Crankshaft balancing and truing (part...)

Discussion in 'High Performance Bicycle Engine Tips and Motorized' started by multipaul, Jul 5, 2012.

  1. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    Balancing a single cylinder engine can be done at home. This is done in two-stroke engines statically. The crankshafts are short. We have only forces to balance and no moments. No expensive machinery is needed.

    1. Everything is fine, as long as every moving part is turning around.
    Such a system of rotating masses can be balanced easily and with100%. Here all masses are concentric. So no counterweight is needed.
    [​IMG]

    2. Now the connecting rod appears. The bottom part of it with the crank pin and its bearing is 100% rotating. So it should be balanced 100%, like a concentric flywheel. The further we move away from the crank pin toward the piston the less the connecting rod turns and the more it oscillates.
    Trick: the lower portion of the rod is added to the rotating part and is balanced 100%.
    [​IMG]

    3. Now let's look at the piston, the rings, the piston pin and it's bearing. They go up and down 100% without any rotation. So does the upper end of the connecting rod. But the more it approaches to the crankshaft the more its parts make an elliptical way. Nevertheless the weight of the upper connecting rod is added to the reciprocating masses.
    If we don't balance these masses the engine will vibrate up and down.
    If we balance these masses with 100% everything is fine in TDC and BDC, but the engine will strongly vibrate at 90 degr and 270 degr. horizontally.

    Against the reciprocating weights I added the green masses. The green arrows show in which direction the forces point.
    [​IMG]

    That's why in praxis the engine manufacturer chose a compromise. Most single cylinder engines use a balance factor between 0% and 60% of the reciprocating masses. The masses to balance can be added opposite the crank pin or mass can be removed near by the crank pin. Often its a combination of both.
    The rotating masses are always balanced 100%. That is why we don't mention them!

    Big single 2strokes (for example the Ossas) with their standing cylinder had balance factors of 58% (57 to 60). These engines were used in motorcross, trail and also as very successfull grand prix racer in the 1970th.

    Part 2 is about the math.
    Part 3 is about the dynamic forces and internal speeds.
    Part 4 is about how I trued my crankshaft. With system but without a lathe.

    (Part 2 and 3 will appear perhaps tonight)

    Multipaul
     
  2. ivan H

    ivan H New Member

    Joined:
    Oct 8, 2011
    Messages:
    619
    Likes Received:
    0
    Cool, looking 4ward 2 rest of
     
  3. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    Part 2a: The math of STATIC BALANCING and doing it in practice.

    The math is different from dynamic balancing. But the result is the same.
    So it often happens that the persons using one and the persons who uses the other method, don't understand each other.

    We now want to calculate a replacement weight, the bob weight. This weight is acting on the crank pin, such as pistons, rings, etc. , but only with the intended percentage.
    When we add this bob weight to the down hanging conrod and the crankshaft stops in each position, the crankshaft is balanced statically.

    Preliminary: Determine the weight of the
    Piston
    Piston pin
    2 piston rings
    2 clips
    Small end bearing
    Reciprocating part of the connecting rod

    The first 5 positions are easy to determine.
    The 6th makes some difficulties.
    Position the connecting rod in that way that it lies as level as possible. The upper conrod end will now be weighed. (The weight of the lower connecting rod does not interest at all.)

    [​IMG]

    ...done. When we add our six positions above we get 110 grams.

    Now we think. We want to compensate the 6 weights together, let us say with 40% of their true mass. 110g x 0,4 = 44g for the bob weight?
    BUT THE CONNECTIONG ROD is hanging there with 100% of its weight.
    It should hang with 28g X 0,4 = 11,2g. But it doesn't. From the rod there are hanging 16,8 grams too much. Problem...16,8g too much. So the 44g bob weight is wrong.

    Don't laugh at me. You know the solution: Reduce the first calculated bob weight by 16,8 grams. 44g - 16,8g = 27,2 grams is the correct bob weight in this example.

    Formula:
    Sum all 6 positions => result1;
    result1 x percentage /100 => result2;
    upper conrod part x percentage /100 => result3
    upper conrod part - result3 => result4
    result2 - result4 => bob weight

    Now take some nuts, washers or whatever you have laying around and fasten it somehow at the conrod. Ribbons, wire, a bag - no matter. But its weight becomes part of the bob weight and must be considered.
    It should only hang freely and must not disturb the rotation of the crankshaft.

    The appropriate base for our work:
    You can position the crankshaft with the main bearing journals on two flat and exactly horizontal surfaces.
    The easier and probably exacter way is to use absolutely new not oiled or greased 6202 bearings and some brake pads. You will be astonished how difficult it is to prevent this unit from rolling off. Even if you thought the base was horizontal.

    The HT engines with 3 part crankshafts have kidney-shaped notches on the inside. This is the balancing from the factory. And this is the area, where we should take away some more material or add some.

    Here is the state of the flywheels from factory

    [​IMG]

    Miniature cutters in action

    [​IMG]

    Here, the state afterwords. 38g less. I can fill the holes again. Either to add some weight for tests or to keep the original crancase volume.

    [​IMG]

    (There will be a part 2b and a youtube video and an excel sheet for you to download. And the parts 3 and 4)

    Kind regards to Biknut. He discribed the procedure correctly in another thread.

    Multipaul
     
  4. Thud

    Thud New Member

    Joined:
    May 26, 2010
    Messages:
    205
    Likes Received:
    0
    hehe, Looks familure
    I run a 49cc engine & just machined the entire web out of the exterior crank parts.
    [​IMG]

    I carefully measured the bob weights off the lathe, they snap into the crank pin:
    [​IMG]

    & then checked for ballance on a set of level & parralell beams(smoothed angle iron on magnets shimed level/par. with paper)
    [​IMG]
    yes all the screws & running parts were in place....
    I used 55% or recipricating mass as a starting point.
    motor is smooth @ 9500 rpms now.
    hope you don't mind me adding this here....your doing a much better job than i would have...excellent documentaion! Thanx
     
  5. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    Part 2b: How to determine the actual balance factor

    When you put the crankshaft onto the apparatus for balancing 3 situations are possible:

    1. The conrod stops in the top position.
    (This is unlikely to HT motors. In other two-strokes this is rule)

    [​IMG]

    Now, find the bob weight by try and error until the crankshaft stops in each position. When this is done, put the complete bob weight on a balance.

    [​IMG]

    Math:
    Sum the 6 positions (piston + +...+ + upper conrod part) => result1
    bob weight + upper conrod part => result2
    result2 / result1 x 100 => actual balance factor [%]

    2. The conrod stops in the lowest position.

    [​IMG]

    Look at the picture below. You have to determine the down pulling force.
    Crankshaft pin and main bearing journal must be exactly horizontally. The conrod itself should press vertically to a scale. You must align and underpin a lot until everything fits. Read the value from the scale.

    [​IMG]
    [​IMG]

    Math: now line 2 is different
    Sum the 6 positions (piston + +...+ + upper conrod part) => result1
    upper conrod part - value from the scale => result2
    result2 / result1 x 100 => actual balance factor [%]

    If the result is negative that means, that the reciprocating masses aren't balanced at all. Even the rotating masses aren't balanced completely.

    3. Theoretical possible: The crankshaft stops in each position. In this case use the formula from case 1.
    Set bob weight = 0

    Now to the primised Excel sheet. I made it in Google-Excel.
    You should download it from there. I can only see the German version.
    In the pull down menu of "Datei" (=file) there is the option "herunterladen als" (=download as?) and then click "Microsoft Excel 97-2003 (.xls)"

    https://docs.google.com/spreadsheet/ccc?key=0Aht55HXLtHuOdEx6QTVVd0dPRF9LS3NSTjBCMU5EQkE#gid=4

    In the last line of the Excel sheet there is a suggestion how much mass to add or to remove. Remember, the area where material is removed or added, has a larger radius than the crank pin. That is why the values seem to be rather low.

    Multipaul
     
  6. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    Hallo Thud

    Never mind.
    My text is for those people who have no lathe. And a little bit for those who are not sure about how to calculate that all. There are different opinions about it from time to time.

    Now I want to talk about some inner forces and vibrations.
    This may know a lot. Then it's just a reminder.

    Multipaul
     
  7. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    One addition. Thud's bill is different. As a result of the horizontal conrod with the wire, the bob weight in that process must be heavier (+ conrod reciprocating part). The rest is the same. More important: the result of balancing is the same.

    Multipaul
     
  8. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    Part 3 Dynamic forces and internal speeds

    In the www I found an interesting link. Had I known sooner, I would have probably saved a lot of work. Here you see the forces that make the engine vibrate. At least the quality and direction. It is in English.

    How to use it.
    - The balance factor is choosable (0% to 100%)
    - rod to stroke ratio
    47/90= 1.9 for 66cc
    and 2.1 for 48cc
    - click "With secondaries" (yes, the length of our conrod isn't endless)
    - choose "Toggle radar plot"
    - choose "Against global angle"

    Then you can play very fine with different values.
    It is the program "Engine balance analysis". I downloaded it and my computer is still okay. Or you can only execute it, if you want to.

    http://www.tonyfoale.com/

    Days before I had started the same calculaten by hand.
    The equation of motion of the crankshaft and the piston gives answers about the piston position, its speed and the acting forces depending on the crankshaft angle. Then I programmed these formula in Excel, every 10 degrees one value. Although now I have the same curves, but the real ways, real speeds and forces of our HT-engines that appear on my charts.

    I realize that for many this is not interesting. It is rarely written about.
    It might interest perhaps one or the other.

    1. The way of the piston (HT 48cc and aproximated 66cc)
    Can be interesting for people who want to rise the ports.
    Can be helpful to make a port mapping.
    At 90 and 270 degr crank angle the piston is not on its half way, but below. This means that the piston is faster from the center of the path over TDC than BDC. That is not important, but one can have heard it once.

    Someone might read about an exhaust timing of 180 degrees. That doesn't mean, that the exhaust is open the half stroke. It is less.

    [​IMG]

    2. Piston speed (HT 48cc and approximate 66cc)
    The average piston speed is uncritical in our small engines. At 7000rpm, which is often called the red line, it is just 9 m/s. The diagram below shows the true speed.

    [​IMG]

    3. Forces at the piston pin bearing
    This is the first time, I watch the 48cc and 66cc separatly. The difference is too big. Notice that both charts show the forces with NOT.

    [​IMG]

    4. Vibrating forces
    That's what it's about a bit in this thread.
    After the ignition, the gas expands in all directions and has therefore no direct influence on the vibrations of the engine.
    I made all diagrams for 7000rpm. Any other speed is possible. The forces increase or decrease the square of the speed.

    [​IMG]

    [​IMG]

    ...
     
  9. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    ...
    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    Puh...Multipaul
     
  10. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    Part 4a: Aligning and truing the crankshaft.

    The last part is the try of truing the crankshaft.
    Up to today I hoped but I was not sure whether I would be sucessfull.This is the first time in my life that I had to true a crankshaft.

    I have no lathe, and I know that there are thousands of people where it's the same.
    When I opened my engine, I suspected immediately that the shaft would be quite crooked. It seemed like the flywheels were splayed apart a bit. This could be caused during assembly into the motor or by pressing the components together. Unlikely that this had happened already in the lathe. The hole for the crank pin could be drilled not perpendicular and so on.

    First, I built a simple device. That is what I needed. The bottom is 19mm plywood.
    [​IMG]

    Initial measurements were 0.4 mm eccentricity of the journal:
    [​IMG]

    I wanted to work as systemically as possible.
    So I identified three possible errors which are described in the following picture:
    [​IMG]

    Fortunately, in my engine mistake #1 did not occur. The threaded holes were all in alignment.

    Mistake # 2 was clearly present. I measured the inner distances between the disks. Then I marked the positions where to pull or to press. I've read about broken HT crank pins. So I warmed it up electrically to about 150 °C (300°F).
    Upper part shows how i pulled, below you see how I pressed. There was only few force required.
    [​IMG]

    With these measures, the error on the journal decreased to 0.2 mm. Trying to make both disks exactly parallel, the error enlarged again.
    So now I had to fight error No. 3.
    Some attempts with little heat were hardly sucessfull. The main bearing journals are not hardened. I was very careful with them and treated them with hardwood and a 500g hammer. Note that only the upper disc is clamped.
    [​IMG]
     
  11. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    (Part 4b)

    Only when I worked at a higher temperature, there were noticeable improvements. I used a soft gas flame, which is intended for brazing copper solder. But I remained below 200 °C (390 °F), so that the steel did not change its color. (The hardened crankpin should not be heated so much!)

