How to balance a Crankshaft???

[wz507]

Biknut,

Did you weigh the large and small end of the rod, or did you just take half the total rod weight and call it the small end? The actual split might be more like 45/55 for small and large ends respectively, or depending on rod design the differential from small to large might be even bigger.

Your approach sounds fine, but one big issue that could lead you astray is how the bob-weight is attached to the crank assembly. The bob-weight mass needs to be located concentrically or symmetrically about the crank pin. It can not be hanging from the rod which would apply the weight very non-symmetrically about the crankpin.

As you saw in the S&S article on balancing a Hog crank, the crank is disassembled, a pilot installed in the crankpin hole, and the bob-weight in turn attached to the pilot. In the case of the Hog crank the total bob-weight mass is divided by 2 (half the total mass for each flywheel), and 1/2 of the total mass used to balance each flywheel half separately.

I assume your crank is a 3 piece pressed together unit, so it must be pressed apart, the rod removed, and a pilot or the crankpin used to mount the bob-weight symmetrically on the pilot or crankpin. Of course the mass of the crankpin is part of the rotating weight, so must be included in the bob-weight calculation just like the heavy end of the rod must be included as rotating weight in the bob-weight calculation.

If you have a means of fixturing each of your flywheel halves separately, you could also use half the total bob-weight and balance each side of the crank assembly separately just like it was a Hog.

In closing, I'll remind you that the total bob-weight for your system will be the following if you want a 60% balance factor.

0.6 (piston + rings + keepers + pin + small end of rod) + large end of rod + crankpin

If the large and small end of the rod have loose bearings/cages, these are also included in the respective rod end weights.

To achieve static balance may require either weight removal from counterweight, via drilling, or may require addition of weight via heavy metal insertion (drill holes in counterweight and press tungsten alloy into hole to add weight).

Good luck and keep us posted.

 

[biknut]

The weight of the piston assembly, and half a rod, is 135.5

So 60% of 135.5 is = 81.3 g

Right now even without a piston on the rod, it already rolls down. Crank pin down.

Please let me know if this is wrong, or what you think.

Trying to worry about the rotating mass, and how that affects the balance factor makes my head spin, so I'm going to disregard it like most people before me, and concentrate on the recipracating
mass. This is piston, rings, wrist pin, needle bearing, clips, and half the rod.

I need to make a weight correction. The piston assy is 105.5 g, and half the rod (small end) is 35 g. These weights are about the same as many other people have posted since at least 08 so I guess things don't change much in china girl land.

105.5
+ 35.0
140.5 g

60% balance factor

140.5
x.60
84.3 g

So how much weight do I add to the bob weight?

 

[wz507]

Biknut,

The subsequent answer assumes the following.

1. The crankshaft is assembled and has no connecting rod on it
2. The bob weight will be symmetrically attached to the crankpin.

Your calculations above for the reciprocating component of the bob-weight are correct. As you note, you have neglected the rotating weight, which would likely increase the total bob-weight by ~ 1.3 to 1.4. Neglecting the rotating mass makes your 60% factor meaningless since you have neglected all the rotating mass. You took 60% of something but threw away 30-40% of something else so the entire exercise becomes meaningless.

The rod must weigh 70 g, if 35 g is 1/2 of it, thus you need to add at least this much more to the bob-weight. You also need to account for the rod large end bearing weight.

So, if you want a crank balanced to a factor of 60%, you have the reciprocating components correct at 84.3 g, but need to add the rotating mass to this, namely the 35 g from the large end of the rod, and the weight of the large end bearings. If you symmetrically attach this bob-weight (84.3 g + 35 g (large rod end) + ??? large end bearing) to your crankpin, you will then have the correct bob-weight to balance to a factor of 60%.

A picture of bob-weight attachment can be found at the following link.

tinyurl.com/ck7qcyq

Good luck and let us know how you make out

 

[biknut]

Biknut,

The subsequent answer assumes the following.

1. The crankshaft is assembled and has no connecting rod on it
2. The bob weight will be symmetrically attached to the crankpin.

Your calculations above for the reciprocating component of the bob-weight are correct. As you note, you have neglected the rotating weight, which would likely increase the total bob-weight by ~ 1.3 to 1.4. Neglecting the rotating mass makes your 60% factor meaningless since you have neglected all the rotating mass. You took 60% of something but threw away 30-40% of something else so the entire exercise becomes meaningless.

The rod must weigh 70 g, if 35 g is 1/2 of it, thus you need to add at least this much more to the bob-weight. You also need to account for the rod large end bearing weight.

