Results 1 to 13 of 13
  1. #1
    Join Date
    Oct 2011
    Location
    Norwood-ish, Adelaide
    Age
    59
    Posts
    6,540

    Default Cutting helical gears

    As someone said with the correct mill and dividing head and gearing a basic operation but I suspect that one off-gear making using a mill is becoming a rare thing - as Andre says, it is not taught at TAFE these days. For starters you need a universal mill* and universal dividing head. You could possibly do it on a CNC machine but that starts getting complicated and expensive. I was talking to a machinist just the other day who was telling me that these days simple jobs are not being done because as soon as you ditch your manual machines for CNC it costs too much.
    Strictly speaking we talk about cutting gear teeth but in reality it is cutting the spaces in between
    So -
    In case there are some who haven't seen a horizontal mill set up for spur gears, here is mine set up in that manner
    P1020874 (Medium).JPG
    A key thing to notice is that the plane of the cutter is co-incident with the axis of the dividing head so as the table feeds, the cutter cuts a straight groove in the blank to produce a gap.
    To get the basic sizes, you use equations for the OD and depth of cut (see attached sheet)Formula.pdf - equations 1, 2 and 3.
    (Sorry about the quality - the pdf writer and the jpegs are conspiring against me.

    Similar equations are used for helical gears except that Normal DP is used as per equation 5. I have a spreadsheet set up to calculate that all for me.
    P1020876 (Medium).JPG

    For the sample gear that I am making, the helix angle is 20 degrees so the normal DP is 12.8 (or as I've done and cheated - M2). Two important things result. Firstly the depth of cut will reduce slightly and secondly the cutter will change. To cut a 12Dp helical gear of this size I will use a M2 cutter and using equation 4, I will select the cutter as if it has 62 teeth. This is because when cutting on an angle the shape of the gap changes and this is to try and get it back to approximately involute (otherwise the teeth would be fatter and the gap smaller).
    The table is kicked over 20 degrees and gearing put on the back of the dividing head. I get the gear selections from tables in Machinery's handbook. The shot is probably a little too close, but the thing that looks like a tyre lever in the first photo is loosening the 4 bolts that pivot the table around (up to 45 degrees either way). To mount the gears as in photo 4 the handwheel is removed, a spacer taken out and substituted with a gear (and the handle replaced). If the gears sizes are not good you may need to add an idler or two to get the train correct. To reverse direction (have the DH turn CW rather than CCW) when the leadscrew is turned requires just one additional idler in the train. Photo 4 is the gear caddy which keeps all but the largest in one place.
    P1020881 (Medium).JPG P1020875 (Medium).JPG P1020877 (Medium).JPG P1020878 (Medium).JPG

    With the table kicked over and the gears in place it is almost time to cut. The cutter needs to be directly above the blank for the cut to be symmetric around the gear centre line. I use the ruler trick and measure the distance the two wings of the ruler are above the table. In the second photo the tip of the cutter is just rubbing the surface of the plastic, demonstrating the path that the cutter will take across the gear blank. Photo 3 is the cutter actually doing something.
    P1020882 (Medium).JPG P1020880 (Medium).JPG P1020884 (Medium).JPG

    Indexing is done as normal for a dividing head but the index plate is only there keeping the relative position of the blank to the leadscrew. The photo shows the device this dividing head uses to allow the indexing plate to rotate. When feeding this whole assembly rotates. While cutting helical gears differential indexing can not be used, so if I was cutting PDW's mythical 73:127 gear set I would have to make up indexing plates with that number of holes so I could index 40 at a time - so it could be done. Herring bone gears are beyond me at the moment though as the gap between the two opposing gears is only one (circular) pitch of the gear so the cutter is too large. To make herring bone gears it would have to be done in two halves and then fastened together. Commercially they are made with gear shapers.