    Here is my final result:
    [​IMG]

    Initially I wrote readings directly on the disc. After some attempts that became a bit messy and confusing. Later I made templates to assess improvement (or worsening) more effectively.
    [​IMG]


    If you have questions or critical comments, no problem. Perhaps there is another and also better methods. In this regard, I am open minded and happy to learn.

    If parts of this posting are not well understandable because of the language and the used technical terms, the moderators may change my sentences to make them clearer where necessary. It would bei nice, if they would keep me informed about changes.

    Finally I'll post a video that shows the achieved results. Remember, we worked down from 0.4 mm runout on the main bearing journals.

    Up tp now that was a complex but cheap pleasure.

    Multipaul
     
  12. ivan H

    ivan H New Member

    Joined:
    Oct 8, 2011
    Messages:
    619
    Likes Received:
    0
    Hi Multipaul, condition 1 u show is "indexing" of the crank, not a home job, & u'd like 2 hope the crank manufacturers got it right. Condition 2 can b caused when assembling the crank in the cases. For this reason, the crank spindles should b "pulled" thru the cases. U can do this with packing washers & the nut/bolt on the spindle ends, it just takes a couple of goes at each side. Great thread, well done. Cheers
     
  13. multipaul

    multipaul New Member

    Joined:
    Mar 31, 2012
    Messages:
    75
    Likes Received:
    0
    Hallo Iwan

    Thank you. In the meantime I'm not sure about this thread. Probably too long and too complicated in some parts.

    Finally I had to do some changes on the little video.
    It shows how I balanced and how I measured the results of the truing. It is about 4:15 minutes.
    http://www.youtube.com/watch?v=ONbEV7O4tls&feature=youtu.be

    Working on this thread, I got some clarity about relationships. For me, there are some conclusions:

    1. The lifetime of the bearings
    The limiting factors are the max dynamic load on one side and the fatigue load limits on the other.
    Barings which are loaded maximal quit their lives (statistically) after 1 Mio revs, which are 70 miles.

    2. Balanced engines compensate forces internely.
    Balanced motors compensate internal forces, although they are still there. They just can't reach the outside. This generally doesn't generate additional power.

    3. Trued crankshafts generate more useable energy than not trued ones.
    Vibratory plates also need a separate drive.

    4. What is the role of the gases. I wrote pretty much none at all.
    Anyone can check this by opening the gas at a certain speed and closes. Changes?

    5. A bicycle should not vibrate stronger while driving at any speed than with the clutch pulled on the stand.
    Otherwise, the vibrations coming from somewhere else. Mostly of the drive, chain, sprockets.

    Multipaul
     
  14. Goat Herder

    Goat Herder Gutter Rider

    Joined:
    Apr 28, 2008
    Messages:
    6,251
    Likes Received:
    0
    This is a great thread! Thank You kindly Sir for sharing it too. Been following it and have enjoyed all technical aspects of this whole set up here immensely .(^)
     
  15. ivan H

    ivan H New Member

    Joined:
    Oct 8, 2011
    Messages:
    619
    Likes Received:
    0
    + 1 GoatHerder, I think its a great thread, & ur right on that last point MultiPaul. I think a lotta vibration comes from that chain tensioner. Also, notice the countershaft bearing stayed in the left case half. Did the right side bearing stay on the countershaft, not in the case. I think its a kinda loose fit in the case 2 overcome that the cases provide no means of aligning this shaft. If, upon assembly, u align it properly & seat the bearing in the case with loctite bearing retainer, u'll notice a lot less rattle from the motor. Cheers
     
  16. Bonnygirl

    Bonnygirl New Member

    Joined:
    Dec 12, 2016
    Messages:
    1
    Likes Received:
    0
    Hi multipaul. Great post but should there not be an allowance made for the radius at which the weight is removed? It was removed from 2 locations either side of TDC which compensates somewhat for the radius being greater than the crank pin but I still think you have removed more mass than is needed for a 50% ratio. My guess (looking at your pics) is that your balance is probably about 70-75% of the reciprocating mass.
     

Share This Page