, combustion, explosions, etc. All this not at the same time of course, and not on the same part of the crankshaft at a time. So why balancing a crankshaft to the tenth of a gram with dummy weights, knowing that in the real world the engine will rotate much faster, with different forces acting on the crankshaft rod journals?
The real world situation will only be worse, so this "low level" balancing is actually quite helpful. Primarily, if I remember correctly - be warned, it's been a while since I've had to calculate an engine - the rotation speed goes in ^2 to the equasion, which exagerates the forces generated by the crankshaft imbalance. So, to limit this affect without needing extremely expensive test equipment (balancer @ 7500rpm would be obnoxiously expensive, if they even make them), they make the compromise of 500rpm. Anyone with experience should correct me.....
“Strange women lying in ponds distributing swords is no basis for a system of government. Supreme executive power derives from a mandate from the masses, not from some farcical aquatic ceremony!” Monty Python and the Holy Grail
An inline 180` 2-stroke with rolling element built up crankshaft?
A V4 360`4-stroke with plain bearing one piece crankshaft?
Flat 4, or inline 90`cross-plane?
2-stroke radial?
& on & on.. .. a more specific data set required M-M.. ..too many variables..
& "explosions"? They are def' to be avoided.. ..since the engine will stop.. ..expensively..
"Well, we knocked the bastard off!"
Ed Hilary on being 1st to top Mt Everest,
(& 1st to do a surface traverse across Antarctica,
in good Kiwi style - riding a Massey Ferguson farm
tractor - with a few extemporised mod's to hack the task).
Well since balancing a rotating system is pretty much only a function of the mass and geometry, if its balanced at 500RPM, then its also balanced at every other speed.
4 inline cylinders, petrol, 4 strokes. By explosions I meant the moment when the spark ignites the mixture (this is how its called in my language). I didn't mean detonation.
But it is the same with other engines, I mean the forces acting on the crankshaft rod journals and therefore the main journals will be constantly changing as the engine is running so why roughly balancing the crankshaft alone is not enough?
The rate at which the engine rpm is changing in an engine driven at anger is quite high.
Tim for sure, it only means if you have a 0.1g error at 500rpm, it will be much bigger at 9500rpm for instance, but my question was more about the meaning of fine balancing an engine bottom end considering you will never have constant loads on the journals (unlike what they do during the balancing process).
Timed burn.. ..expansion.. ..good..
Explosion.. ..destructive.. ..bad.. esp' for SI mills..
Again, there are so many variables, B X S ratio, design piston speed,valve train harmonics, crank counter-weights..
..number of main bearings/crank-whip & etc..
"Well, we knocked the bastard off!"
Ed Hilary on being 1st to top Mt Everest,
(& 1st to do a surface traverse across Antarctica,
in good Kiwi style - riding a Massey Ferguson farm
tractor - with a few extemporised mod's to hack the task).
I'm not an engine man, I'm just applying some engineering principles so that's my caveat.
I don't believe that the goal of balancing is to deal with loads coming from combustion. Like you said they are always changing. The goal is (likely) just to take out mechanical imbalances which do nothing but reduce component life.
In fact I'd guess that the balancing process doesn't even try to deal with valve train harmonics, since this is something that needs to be sorted at the design stage. I believe its purely for minimising the radial loads on the bearings that come from geometric out of balances.
in principle the peak inertia forces at primary frequency are zero (and will be with equalised piston masses etc as tuners also do)
but out-of balance forces from the rotating crankshaft are non-zero
the raw crankshaft will be out of balance, this imbalance will likely be worse with a raw counterweighted crankshaft
ie counterweights may need checking and machining
traditionally a 4 inline would not have a counterweighted crankshaft
the tuner might anyway have some ideas of non-standard counterweighting eg to help the crank or a main bearing
I am in a similar position to Tim here - a long time since I worked on this directly so language may not be precise, most people on the thread are right - in different ways.
Essentially, think of it as "types" of balance - static and dynamic. Static balancing - at this case at 500rpm rather than 10,000 RPM, is looking mostly at first order effects - are the masses in line, are they exactly equal, are they all acting along the same paths?
in a frictionless system you would do this at 1rpm, but we don't live in a world like that, so slow speed like 500rpm is used.
This is where the balancing of pistons etc happens - but you will be balancing to a closer more than 0.1g - perhaps 0.01 or 0.005 - remember that the G forces on the piston at 10,000 rpm are enormous, and all those fractions very rapidly grow with exponential speed - can be 10's of kilograms when all is said and done - especially in a racing engine which are designed for rapid transients in engine speed.
Also - remember that this is an racing engine, not a road engine - as well as the tolerances and physics you have the car designer moaning about things like centre of gravity and wanting less weight all the time - and an owner of some sort who wants to go as fast as possible for very little money and would like everything done yesterday. So if you're dealing with a fractionally light piston for example, you might mill out the OTHER pistons to balance. If weight needs to be added, it's going to be added as low as possible and in what ever way is best for improved acceleration. So - you may end up with a system which isn't perfectly symmetrical in it's component parts, but is close to perfect when viewed as a whole.