So, if you want a crank balanced to a factor of 60%, you have the reciprocating components correct at 84.3 g, but need to add the rotating mass to this, namely the 35 g from the large end of the rod, and the weight of the large end bearings. If you symmetrically attach this bob-weight (84.3 g + 35 g (large rod end) + ??? large end bearing) to your crankpin, you will then have the correct bob-weight to balance to a factor of 60%.

A picture of bob-weight attachment can be found at the following link.

tinyurl.com/ck7qcyq

Good luck and let us know how you make out

I totally agree with you, and also thank you very much for this discussion.

Let's assume everything you've said is true. I've come to the conclusion that since these motors are so far off to start with, any improvement will be beneficial. I guess we're going to find out what getting close is like.

What you're suggesting is adding even more weight to the bob weight than what I propose. That makes the stock balance look even worse, if that's possible.

I was trying to get the crank to balance with a 49 g bob weight on the rod. 49 g bob weight + 35 g half rod = 84 g
I was only able to make a 40 g change. 6 to 9 g short of the goal.


At least now the bob weight side hangs down (crank pin up) without a piston on the rod. Before the crank pin always was down, so it's getting close.

So you're right about this not being balanced to 60% balance factor, but I ask does it really matter considering how bad most of them are?

That's a question we're going to find the answer out soon.

I'm going to try to post some pics

 

[biknut]

On the left is the stock fly weight like from the factory.

This is how it looks now.

 

[biknut]

The rod must weigh 70 g, if 35 g is 1/2 of it, thus you need to add at least this much more to the bob-weight. You also need to account for the rod large end bearing weight.

So, if you want a crank balanced to a factor of 60%, you have the reciprocating components correct at 84.3 g, but need to add the rotating mass to this, namely the 35 g from the large end of the rod, and the weight of the large end bearings. If you symmetrically attach this bob-weight (84.3 g + 35 g (large rod end) + ??? large end bearing) to your crankpin, you will then have the correct bob-weight to balance to a factor of 60%.

A picture of bob-weight attachment can be found at the following link.

tinyurl.com/ck7qcyq

Good luck and let us know how you make out

Let's talk about this rotating weight. I can understand how the reciprocating weight works, but the rotating weight seems different.

Reciprocating weight inertia is always up and down, but rotating weight inertia, is a lot less in the up and down plane. I'm thinking it would add to reciprocating weight at less than 100% of the total rotating weight .

Wouldn't there need to be some kind of forumla to calulate the effect of rotating weight?

 

[wz507]

Regarding the effect of rotating mass and the need for a formula to treat this uniquely, it's already accounted for in the approach you are following, i.e., the balance factor you use and the treatment of the small and large end of the rod.

Certain components of the reciprocating weight (piston, rings, wristpin) move only normal to the crank (as you note - straight to or straight away from the crank centerline), but the rod moves at all angles from 0 to 90 degrees from the crank centerline, and that is why the rod is apportioned into both the reciprocating and rotating weights of the system. Aside from the rod, the rotating weight simply refers to the fixed mass of the flywheel assembly, which is designed by the manufacturer.

In the case of a full circle crank (like yours, Harley, Honda singles, etc, etc) there is generally a thicker section opposite the crankpin to balance the system to some given factor. In the case of a traditional porkchop shaped flywheel assembly (most multi-cylinder automotive engines and Briggs & Stratton small engines), flywheel cheek material has been removed from that half of the cheek that the crankpin is mounted in for the same purpose (to add weight opposite the crankpin for balancing purposes). In the case of a Briggs engine, it is typically balanced to provide smooth operation at the governed speed. It will not however be smooth at all speeds, nor will any engine. Balancing is always a compromise and done to minimize vibration at the use speed, not all speeds.

If you took your little engine, removed the rod, assembled the crank in the cases and spun it at 2000 rpm, you could not hang on to it and it would jump right off the table. If you now wound solder onto the crankpin to provide 120 g of weight (static balance?) and repeated the experiment you would be amazed at how much the vibration would be moderated.

The bottom line is the following. If you symmetrically attach a bob-weight, consisting of 50-60% of the reciprocating weight plus the weight of the rod large end and large end bearing, to your crankpin and modify the weight distribution of the crank to achieve a static balance, you will have a relatively smooth running engine. But even then you will experience sweet spots where the motor is exceptionally smooth at a given rpm range and other operating regimes where resonance becomes much more pronounced.

Regarding the crank you intend to balance, I'm curious to know what factor it is balanced to at the factory. You could easily determine this by removing the rod, reassembling, and winding solder on the crankpin until it balances statically. Then weigh the solder, subtract the rotating weight from the solder weight to determine what reciprocating weight the factory used. And finally figure out what percent (balance factor) of your reciprocating weight this represents.

Hang in there.

 

[biknut]

Regarding the crank you intend to balance, I'm curious to know what factor it is balanced to at the factory. You could easily determine this by removing the rod, reassembling, and winding solder on the crankpin until it balances statically. Then weigh the solder, subtract the rotating weight from the solder weight to determine what reciprocating weight the factory used. And finally figure out what percent (balance factor) of your reciprocating weight this represents.

My wild guess works out to less 20% balance factor stock. That should make the motor real smooth at about idle speed, and that's about all.

If we can guess some of the weights, we should be able to estimate the stock balance factor.

105.5 g.....piston, rings, clips, wrist pin, needle bearing = 105.5 g

+ rod, big end bearing, and crank pin ????. I know a big end bearing is 7 g. A rod is about 70-75 g. I'm guess a crank pin is probably about 65-75 g

The stock fly weights come from the factory with about 20-25 g removed from each one.

I'll add it up. Feel free to make corrections.

105.5 g piston assy/ wrist pin, needle bearing, clips, rings
70 g rod
75 g crank pin
7 g big end bearing
257.5 g

The factory only removed about 20-25 g from each bolt on counter weight. The lightened part is closest to the crank pin.

 

[biknut]

My wild guess works out to less 20% balance factor stock. That should make the motor real smooth at about idle speed, and that's about all.

If we can guess some of the weights, we should be able to estimate the stock balance factor.

105.5 g.....piston, rings, clips, wrist pin, needle bearing = 105.5 g

+ rod, big end bearing, and crank pin ????. I know a big end bearing is 7 g. A rod is about 70-75 g. I'm guess a crank pin is probably about 65-75 g

The stock fly weights come from the factory with about 20-25 g removed from each one.

I'll add it up. Feel free to make corrections.

105.5 g piston assy/ wrist pin, needle bearing, clips, rings
70 g rod
75 g crank pin
7 g big end bearing
257.5 g

The factory only removed about 20-25 g from each bolt on counter weight. The lightened part is closest to the crank pin.

What this tells me is the factory expects you to run these motors at very low speeds.

 

[wz507]

Biknut,

To probe what the factory balance factor is, let's approach it in the following manner. If the crank is assembled with no rod on it, the large end bearing is on the crank pin, and we are balancing with the bob-weight symmetrically attached about the crank pin/large bearing, the only rotating mass we are concerned with is the big end of the rod, i.e., we don't need to be concerned with the weight of the crankpin, large bearing or the stock balancing plates, as these are fixed parts of the rotating mass/flywheel assembly as designed by the factory (they're all part of the factory balance system).

The following weights are the only weights relevant to the calculation.

Total Reciprocating Weight - 137 g
105.5 g - piston assy/ wrist pin, needle bearing, clips, rings
31.5 g - small end of rod (45% of 70 g)

Total Rotating Weight - 38.5 g
38.5 g - large end of rod (55% of 70 g)

The table below shows a variety of bob-weights with corresponding balance factors. All you need now to determine the factory balance factor is solder and a crank with no rod on it. If you don’t like the large end bearing being involved, just add its weight to the rotating mass and deal with it there.

As always - good luck.

 

[nightcruiser]

Biknut,

To probe what the factory balance factor is....

I would be interested to see how the factory balance turns out on the 4 flywheels that have been used for experimentation, and how similar they are to each other....

 

[biknut]

Biknut,

To probe what the factory balance factor is, let's approach it in the following manner. If the crank is assembled with no rod on it, the large end bearing is on the crank pin, and we are balancing with the bob-weight symmetrically attached about the crank pin/large bearing, the only rotating mass we are concerned with is the big end of the rod, i.e., we don't need to be concerned with the weight of the crankpin, large bearing or the stock balancing plates, as these are fixed parts of the rotating mass/flywheel assembly as designed by the factory (they're all part of the factory balance system).

The following weights are the only weights relevant to the calculation.

Total Reciprocating Weight - 137 g
105.5 g - piston assy/ wrist pin, needle bearing, clips, rings
31.5 g - small end of rod (45% of 70 g)

Total Rotating Weight - 38.5 g
38.5 g - large end of rod (55% of 70 g)

The table below shows a variety of bob-weights with corresponding balance factors. All you need now to determine the factory balance factor is solder and a crank with no rod on it. If you don’t like the large end bearing being involved, just add its weight to the rotating mass and deal with it there.

As always - good luck.

View attachment 43365

Your graph shows a 60% balance factor calls for a bob weight of 120.7 g.

I just want to point out that the actual bob weight from the factory is only about 45 g by my calculations.

 

[biknut]

Your graph shows a 60% balance factor calls for a bob weight of 120.7 g.

I just want to point out that the actual bob weight from the factory is only about 45 g by my calculations.

When I think about it, 45 g is probably about equal to the holes for the crank pin.

