Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Gear Box Design Help

  1. Jan 10, 2009 #1
    I am trying to design a gearbox for a project i have.

    I have calculated a fair few things about what i want to get from the gearbox, but now i have came to design the gear box i have got bogged down with all the various things, so help please if you can.

    Data is

    Input:7850 rpm
    Output: 1975 rmp

    Power that is transmitted is 29.75kw

    Permissable bending stress is 380mpa and shear stress 280.

    I am going to use a modulus of 3 for the gear tooth.

    I want to use 4 gears, but little unsure on what gears to use? Application is to take a motion from an electric motor and output, much like a gearbox is a Drill.

    I am a little unsure, well lost really, on how to calculate the ideal number of teeth for Gears 1, 2, 3 and 4, and there diameter and width.

    Is there a minimum width to dia ratio?

    Application is a small generator.

    Thanks for any help.
  2. jcsd
  3. Jan 10, 2009 #2
    you dont have to calculate the ideal number, u have to base your calculation on the minimum number of teeth to avoid interference. then assume the shaft distance based upon the constraints & find the width of the tooth.

    Consult some data handbook, for a start.
  4. Jan 10, 2009 #3
    also, select the type of gears you would want to use, the choice would depend upon the application type & $$.
  5. Jan 11, 2009 #4
    Thanks for that.

    What gears would be suiteable for the application of a simple reduction gear box?

    $$ dont matter too much as its a 1 off project and as long as i can get them of the shelf there wont be any issues with cost.

    Not having the time to study this in depth, would be great if you could provide a formula or two if someone knows them off the top of their head, or knows where to find them/.

    How would i calculate to obtain the desired output speed?

    thanks again
  6. Jan 12, 2009 #5
    its not a question of suitable gear, its a question of application. For example, spurs are the least costly, they are noisy too, helical are smoother, they are a bit heavy on $$, & also axial thrust comes to play, so its sort of an optimization question, well not entirely optimum, but a smart choice has to be made.

    Also, this is a project work, it aint going to the production lines, so take that into account also.
    i wouldnt bank on that. It would be really helpful if you go through a design book. Get some book, like Joseph shigley, Mechanical Engg Design, it has a complete section devoted to gears.
    Get a data handbook or you may consult AGMA guidelines, but better to study first.

    you already have it, it is 1975 rpm(it isnt speed, its angular velocity, are you talking about the pitch line velocity??)
  7. Jan 12, 2009 #6
    Ok, all points taken.


    I do have the output speed =) your right

    What i should have said is how do i calculate the size of the gears and the value of teeth required?

    I know what the reduction ratio is, but because of the torque and stree i have decided that i want to use 4 gears to reduce the speed.

    1975 is my desired output RPM, i have not calculated to pitchline or angular velocity.
  8. Jan 12, 2009 #7
    you have a a moderate gear ratio,being ~4, you can get it in one go, why do you want to do it in two stages?

    its 18 for spurs, & similarly some number for helical depending upon the helix angle. I suggest you take a look in a data handbook.

    Now you have G, gear ratio, & number of teeth on the smaller gear. assume a module & calculate pcd.

    Do you plan to get them made in workshop? Have you considered using chains??
  9. Jan 13, 2009 #8
    Space is limited, so chains are not going to be used, but thanks for the thought.

    Gears will from stock/shelf as my calculations so far would indicate they are pretty standard size, and this just eliminates extra work in machineing.

    I have deiceded to use spur gears and Ball Bearings.

    My attention now turns to the stress that i will put upon them so i can choose the right ones from stock.

    Could FEA be used with relative easy to calculate stress? could it be used to predict wear too?

    If i use an equal number, then the same teeth will be hitting, thus i can predict a wear patten right?

    Got a databook, but its just data with little explanation, so here is a dumb question:

    Helix angle equations only apply to Helical gears?
  10. Jan 13, 2009 #9
    since you ll be using the gears off the shelf, module is already defined, only optimization left I can think of is to choose a face width & the pcd of the gear(& both are interdependent!!). Go for a pcd & calculate the face width using the lewis equation(calculate carefully the different factors used in lewis equation). Choose the one which is equal or larger(generally, if you are picking it up from elsewhere) than the dimensions calculated. Thats pretty much it.

    i fear not, it is pretty much difficult & is most probably not the way, you imagine it is . As I suggested earlier, get a design book.

    throw some light here, i am in dark, lolz

    yups, you shouldnt be worried about it since you are only dealing with spur gears.
  11. Jan 13, 2009 #10
    You strongly want to avoid having a common factor in the tooth numbers of gears that engage. If there is a common factor, the same teeth will engage repeatedly, setting up a wear pattern as was suggested, and this is very much to be avoided. Thus the tooth numbers for any gear pair should be relatively prime which is to say that they have no common factors.
  12. Jan 13, 2009 #11
    Ok, few more questions.

    Why is 18 or more teeth advised?

    If i used prime numbers for teeth values, then the wear pattern wouldnt be predictable right? This wouldnt be advised for what reasons?

    Is it easy to get a wear pattern with prime number as i would want to establish when its right to replace the gear set?

    Sorted most other equations out now.

    Working on lewis factor at the moment, but now turning my attention to shaft fit and keeping it free, how would i go about choosing ball bearings for the case?

    Thanks for all again
  13. Jan 13, 2009 #12
    There is a minimum number of teeth allowable on the pinion in order to avoid what is called undercutting. This is where the cutter forming the teeth actually cuts away a part of the involute tooth form outside the base circle in the process of forming adjacent teeth, resulting in a non-involute profile for that part of the profile near the base circle.

    The use of relatively prime tooth numbers is to spread the wear evenly over all gear teeth. If this is not done, fairly rapid wear may occur requiring very early replacement of the gear set.

    Note the difference between being relatively prime and prime. Two numbers are relatively prime if they have no common factor. Two numbers are primes only if each of them is a prime number. It is far too limiting to require that only prime numbers be used for tooth numbers.
  14. Jan 14, 2009 #13
    wow!! never thought of that before:surprised
  15. Jan 14, 2009 #14
    Ok, well i am very thankful for all this help.

    I think i will be back here asking more questions on this, but will post some more specific problems based on my design.

    Thanks gain all
  16. Jan 26, 2009 #15
    I have managed to get 17 different combinations to get my output speed.

    So how to i choose?

    Am i looking for a lower ratio on the first gear pair?

    Am i looking to use all even, or all odd teeth numbers or can odd and even teeth numbers be mixed with little issues?

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook