To Balance or Not To Balance Crank?
 

I have my 993 engine totally apart. I am replacing the crankshaft with a new (used) one from a lower mileage car (original was damaged). The crank was inspected, polished and magnafluxed at Ollies. I have balanced the rods and pistons already. The pulley is new. I'll be re-using flywheel, pressure plate and clutch disc.

Is it worth the trouble, effort, money, etc to get the crank, pulley, flywheel & PP balanced?

My goal is a smooth running engine that will last another 100K or more. I do not race it. It is a daily driver. I just don't want to regret NOT doing something that I could have done pretty easily given it's all apart. Also, does the 993 crank follow the standard rule of 100% rotating weight plus 50% reciprocating weight?

Opinions?

An opposed engine is in pretty good balance to start with.
As long as you don't mix types of barrels and cylinders (size, weight) you should be OK for the normal RPMs.
We tried using mis-matched barrel/piston combinations on a VW once (to cheat the rules in class racing...bad us!)...and we put 66mm barrels on the rear (pulley end) of the case...and 92mm barrels on the flywheel end.
As long as they were opposed the engine ran very well.
And yes...we got caught...LOL.
But it did prove that if you match opposing P/Cs you can get away with a lot....just make sure that the heavy ones are on the same end of the crank...etc.
Bob

The pistons, cylinders and rods are an original matched set. The pistons, however, were 3 grams apart from heaviest to lightest (without rings or pins) and my machinist said the rods were also out of balance (now in balance).

So...pair up the heavy pistons and use them on opposing rods....and so on.
Bob

Yessir, it sure is. :)

Steve,

Is it customary to component balance the flywheel, crank, and pulley, then final balance as an assembly? Or do you just do a single balance job on the complete assembly? If necessary, where is material removed from on the crank?

Crankshafts are balanced, factoring rod & piston weights. Flywheels, pressure plates, and pulleys get zero-balanced so they can be replaced without having to disassemble the complete engine and re-do everything.

Material is ground off the counterweights in various placed to achieve objectives.

Steve,

Does the 993 crank follow the standard rule of 100% rotating weight plus 50% reciprocating weight?

I don' think that the 50% rule is applicable to a Flat 6.

I believe that a Flat 6 Crank has good primary and secondary balance and can be balanced as a simple rigid rotor to a maximum imbalance of 10 g.cm (0.139 in.oz) without bolting on additional weights.

I would balance a flywheel separately to ISO G2.5 and the same with the clutch cover.

I would match weigh the pistons and rods (I would also match little end weights)

The crank pulley is such a small diameter it may not have much influence.

I am not sure what 'zero balanced' means - is this a Class 0 according to the German Standard VDI 2060?

A crank itself should be balanced by design. I do not see how it could become imablanced unless material is removed from it, which is exactly what is done to balance it with pistons, rods, etc...But then, that particular crank becomes imbalanced by itself, and will not function correctly with another set of rod and pistons, right? It makes a lot more sense to me to balance the whole assembly via the piston pins weights, and removing material from rods whenever possible. Please enlighten me...

You are correct as it pertains to flat or opposed engines.
With an inline or V8/V6 it is a different matter.
As long as you are not pulling high RPMs IMHO you can forgo the balance.
If you plan on racing in any fashion..by all means spend the cash and balance it.
Be advised...there is balancing and there is BALANCING!!
The amount of offset or allowance is where the big bucks come into it.
To balance to zero grams...you are in the stratosphere (and remember...the clutch assembly is part of the rotating mass...as well as the pulley, etc).
The engine will be glass smooth...and probably last a lot longer...so it just depends on your wallet...LOL
Bob

I'm pretty sure Zero balance, means component balance. In other words, each component is dynamically balanced individually. This way, if one needs to replace the fly wheel, it can be done so "without affecting" the engine balance.

The truth is that you balance to a specification, such as ISO1940 G2.5. or whatever you prefer, there is always an inherent in-balance. The in-balance is a small amount of course, and it is within spec, but it is still there.

If you only component balance each peace and you don't do an assembly balance then you have to make sure you know the phase angle (location) of the in-balance so that when assembled you don't align the in-balance of each component into the same phase and potentially exceed the spec you are trying to achieve.

Otherwise, you component balance each piece, then you assemble and balance the assembly. This is the way large turbomachinery is balanced, except that in large machines, you may also consider an at-speed balance if the rotor operates above the first critical (natural frequency). In a car engine, the run speed is always below the first critical.

This is also why on any rotating equipment, you want to match-mark one piece in relation to the other...so that the balance of the assembly is not disturbed.

