Diagonal Hobbing

[coreywadeadams]

Multus u4000 with p300sa control.... it has it's own built in gear hobbing/skiving software but it is limited to only using a straight hob cutter as far as i can tell... u can make crowned gears using z and y axis feed motions just fine, but not "tapered" gears or splines using x and z axis feed motions. It has only one hob diameter input...

What im wanting to machine is a tapered root spline using a tapered hob cutter but im stuck and our okuma service personel has evidently given up on us. Have been in contact with them about this problem for months but no answers as of yet. I need to move the z and x simultaneously while maintaining the index sync ratio.

Im not a gear guy nor very fluent with okumas but have lots of cnc experience. The math involved in gear calculations make me cringe lol.

Does anybody know what im talking about?

Similar Threads:

• Diagonal movements
• Problema al cortar en diagonal
• MB3-screen goes diagonal sometimes
• Newbie- Mach3 shooting off in diagonal

[coreywadeadams]

A few pics... I'll get the part specs posted asap if anyone replies

[Jim Dawson]

Looks like a really good job for a horizontal mill with a 4th axis.

[deadlykitten]

hello corey

u can make crowned gears using z and y axis feed motions just fine

please, can you share a photo with such a part and the tool used ? i dont know exactly what the Y feed is doing

What im wanting to machine is a tapered root spline using a tapered hob cutter

you mean the attached part ( from your post ) ? also, if possible, please share a photo with your tool, and the drawing of the part section, or even better, 2 drawings of your part section, located at a certain axial distance ( eq 10 mm, etc )

without knowing all the data, this is what i think of :
... simple milling operation, one tooth at a time :
...... at least 3 C axis values for each tooth
...... a single C axis values for each tooth
... broaching with custom tool
... using a hob : requires S+M+Z+X sync ( like you also wish )

kindly

[deadlykitten]

please excuse my curiosity : i put 2 questions in attached images

[broby]

please excuse my curiosity : i put 2 questions in attached images

Yep this is a Gear/Spline HOB cutter.

[deadlykitten]

hello, taking a closer look at your photos :
... image 1 : seems that the part may require special profile among the tooth flanc
... image 2 : wavy surface

a hob operation is fast, but it requires a holder with stacked axis ( = randament lost; better use 2:1 ratio ), and near the operation end, the cutting effort increases

it may be possible to use a set of tools ( roughing and finishing ); the finishing tool has custom profile, and it can be like an end-mill, or T-shaped; in both cases, this tool would be driven directly by the M spindle, so no more stacked axis
... cons : cycle time may increase
... pros : setup cost will lower, required machine effort will lower, surface quality will improve, finishing tool will cut less material

kindly


ps : my bad, on a multus you may not need a stacked holder; however, check also this video, which shows what i was trying to suggest : cutting is smooth and no hobs are used :

[coreywadeadams]

We're going to machine this part complete on the multus using the tapered hob in the first batch of pictures... note that major dia."of the workpiece" Is straight and only the root is tapered. The grooves get wider as cut gets more shallow. This is generated with the tapered hob cutter as it feeds in both x and z as it rotates in sync..

[coreywadeadams]

As promised....

[deadlykitten]

my god, now i see that flat turn is an option, and available ratios are 1:1, 1:2 - 1:6

but there is also an M238, and i dont know how to go cw-cw, ccw-cw, cw-ccw, ccw-ccw, because SB can not be issued during flat turn ( hmm, M238 may reverse the M spindle )

i must only fix the ratio problem; mr Wizard, someone, please, a bit of help

i am sorry correy, i will do my best to help you; kindly !

[deadlykitten]

mr corey, i was googling to better understand what is a tapered hob and how it works; i have mistaken : in image 1 of post 5 is not a part, but your tool pfff

i guess you need feed among Y and not among X ( pls check attached image )

i would input Y feed inside the code for a normal hob, and maybe use X axis only for rapids when positioning, so to reduce stress on tool

also check this :

[coreywadeadams]

Looks like a really good job for a horizontal mill with a 4th axis.

We have a ma500 space center but it has tombstones on it. I could replicate the same setup on the multus and 3d the damn thing but it's such a waste of time when ive got the $1000 hob that will do it in a single pass. The machine is capable, i know it is, but the included software to make the code is not capable of tapered root splines. I'm going to post the code that it gives me for this part. It is good code and it makes a very nice "straight","flat root" spline. I have also tried editing the code by adding in an "x" move when it feeds across in "z", but this resulted in grinding all the spline teeth off the part!! lol

[deadlykitten]

hello again, please check attached image long story short : sync zy

the time required for real trials on that multus-u4000 is much shorter than the time that i need to fully understand this kind of process and develop sharp code

if the hob is spinned directly by M axis ( like and end-mill ) , i recomand using the hob in the area near the tool shank, for better cutting rigidity

when wear appears, simply move towards the top ( maybe also reduce cutting specs ), so a fresh row of teeth will be delivering the finished surface

i may be wrong, since flipping the tool requires reversed sense of the M axis, and it may possible that a new tool could not be used towards the shank, but only towards the top, so a bit of rigidity will be lost, but it may be just fine; kindly

[coreywadeadams]

hello again, please check attached image long story short : sync zy

the time required for real trials on that multus-u4000 is much shorter than the time that i need to fully understand this kind of process and develop sharp code

if the hob is spinned directly by M axis ( like and end-mill ) , i recomand using the hob in the area near the tool shank, for better cutting rigidity

when wear appears, simply move towards the top ( maybe also reduce cutting specs ), so a fresh row of teeth will be delivering the finished surface

i may be wrong, since flipping the tool requires reversed sense of the M axis, and it may possible that a new tool could not be used towards the shank, but only towards the top, so a bit of rigidity will be lost, but it may be just fine; kindly

On this machine "X" is up and down... "Y" is front to back, z is left to right.