When all that is settled and the engine is on the dyno, you then test balances at higher speeds - and you don't just open the throttle - you go up in increments of perhaps 100-200rpm.
If you hit any high speed issues they will often turn up in regions - maybe you'll be clear up to 5,200 RPM, then have 600RPM of lumpiness, and back down again. These are much more annoying - it's harmonic issues, often 2nd or 3rd order, and it could be from all sorts of things - resonance in a journal, pulses in the oil gallery, one crank being fractionally hotter than another due to cooling effects etc - so it's a "balance" issue, but not the same type as you were dealing with at the static phase. Because it's much harder to measure the innards of an engine at high speed, it's harder to diagnose where the issue is.
When this type is problem happens, everyone says a few swear words, puts the kettle on, and then it's a process of elimination and lots and lots and lots of retesting. Again - this is a racing engine. If you had a problem like this on a road engine, you might consider just thickening up a wall somewhere or adding mass in some other way to damp it. However, you have the designer demanding miracles with weight, so thats not an option.
If your very unlucky, when you find the problem, it's something to do with one of the moving parts that you balanced earlier. So then you monkey with those, and your static balance goes, and you start everything over again - cue more swearing and more tea drinking.
If your excpetionally unlucky, the part that needs work is either a heat treated titanium component in some god awful phase, or something with a special coating on it, so any work you do is going to be an utter pig of a job, and will take 5x as long as anyone expects, and is very expensive. In these cases, along with swearing and tea, you stop for biscuits as well and try and think of ANY other way of dealing with the problem that will 1) work, 2) the designer will accept and 3) the owner will pay for.
Real world engine balancing experience is that if you don't do a static 2 plane balance on the assembled crank and flywheel then the first order vibration will destroy your engine mounts and upset your passengers. If it were severely out of balance I'd expect internal catastrophes at high speed - balancing forces are strong.
You don't need to simulate the reciprocating parts for static balance, so long as they are balanced in their own right.
The dynamic balance of the engine is, more usefully, a way of reducing the loads on the individual main bearings. That's when you need to simulate, or have, the reciprocating parts. That is adjusted by drilling webs out or adding slugs of dense metal to them.
You can balance a crank perfectly, statically, 2 plane, with no counterweights at all. They are purely there for higher rpm.
I developed the initial balancing strategy for the LT5 corvette engine, and was closely involved in production balancing of engines for several years, and designing a new crankshaft.
To give you some idea, an I6 engine will vibrate at around 5g rms at 5000 rpm. A 4 cylinder in line without balance shafts will vibrate at 15g rms at the same speed, if it doens't have balance shafts. The inertial effects of the poor dynamic balancing of the I4 are therefore responsible for the vast majority of the vibration. I don't remember the first order component of an engine at high speed, if you know the balance mass, then mrw^2 gives you the out of balance force. Balancing the flywheel is crucial.
But that still doesn't explain to me how balancing a crank with dummy weights will replicate what you see IRL with combustion and constantly changing loads ont he crankshaft. I am not an engine expert but I would like to understand the meaning and reasoning behind that.
Is it because we assume that as the engine rotates so fast with so many explosions per seconds the loads are equally spread on an average basis on the crank?
On a 4 cylinder engine, 4 strokes, we have 1 explosion every rotation of the crank (nothing new there heh), and each cylinder will produce power every 4 engine revolution, which means for e.g. 4800rpm (which equals to 80 rotation per second), each cylinder will produce power 20 times each second.
Balancing a wheel make sense because the load is always applied at the same point, but on the crankshaft the loads are applied on a different conrod journal.
combustion in effect causes an upward gas load on the engine (trying to lift the head)
and a equal-and-opposite downward load on the crankpin
so the resultant force on the engine bulk is zero (each cylinder 40 times per second in your case)
the reciprocatory motion of eg the piston causes inertial forces that may or may not be cancelled by motion of other pistons
in your inline 4 the major part of these forces (the part at engine rpm) is so cancelled
in this situation it may be worth balancing the crankshaft as you have identified
it is not precision made throughout and will be rather out of balance, so basic balancing (also flywheel) will be done on manufacture
When the crankshaft turns, the conrods are "converting" the rotation into a reciprocating movement. A part of the conrod mvoes up and down, and the other part rotates with the crankshaft. Now you have an additional mass attached to the crankshaft, which has to be balanced.
Due to the conversion of the rotating crankshaft to the reciprocating movement you get additional mass forces, which are "revolving" with one,two,... and more times the rpm (the higher order mass forces are usually neglected, so you usually only consider first and second order).
So therefore you can calculate how much of those masses you want to consider for your first order balancing and attach those dummy weights to your crankshaft and balance it.
No, your first order, static balance, is not affected by the reciprocating parts unless they are of different masses, or the crank is not symmetrical. So on any multicylinder engine I've seen you don't need the dummy weights, except maybe on 3 throw cranks - too late at night to think about them.