Always shoot for "0" Its cheap, it right, there are no wrongs .1 gram is a LOT of weight at RPM

I would consider balancing the reciprocating parts a basic part of rebuilding an engine. How do you know the conn rods are end balanced.... other than assuming they are or measuring them?

Have engine balancing costs increased significantly (~$250-$300 to elect not to perform? The more you know about the engine, the less you have to assume.

Sherwood

Crankshaft should be balanced for factory rod & piston weight. The biggest question is the pressure plate balance but then again no one check balances these except oddly enough the 356 guys whose engines turn 2000 rpm slower :). Considering the cost of a 993 build checking the balance seems like a good investment IMHO.

The reciprocating parts are already balanced. Supertec balanced the rods and I did the pistons. It's the rotating parts that need balancing. The only thing keeping me from balancing the crank is knowing which formula to use. My local machinist has a list of various makes/models provided by Hines but it does not include Porsche. According to Hines website if it is not on their list then the standard formula of 100% rotating and 50% reciprocating. All I need to do is confirm this and I can go ahead.

I have searched all over the interwebs and can't find this info. The machinist is not a Porsche specialist but otherwise is an excellent shop and well regarded. I am also amazed that there are not many shops that do crank balancing (at least locally here in the Bay Area).

I had an interesting conversation with Ted at German Precision. He said people are just not rebuilding as much as they used do so demand for this service has declined. This, coupled with California's stringent chemical laws, may be why automotive machine shops are on the decline in CA.

I agree. See my last post.

If the rods and pistons are balanced, then they should be a null factor when balancing the crankshaft. Not sure why there's such a mystery in balancing a flat six crank despite the reduction in automotive balancing shops. Contact a Porsche engine rebuilder/machine shop like Ottos, Competition Eng., Patrick Motorsports (or equivalent) and ask them the pertinent question(s). Don't assume a Porsche 6-cyl. crank is inherently balanced w/o confirming. It's a relatively minor expense compared to the overall budget.

Sherwood

My understanding is that balancing a Porsche 911 crank does not involve bobweights. I'm pretty sure that's what the balancer guy I used to use (RIP, alas) said. He took only a very little off here and there for race use.

And he was a stickler. He wanted to balance the pulley, flywheel, and PP together, and had to be persuaded that I wanted those components zero balanced so, as has been mentioned, they could be separately replaced.

Pistons can be an easy DIY balance job with an inexpensive digital gram scale. I don't think it is worth getting pistons within one gram, because carbon willl deposit unevenly and it doesn't take much of that to add a couple of grams. And by matching weights, you get as good a balance as you could anyway. Maybe at 20,000 rpm it matters, but I don't think having pistons which are close enough to the same weight get a bit lighter as they move out from the flywheel is going to put a twist on the crank.

Not so rods, which require a jig for end for end balancing, and have only a few small places where you can safey grind off material. The web on the bottom is not like a lot of Detroit iron - a big blob there to be ground down for balance. A shop once ruined a set of nice stock 911 rods by shaving away on the web. They weren't used to 911s.

Perhaps too late, but the 50% vs 100% sounds like a general design rule. I searched too, but could not find any info. I would recommend you contact a couple of race shops and ask them what spec they use for balancing their cranks. It might be beyond what you need, but it more than likely will be a spec that allows x amount of imbalance. Come to think of it, Costa Mesa R&D has a video on balancing cranks. They would be able to help you with a spec for your crank.

In machines such as compressors and steam turbines, it is a fairly common practice to balance rotors to 4 x W / N, where W is the weight and N is the max run speed. This calculation will be in oz-in. This is essentially an Navy spec. It may be of some guidance. You can read more here: http://www.irdbalancing.com/downloads/shoptolerances.pdf

What did you end up doing?

P.S.: found a site that explains it better than I did and it is focused on cranks....http://www.cwtindustries.com/whybalance.htm

Notice that their computer screen shows two balance planes, which means they are assuming for the purposes of balancing and correcting weight that the rotor is supported by two bearings, one at each end. So to apply the above formula, W is the weight at each end or essentially half of the total weight. The oz-in value is the max allowable in-balance. The balance software will calculate that for you and tell you at what angle the imbalance is so that if there is a correction needed you know where to add or remove weight. Anyway, probably clear as mud now.

Alfonso, Haven't done anything yet. Will contact Costa Mesa and see what they say. My case is on its way back from Ollies so I have to get to this soon. I saw the video you mentioned - it was for a Chevy crank.

I think the correct answers have been stated above. Here is some more background:

The 50% reciprocating balance rule does not apply to horizontally opposed engines. They have an equal reciprocating mass on each bank, moving in opposite directions, thus canceling each other out.