And what u said about moving to a new row of teeth when it becomes dull is exactly what i want to do. But i dont want to shift it occasionally, i want it to shift and use the entire length of the tool every pass/part. Thats why its called diagonal hobbing. It increases tool life, uses all the teeth on cutter, and its also the method for creating "taper root splines" by means of a conical hob tool.

Heres the unedited code straight off the multus using the conversational gear software.

[coreywadeadams]

Note that the name "taper spline" does not reflect the actual code. Its what i named it, the code produces no taper whatsoever.

I realize this is a difficult mess to wrap your head around. Even when its right in front of u.. its kept me up at night and been on my mind constantly bc challenges like this is why i chose the trade... I will get this done or scrap every part trying.
Im not currently setup on this job but it will be back aroundin the near future. The guys that are doing the spline for us are 600 miles away and really sticking it to us on an order of these, from what im told anyway.

[deadlykitten]

On this machine "X" is up and down... "Y" is front to back, z is left to right

hello corey i think i know how this axis are named; a normal lathe has XZ, and Y is an option

moving to a new row of teeth when it becomes dull is exactly what i want to do. But i dont want to shift it occasionally, i want it to shift and use the entire length of the tool every pass/part.

this is how i see things : tool has active length of 100%; when tool cuts, it uses only 10% of its active length, and this 10% is moving between 0% and 60% ( or 60%-0% ) as the turret is moving to left

after a while, wear occurs, so is no longer possible to use 0-60%, but [0+5]-[60+5]%, where 5% represents the portion of a new row of teeth

i dont know exactly if you must go from bigger diameter to the smaller diameter among the tapered tool, or viceversa, but tool requires feed among Y among a certain portion, and this portion may be shifted as wear occurs i hope that i explained it well

*all numbers used are theoreticall

i want it to shift and use the entire length of the tool every pass/part

i dont think that you should wish for this / if the process requires a tool with an active length AL=n, than the real tool must have AL=n+2*clearance, so to have a bit of comfort

the longer the tool, more comfort created, but also rigidity is lost

Heres the unedited code straight off the multus using the conversational gear software

i have not yet open that file; all i need is just the code outputed for a simple ( cilindrical ) hob, for a single Z pass

after that, we can develop

if the code is created for this tapered hob, but it does not perform well, than onestly, i dont think that it is a good idea to remove errors and add good code; i simply wish to develop from a good working code

also, i am not sure if i will succed, but i will try

its kept me up at night and been on my mind constantly

there are better ways to have white nights

Note that the name "taper spline" does not reflect the actual code. Its what i named it ...

i am really waiting for the day when i will simply create a file with a particular name, without any content at all, and the cnc will machine parts

i would call it : just_get_it_done_and_clean_yourself_and_do_whateve r_without_letting_me_know.go_now

[coreywadeadams]

; all i need is just the code outputed for a simple ( cilindrical ) hob, for a single Z pass

thanks for your time! Much appreciated...
Yes that code in the file is for a simple hob with one z pass. That is all the machine will output.

However, i "think" in could be edited to feed in both x & z at the same time... on this particular machine and part, y is only used for an approach or retract, the variation in tooth depth will and must come from the big to small diameter change of the cutter as it travels to the left in z and upwards in "x". Y is to remain fixed other than lead in/out of the cut.

The cutter has extra markings on it from a normal hob. Has 2 diameters and an extra "L"...

[deadlykitten]

good day corey you desire for extra movement on X, while i thougth of Y

if hob position, relative to M spindle axis, is :
... coaxial ( just like an endmill ), than X is required ( i guess this is your real setup )
... perpendicular ( just like when using an attachement ), than Y is required ( this is what i have imagined : hob suport from both ends )

i will look soon over your code / kindly

[deadlykitten]

... coaxial ( just like an endmill ), than X is required ( i guess this is your real setup )

if this is your case, and you have allready feed X axis ( and it wiped the teeth ), than it simply means that it was a wrong ratio

i dont know how to compute the ratio between the 2 linear feeds, but last video that i shared show the process that you need, so maybe the guys that posted it can provide the answer

[deadlykitten]

hello after my own math, which i can not test it, ( pls find attached draw ), i would recomand a 2nd feed travel of 64.776mm, coresponding to a 1st feed travel of 63.094mm(=2.484inch) so to achieve a part with 3.279 degress taper, using a hob with 3.199 degress taper

what is the length of your hob ? is it at least 65mm ?

at this stage, i would measure the conicity of the hob ( because this approach is based on that value ), compute a 2nd feed travel, and do a part

i still have no clue about :
... the theory about how to compute the exact ratio
... how the cnc syncs M&S ( initialty i thought it is flat turn, but now i doubt it )