 

[biknut]

I would be interested to see how the factory balance turns out on the 4 flywheels that have been used for experimentation, and how similar they are to each other....

Right now I can only give you a general rundown.

Of 4 cranks I've sampled, 3 of them were like the one on the left and 1 was a new style like on the right. The 3 that were alike, one had bolt on fly weights that looked like they might have been from a different factory.

The only difference I can see is the color of the fly weight in the above left is shiny, and the one in the lower picture is dull iron looking.

Both cranks in the first picture had heavier bob weights than the 2 cranks I have that match the one in the lower picture.

 

[wz507]

Your graph shows a 60% balance factor calls for a bob weight of 120.7 g.

I just want to point out that the actual bob weight from the factory is only about 45 g by my calculations.

I'm not following. What do you mean when you say "the actual bob-weight from the factory is only about 45 g by my calculations"? The bob-weight is something you add and statically balance the crank to, not something from the factory.

I've never seen one of these cranks, but from the pictures it appears there is no design feature that adds any counterweight (weight opposite the crankpin) other than the plates that screw on? Am I seeing this correctly? If so, you could remove lots of weight from the crankpin side of the flywheel by drilling horizontal holes through the outer rim of the flywheel, which is much easier and much lower cost than adding heavy metal oposite the crankpin. Of course this creates extra undesirable free volume in the crankcase which you wouldn't want on a 2-stroke (this is a 2-stroke engine right?).

Still eager to see you press apart a crank and get to the bottom of it with actual data once and for all.

 

[biknut]

Take a look at the bolt on fly weight on the left. This is all the factory has done to balance the crank. They removed about 20 g from each fly weight on the crank pin side, much like you suggest. Just a lot less.

This makes the counter weight side of the fly weight about 20-25 g more than the crank pin side.

Once you understand what I'm talking about you see the magnitude of the poor job the factory has done trying to balance these cranks.

 

[biknut]

Ok, I finished building a new motor with this "balanced" crank, and installed it my Cadillac AV Sport. So far I've put 10 miles on it and got up to 27 mph. With my 36 T sprocket, that's not very high rpm. About 5000 maybe.

The report so far is, very smooth operation. Of course the real test will come later when I can rev it up to 7000, but for now it feels real good.

The old motor, a X80A from LEB, had a lot more vibration in comparison to this modified T80 from BGF. The cranks were the same in both motors.

Once I get the motor broken in and tuned up, I'll come back and make my final report and conclusion about the balance.

 

[biknut]

It turned out after all my effort of truing and balancing, the crank was pretty smooth, and the motor was mechanically quiet, but it still ended up a failure. I kept losing small end needle bearings. Ruined 3 in all, in 500 miles.

Turned out the rod was not properly machined on the small end, and was rough on the inside dia. This caused the needle bearings to fail.

Back to the drawing board.

 

[ivan H]

I noticed in the posting by Big E, page 2 describing fitting bearings 2 the crank then installing in cases that no mention was made of setting crank "end float". Crank end float on a Chinagirl should b set at between 0.003" & 0.008". The prescribed float is acheived by placing correct thickness shims between the bearings & crank wheels, also centralizing the crank in the cases. Cheers

 

[crmachineman]

Hi everybody! I've been very concerned about vibration as well. I have investigated the crankshaft and found a few things that were wrong. First, the worst thing I found was that the crankshaft assembly had angular misalignment. All three engines that I have done had pretty much the same defect. As you probably know, the crankshaft is an assembly that is pressed together; well, for understandable reasons, when the assembly is pressed together, it is not perfect. If you were to put one of these crankshafts in a lathe and spin it, you would be shocked!, I mean really shocked! The misalignment is very pronounced. You don't need a dial indicator to see it; it is totally visible! The crankshaft ball bearings were never designed to operate in conditions like that. The angular misalignment also throws the flywheel weights off, so that just adds to the vibration. When I build an engine, I straighten the crankshaft, then take a skim cut to true up the flywheels. It is a quick way to get reliability from the engine. The bearings just cannot cope with the shaft torquing from side to side due to that angular misalignment. All the crankshafts I've seen are bent the same way, and I'm sure your crank is bent too! It's terrible! Another thing I would like to share with you all is that the flywheel weights are lightened from the factory on the connecting rod side. On the GT-5A (and I presume the GT-5SR as well) motors, you can see the drill holes, but on the older models, the flywheel weights are held in place by 3 screws. If you were to remove one of those weights, you will see that they milled some slots on the inside of those weights for balancing purposes, and when they are attached to the plates, you can't see them. I tried further lightening the flywheel weights on the connecting rod side, and it made the vibration worse. I might start selling reworked crankshafts for people who want to redo their engines, but don't have a machine shop. I predict I'll get to that in about a month.
Thanks,
-Fred