This is not true on inline engines or v engines, so counterbalancing the reciprocating mass is required. While the piston clearly moves up and down, the motion of the rod is more complex. It both rotates and translates. If you counterbalance all of the rod and piston mass (100%) then the engine is overbalanced. This makes for a very heavy crank. If you balance at 0% the engine is not balanced for the reciprocating mass. A compromise is used. This is where the 50% rule comes from.

We don't care. As I mentioned, the reciprocating masses cancel each other for the flat six.

The porsche crank is born balanced. The little counterweights are only to reduce bending force. It is 100% balanced as is, as long as nothing is broken off of it. It is a light, elegant part.

Yes, the design of the flat six crank maybe balanced, but you are speaking only in design terms that for example a flat six crank is better balanced than an V engine crank.

However, dynamically balancing a rotating part has to do with an actual measurement of how much in-balance there is in that component. If you measured this from one flat six crank to another coming off the assembly line after balancing you would get different measurements.

Some possible variables in why one crank would measure different in-balance compared to another could be:
1. variables in forging process
2. stacking of machining tolerances that locate and machine the journals
3. operator (human) variables such as how close one meets balance parameters vs another operator, etc
4. balance machine variables such condition of machine, machine sensitivity, etc

Speaking from job experience, a person in my group once went to inspect a gearbox (about $100,000 unit). We like to do a residual balance test which means that once the balance machine operator says that the rotor is balanced, we ask them to put a known amount of weight, at a known angle (usually 0 deg) and physically measured distance, then we ask them to spin it up. The software should tell you what you already know, that there is an in-balance of X amount of weight at so many degrees. Guess what, the software said something totally different which meant that their balance machine had an error. They quickly found that one of the four sensors had failed. They replaced the sensor and rebalanced our gear rotor. But there was also an inventory of previously balanced gearboxes ready to ship sitting right in front of that balance machine. I wonder if they ever went back and re-checked balance on all those crated up, ready to ship gearboxes.

The point is that it's a check that does not cost a lot of money. I would recommend to do it on any build where you have no history.

I contacted Henry at Supertec and he was kind enough to respond. He said the 993 crank is a 180 degree crank. No bob weights are used. He said most any shop that has a balance machine can do this. Here are the net takeaways for a 993 crankshaft balance:

1. Crankshaft gets balanced by itself without bob weights. I guess you could call this zero balanced.

2. The Flywheel, Pressure Plate and Pulley each get zero balanced individually.

3. The pistons including snap rings, rings and wrist pins optimally should be the same weight ( I read somewhere they all should be within .5 grams, this is what Costa Mesa R&D does). All six are weighed and the lightest one put aside. The remaining 5 have material removed from the wrist pin bosses or the inner skirt.

4. Both ends of the rods must be balanced separately. A special jig is used to weigh the small end and material is removed from the small end's outer-most surface. The lightest of the six is put aside and then the other 5 get material removed to within .5 grams of the lightest small end. To balance the big end all the rods are weighed (this time without the special jig) and the lightest put aside. The other 5 get material removed from the big end's outermost surface. This could be done with the bearings in or out because the weight difference of these is negligible, particularly if they are new. To see this being done look here: http://www.youtube.com/watch?v=QLpiF0E0EJU

Green 993 better call Henry back I think something was lost in the translation

regards

To re-summarize, the flat six Porsche engine crankshaft is inherently balanced by design. However, this doesn't mean the rotational balance at speed is within specs. Even a round tube can be out of balance, and to assume a flat crank is automatically in balance at operating speeds of 6000-7000 rpm (or beyond) is either too lazy or overly optimistic. An engine builder has many options and paths to putting one together. There are only a few preferred methods, but many more besides that. YMMV.

S

Racing 97 - do you suggest Green 993 should have said 993 (for that matter, any 911) crankshafts are balanced without bob weights? Rather than without counterweights?

What is wrong? I edited the above to change 'counterweights' to 'bob weights'.

Good edit on the counterweights Green, The Porsche engine is a boxer but it does not have a 180 crank the 914 does the porsche 911 has journals placed at 120 intervals and is not on a flat plane 3 up 3 down it does have aspiration periods that complement each of the banks that allow you to tune 123 into a collector as well as 456

regards

There is a common suggestion to identify pairs of equally-weighted piston/rod assemblies and place them in opposing cylinders (i.e. 1 and 4, 2 and 5, 3 and 6)

If the combined weights of an opposed pair of piston/rod assemblies (e.g. 1 and 4) aren't equal to another set of opposed cylinders (e.g 3 and 6), is the crankshaft assembly still dynamically balanced or has this introduced a secondary imbalance?

If so, is the ideal solution to have all six piston assemblies weigh the same?

Sherwood

The 911 engine is a boxer engine. The crankshaft has individual rod journals placed 180º apart. However, each cylinder does fire at 120º intervals which creates an overlap to help smooth power applied to the crank.

While inherently balanced, opposing boxer cylinders are physically offset from each other (not totally opposite) which would introduce some secondary imbalance. However, there are six cylinders, enough to compensate, and more tend to minimize the cylinder offsets. Boxer engines with fewer than 6 cylinders, e.g. 4 cylinder BMW engines as used in their motorcycles, must deal with this cylinder offset and resultant imbalance in their own way.

This link contains a chart that describes most engine configurations and their basic inherent characteristics:
Flat-4 engine - Wikicars

Sherwood

I know it is interesting the cylinders are at 180 the crank fires at 120 each journal has an 180 offset to its next journal but the firing order is 120 ergo it is not a 180 or flat plane crank.

regards

I appreciate your viewpoint, but the crank is not installed without balancing at the factory. It is balanced at the assembly point just to take into account the factors you mentioned above. Then it is picked up by human hands (for all 993's) and put into the main bearings, the case is sealed, and it spends the rest of its days there, until someone removes it. If no carnage has taken place, it will retain the factory balance indefinitely. Now, for higher rev than stock application, one would want to bring the balance tolerance close to zero perhaps, but for a daily driver, an undamaged stock crank is fine.

The point about indexing is an interesting one. I think this is usually skipped because nobody wants to destroy the ferritic nitrocarborizing (Tennifer treatment) just to tweek the journals by a tenth of a degree.

The firing order of an engine is a separate element from the physical layout of the cylinders. Regardless of the number of cylinders, all must fire within 720º of the 4-stroke cycle (thus 4 cyl= 180º; 6 cyl=120º; 8 cyl= 90º, 12 cyl=60º). Each cylinder of a boxer engine (including the Porsche flat-six engine) has it's own crank pin. FWIW, the flat 12 Porsche racing engine uses a V-type crank. On that engine, each crank pin is shared by the con rods of the opposing cylinders, thus technically isn't a boxer engine.

917 crankshaft:
http://forums.pelicanparts.com/uploa...1369352411.jpg

<iframe width="420" height="315" src="http://www.youtube.com/embed/IEgXUt1PSgg" frameborder="0" allowfullscreen></iframe>

Sherwood

Thanks, just trying to share my experience, as you and others here.

You are right that the crank is balanced at the factory. The only point I try to make is that a human has input into the quality of the balance. The condition of the balance machine is also a factor.

Most balance machine operators do not understand the math and simply follow what the software tells them. If something goes wrong but the software does not catch the error, then the operators are usually not smart enough to catch it either.

Maybe Porsche trained their balance guys back in the day to know the math and to know how to spot a bad balance despite the machine telling them it was good, but I think it's worth a few bucks to know that they got it right.

Now, if you check balance today and then in 2 years decide to upgrade your 3.6 to a 3.8, I agree that there is no need to check the balance.

EDIT: I guess in an engine build, as a DIYer, there are things one may choose to assume and things they choose to know. Is it important to know the weight of each piston? each rod? Is it important to know the diameter of your cylinders? Is it important to know if your cylinders have taper? etc, etc. You can either assume or know. To measure is to know. I prefer to measure as much as practically possible.

There isn't as much possible error in boxer motor cranks. You don't even need counterweights, as witnessed by the 911T. So you are only talking about a few grams rather than needing to re-grind counterweights after installing lighter rods and pistons as you would need to do on an inline engine, assuming it was balanced at the factory.

A significant difference in static component balance (weight difference) should show up as a primary harmonic since the weight difference will be noticed once per revolution. Forces that cause second order harmonics are noticed at twice the running speed of the crank.

From what I have read, if you keep the weight of the piston/ rod assemblies to within 1 gram of each other, you should be fine. There are many other factors, like friction, oil film, windage, crank flexure that have an effect as well, will make weight matching pistons to the .0001 g not necessary.

The boxer crank has inherently good first and second order harmonics characteristics. That does not mean you don't need to dynamically balance it.

On the other hand, if a few grams difference from one journal to another, spinning at 6500 rpm don't bother you, then that's entirely your dollar.

All I say is, if i don't know the history of the crank, I will ask to have it balanced, and preferably, I will watch and ask questions. Just my opinion and my dollar.

Boxer cranks are inherently balanced for all orders- 1,2,4,6,8,10,12,etc.

Here are the diagrams from my Porsche Boot Camp presentation:
http://forums.pelicanparts.com/uploa...1369392988.jpg