    P1020886 (Medium).JPG


    *I suppose you could do it on a vertical machine with a tilting head but then you still need a leadscrew take off to drive the dividing head
    Michael
    Last edited by Michael G; 29th Mar 2015 at 07:47 PM. Reason: adding more

  2. #2
    Join Date
    May 2011
    Location
    Murray Bridge S Aust.
    Age
    71
    Posts
    5,942

    Default

    Thanks for that Michael, now I know how it's done. Way to technical for me,at the moment, but a very interesting read, thanks again.
    Kryn

  3. #3
    Join Date
    Mar 2012
    Location
    Western NSW
    Posts
    543

    Default

    Hi Michael,

    Thanks for the great post and photos. Cutting gears is something that I would love to be able to do but it does involve putting together the appropriate tooling and information.

    An appropriate milling machine (preferably with horizontal milling capacity and for helical/spiral milling needs a universal table), dividing head (and again for helical gears you need a DH that can be driven by gears through the milling machine leadscrew), mandrels for mounting the gear blank, tailstock and of course the cutters (of which there are 8 per DP or metric module pitch.... so even having all the common sizes on hand takes some doing) not to mention a means to measure the accuracy of your finished gear (gear tooth verniers, gear wires or other means depending on how accurate you want to be). So few hobby workshops and not too many jobbing engineering shops are still making one off gears ( from my farm equipment experience you just order one in unless it is something unusual in which case you try to repair the old gear if it is just a tooth or two missing or pay an arm and a leg to get a specialised gear manufacturing outfit to make one for you). No fun in that so I want to try and make them myself.

    I would like to think this thread might be a conduit for those who can cut a gear to teach those of us who would like to learn how to cut a gear.

    So I might start by throwing a few questions out there to get things rolling. I might give my current understanding as to the answers but remember I have not actually cut a gear myself but I have watched and gone through the process with my machining mentor who lives near me (old school but ridiculously generous with his time and equipment and puts up with me asking all those beginner questions....... and the more I learn the more I appreciate how much he knows and how much I still have to learn..... and I hope maybe one day I can have the same influence on someone else that he has had on me.....but I digress)

    So in general I am generally going to want to either reverse engineer a gear I have broken or is worn out, or make a gear that is missing or an additional gear for a set. This means I need to do the initial calculations to work out what exactly I need to make and what I need to make it.

    Lets start with straight spur gears and later can move onto the more difficult helical type gears. I'll start with imperial gears as I find this easier but can look at metric if anyone wants.

    Now remember this is me thinking aloud and not a list of instructions on how to do it. Please feel free to criticise/add comments/discuss an alternate way of doing it as most of this has come from reading and watching rather than doing (which hopefully will follow soon).
    So what information do I need to get started.

    To begin with I need to work out or measure a few things. The first thing is I need to know 2 of the following 3. Diametrical pitch,number of teeth or outside diameter. This is because if I know any 2 I can work out the third from

    Outside diameter = Number of teeth + 2
    -----------------------
    Diametrical pitch

    Now number of teeth is generally going to be easy and the outside diameter can be roughly measured to make up a whole DP or if I have a gear I can check its diametrical pitch with a gear pitch guage if I have one. So that gets me started.

    Next is to work out the pressure angle of the gear I want to make. Imperial gears are generally either 14 1/2 or 20 degrees. Again the gear pitch guage can help identify which you have. Otherwise you can download a full size model of the planned gear from the internet or special gear programs and compare. Any other methods?

    Next thing to consider is how much backlash you want in your gear. There are a whole lot of recommendations for this in the machinery handbook depending on the use and size of the gear you are making.

    With this information I can work out the size of the gear blank I need to make, the depth of the cut I need to make and the cutter I need to use. These formulas are all in the machinery handbook.

    Before I get started I just need to work out how am I going to measure my gear to know it is within the tolerances I want. You can just cut the gear and see if it works but for completeness I like to be able to measure things to know they are going to fit. Gear tooth verniers or over the wire measurements are two ways and I would be interested in what people think about the best way to measure your newly made gear. I would like to be able to measure over the wires as this allows you to have a minimum and maximum size (ie a tolerance) and also allows you to accurately incorporate a backlash figure into your measurements. I can go into this more but I would like to see what the forum members who actually make gears think.

    So Michael and anyone else, is this sort of how you think when you approach making a gear? Do you have anything to add? I know this is probably basic to the old hands hear but I am keen to get it more or less sorted in my head as to how to best approach this before I start. Will move on to the actual process and the thinking around helical gear cutting later on.
    Look forward to hearing from anyone who can set me straight.


    Regards

    Mark

  4. #4
    Join Date
    Oct 2011
    Location
    Norwood-ish, Adelaide
    Age
    59
    Posts
    6,540

    Default

    Quote Originally Posted by markgray View Post
    ...Next is to work out the pressure angle of the gear I want to make. Imperial gears are generally either 14 1/2 or 20 degrees. Again the gear pitch guage can help identify which you have. Otherwise you can download a full size model of the planned gear from the internet or special gear programs and compare. Any other methods?
    I usually work on the rule that DP gears are 14 1/2 degrees and Module are 20. Occasionally a DP gear will be 20 though just to throw you. Another method I've heard used (only ever tried it once with mixed results - probably practice needed) is to roll the gear on a strip of blutak or plasticine and measure the angle of the rack that you create.

    It's worth saying that a gear made with a milling cutter is not going to be 100% accurate. The cutters are made so that a gear with the lowest tooth count for the cutter will be but as the tooth count gets bigger the form becomes less and less accurate until you jump to the next size. If you want better accuracy you probably need to look at either hobbing or a gear shaper.

    For the gears that I make, I rarely worry about measuring them post milling for several reasons
    • Recognising the limitations of milling cutters (and especially for helical gears) I'm likely to be disappointed
    • If I start with the correct OD*, the right DOC and right cutter, the result should be correct
    • Most gears I make are mounted on centres that can be adjusted (eg change gears) so mounted in a way that minimises any systematic error (within a small range involute gear centre distance can accommodate minor variations too - trade off is more backlash)
    • For the small quanitites I cut, I can't afford (time or materials) to cut a trial piece just to set up a process. It would be different if someone wanted 10 off gears and was paying. Having said that, the gears I've made for myself and others seem to work.


    *If you cut the OD slightly under, reduce the DOC by that radial amount. A couple of thou less on the tip of the tooth will not matter.

    One thing I do do though is once having cut all the tooth gaps is index around to the first gap cut and recut. If the indexing has been correct then you have not got a problem. If it is out then you start cutting more metal (and cursing). People do say that you can cut gears on a plain mandrel without a key. For small teeth perhaps but for larger ones the vibrations can be enough to move things if not 100% fixed and secure. This is doubly so for helical gears as the cutting forces are pushing the blank sideways as well as along the mandrel. I cut the helical for my SG on an expanding mandrel and it worked but only just.

    P1020887 (Medium).JPG
    This is the end result of yesterday's efforts. The texta marks are another QC measure - I mark off groups of 10 teeth as I go so that I have a fair idea when I've finished that I set up the indexing properly. The tooth form does not look quite right but that is probably due to the approximations in the process more than anything.

    Michael

  5. #5
    Join Date
    Jun 2007
    Location
    sydney ( st marys )
    Age
    64
    Posts
    4,887

    Default

    As Michael has covered most of what's relevant to spur gears the only other factor to consider if repairing or making gears for an existing machine/application is the OEM may have made gears to suit there particular application and not to the standard sizes.

    This is sometimes the case with early lathes and other machine tools.

    Vairences can often be found in there OD, tooth counts and depth of tooth.

    If you have a sample/ the damaged item it is not generally a problem but can rare its ugly head if not aware.

    Hercus is one manufacturer in particular that strays from standard to suit there particular applications.

  6. #6
    Join Date
    Jun 2011
    Location
    Australia east coast
    Age
    71
    Posts
    2,713

    Default

    Quote Originally Posted by Michael G View Post
    People do say that you can cut gears on a plain mandrel without a key. For small teeth perhaps but for larger ones the vibrations can be enough to move things if not 100% fixed and secure. This is doubly so for helical gears as the cutting forces are pushing the blank sideways as well as along the mandrel. I cut the helical for my SG on an expanding mandrel and it worked but only just.
    I've never used a keyed mandrel for cutting spur gears up to 8DP. I can see it could well be an issue with helical gears as there is a significant rotary force.

    For sizable gears I use an angle plate bolted to the table to back up the gear and take some of the thrust/vibration away. Clamp optional - you always forget to release it when indexing for the next cut. The backup is especially useful if I've blocked up the DH & tailstock to get more swing.

    At various times I've cut a lot of gears, then years go by before I need to do it again. I don't have a universal DH ATM so the helical gear cutting capability swivel table of my Victoria U2 mill goes un-utilised.

    PDW

  7. #7
    Join Date
    Oct 2006
    Location
    Armidale NSW
    Age
    52
    Posts
    915

    Default

    Interesting thread.

    Where I have needed to cut a complicated internal keyed hub (after cutting the gear), I've made the gear a little thicker and tack welded the gear to the mandrel with a TIG welder. Just machine off the welds when done.
    Cheers.

    Vernon.
    __________________________________________________
    Bite off more than you can chew and then chew like crazy.

  8. #8
    Join Date
    Jul 2006
    Location
    Canberra
    Posts
    288

    Default

    Cutting one-off helical gears at home would no doubt be a triumph. If it is the most important thing to do among the many things most of us have planned then go for it. On the other hand, you can buy stock size helical gears from Boston Gear in the US. Their gears can also be bought on Amazon. I bought a pair for a machine part that was worn out. Great quality. I had to resize the bore, and found that turning was very smooth. The teeth however are hardened. The price for memory was under $100, for a 1.75 and a 2.5 inch pair delivered. I still needed to read up on helical gear design in an advanced textbook to be sure of my choice .

  9. #9
    Join Date
    Jun 2011
    Location
    Australia east coast
    Age
    71
    Posts
    2,713

    Default

    Quote Originally Posted by Gunnaduit View Post
    Cutting one-off helical gears at home would no doubt be a triumph. If it is the most important thing to do among the many things most of us have planned then go for it. On the other hand, you can buy stock size helical gears from Boston Gear in the US. Their gears can also be bought on Amazon.
    Yep. *Stock* size gears. I agree, not worth making them if you put any value on your time - except as a learning experience for when.....

    You try pricing gears from 600 Machinery for a Colchester lathe. You'll see why making your own has a great appeal. Also owning a slotter so you can do the internal splines.

    And I'm not even talking about helical gears here WRT outrageous pricing at 600 Machinery.

    PDW

  10. #10
    Join Date
    Oct 2011
    Location
    Norwood-ish, Adelaide
    Age
    59
    Posts
    6,540

    Default

    Most of the gears I've cut have been change gears for the lathe and dividing head. For example the DH requires 12 gears from 24t up to 100t (they are the ones in the caddy and on the mill/ dividing head). The "Simon set" of lathe change gears to allow me to cut worms is another 28 gears from 30 to 73 teeth. I'd much prefer to cut those myself rather than pay someone else several hundred to provide and based on the cost to buy them, I see cutting my own gears as a necessary saving while my workshop gains capability even if it can be slow and tedious.
    As pointed out by Pipeclay, machinery makers had unique sizing for their gears too so finding a stock gear the right size and shape is usually the main issue.

    PDW's comment about a backstop is a good suggestion. I find cutting large steel gears they can tend to ring a bit so a backstop helps damp that out too.

    Michael

  11. #11
    Join Date
    Feb 2013
    Location
    Laidley, SE Qld
    Posts
    1,038

    Default

    Being able to cut helical gears at home is pretty damn impressive.

    Query about the gear train between the lead screw and the dividing head. I'm guessing that for the gear you are cutting the gear train rotates the dividing head only by half a degree or so as you cut each tooth?

    How do you cope with backlash in the gear train as you back the cutter out, drop the table each time before you back out?

  12. #12
    Join Date
    Oct 2011
    Location
    Norwood-ish, Adelaide
    Age
    59
    Posts
    6,540

    Default

    Quote Originally Posted by bob ward View Post
    Query about the gear train between the lead screw and the dividing head. I'm guessing that for the gear you are cutting the gear train rotates the dividing head only by half a degree or so as you cut each tooth?
    The amount of rotation varies depending on the helix angle and the work piece diameter. One way of thinking about it is like a boat (or aircraft) propeller where one revolution travels a certain amount (the pitch). In setting up a dividing head for helical/ spiral milling that is referred to as the lead. The gear train is varied depending on the ratio of the work lead to the table lead (that is, the distance that the table would have to travel for the DH to travel one complete revolution with 1:1 gearing. In the case of this mill the lead is 200mm (metric mill). The tables I showed earlier are for a 10" lead machine so a minor bit of maths is required but it can be worked out

    Tan (helix angle) = PCD x Pi / work lead

    Ratio of leads = work lead/ table lead
    (Multiply this figure by 10 to get the ratio listed in the machinery handbook tables)

    In the case of my sample gear, the lead was 895.23mm, giving a handbook ratio of 44.76. The closest ration listed was 44.79, so the helix angle followed was really 19.98 degrees not 20


    Quote Originally Posted by bob ward View Post
    How do you cope with backlash in the gear train as you back the cutter out, drop the table each time before you back out?
    All the books say that you should drop the table but that gets really tiring really quickly (literally) and can lead to varying DOC. This mill has an anti-backlash screw on it and I find that provided the gear train is properly adjusted (to minimise the backlash in the gears) the combination of the two is not too bad. The reverse journey perhaps shaves something off the side of the tooth but not terribly much. Again, given that the form you cut is (close but still) approximate and the as shown above there will be other errors such as the gearing ratio, it's another thing you know is there but not necessarily anything worth worrying about. The microscope racks and pinions I made for BT for example would have had the same errors but meshed fine. Similarly the little gear for the sander seemed to mesh satisfactorily.

    The biggest problems I have with cutting gears that leads to inaccuracy is either not having the blank mounted so that the tooth PCD and the bore are concentric (large commercial gears will have a high spot marked - all gears have this problem; it's just the order of magnitude that is the issue) and as previously mentioned, errors in indexing.

    Michael

  13. #13
    Join Date
    Aug 2008
    Location
    near Rockhampton
    Posts
    6,216

    Default

    Cutting them on a CNC mill with fourth axis is extremely easy to do with the gearotic program...

    However buying a CNC of suitable rigidity and keeping it running plus a fourth axis is a whole new kettle of fish compared to buying a now obsolete by modern industry standards horizontal mill and universal dividing head.. it is more cost effective for the home shop to do it the way Michael has described...

    There are pro's and cons to each side of how to do it and there is no right way...
    Gold, the colour of choice for the discerning person.

Similar Threads

  1. Helical gear confusion
    By Ueee in forum METALWORK GENERAL
    Replies: 4
    Last Post: 10th Apr 2012, 01:10 PM
  2. Helical milling
    By Michael G in forum METALWORK GENERAL
    Replies: 1
    Last Post: 21st Oct 2011, 01:37 PM
  3. Cutting gears?
    By neksmerj in forum METALWORK GENERAL
    Replies: 17
    Last Post: 18th Mar 2011, 06:22 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •