People only ever talk about this formula for treadmills but fail to apply it to bores. Its the same principles and it works amazingly. Mirror finish every time!
Thank you for putting this out there! I’ve always thought that threadmills and bores weren’t running “correct” and had figured this out one day. All the old guys in the shop insist “federate is feedrate” but there’s a severe notable difference. I wish fusion accounted for this, especially on things like dynamic roughing. Dropping into a pocket and starting to rough out kicks out huge loads and breaks tools often. Video idea, what about what tap to use when? I can’t explain to our old school guys that there are aluminum taps, steels, stainless, etc. they keep throwing way wrong taps into way wrong material, like a green band osg spiral flute tap going at 1000 rpm into 304 stainless and then being confused on “what happened”. Especially with modern, fast, purpose driven tooling
You can use "feed optimization" in fusion and adjust the "reduced feed radius" to be slightly bigger than the tool radius and that will reduce the feedrate after the helical entry. On deep pockets, relative to the tool diameter, I set mine to 1.5 times the tool radius so the first couple of turns are done at the reduced feed rate, about 2/3 in my case. Now keep in mind that this will reduce the feedrate in other parts of the toolpath that fit the criteria.
Just got my Saunders fixture surface for my Shapeoko 3 and some of your vises. The threads are cleaner on your work than any other item I have ever owned.
Needed some threads in a 30mm hole and needed to make a plug that screws into it. Only had a 3/8 x 16 threadmill (1.587mm Thread Pitch) handy so I used your sheet to calculate the sizes. First try.. perfect. I don't mean OK, I mean perfect! Thank you!!!
Hi, I am a beginner in CNC machining and never (yet) milled thread myself, and I have no doubt that your (and others) experience shows that greatly reducing the feedrate leads to less tools breaking (especialy on less rigid machines). And initally the explanation seems to make a lot of sense. But I did some calculations and some drawing in a 2Dsketch in CAD and I have to say that the justification is actually wrong. The actual feed per tooth will never be greater (and most of the time it will be less) than the one specified for calculation. It is easier to explain with a drawing but the main point is that a chip is not uniform in thickness (conventionnal vs climb milling figures often show the chip shape very well), and even though the material effectively advances faster at the maximum hole diameter, the maximum chip thickness depends only on the tangential feed relative to the tool center's motion (which is the one calculated by Fusion). The chip thickness at the maximum diameter is always zero (cheap breakage if climb milling), exactly like in linear motion. That said, I may be more of a theoric person but if experienced people say that I am gonna break less tools with this formula, I am gonna use it. Even if the justification does not make sense.
I wouldn't cry if you just found a expression to put in on the hole ø, or something... so it doesn't matter what you put it... Juuuuuuuuust saying... Fusion can help us there..! edit: Never mind... i'm blind... you just did! Thanks !
I've been doing CNC as a hobbyist for almost a decade, now, and it never occurred to me that you need to adjust the chip load on tightly curved paths. Now that you point it out, it's obvious.
Any thoughts on a double lead thread milling operation? I have a part that requires it and we have always just tapped. Quality of finish usually leaves much to be desired.
Why aren’t you using in control compensation? Surely this would make the federate correct at the cutting edge as the toolpath coordinates would then be cutter edge not cutter centre? This also allows you to compensate for the correct thread fit.
Many of the hobby machines do not have that option and machines that do, can usually rigid tap. You are correct though, but one thing to add is that it can complicate few things including lead in/outs.
Nice. I don't make any toolpaths yet (been working in the shop since almost 2 months, 2/3 days a week, next to mech eng uni course) and we don't even use Fusion, but this is useful knowledge anyways
This can still apply to you big time though. Use this formula when using endmills to bore holes and it will give you PERFECT finishes and no more push off than you normally would get. Even more critical when using long tools. This formula is not unit dependent so as long as you keep your unit the same you're good.
no, thats the beauty of thread milling is that it can be done on just about any machine, unlike tapping which requires spindle encoders or special tool heads
Thank you for your dedication to the trade, you produce quality videos. Quick question: is there a central dispatch for cnc work bidding? Iam new to this business, thanks.
Why not just use g95 and program in ipr? Number of flutes times chip load. I have never understood programming in ipm. There really is no logical reason to program ipm. Ever.
People only ever talk about this formula for treadmills but fail to apply it to bores. Its the same principles and it works amazingly. Mirror finish every time!
Thank you for putting this out there! I’ve always thought that threadmills and bores weren’t running “correct” and had figured this out one day. All the old guys in the shop insist “federate is feedrate” but there’s a severe notable difference. I wish fusion accounted for this, especially on things like dynamic roughing. Dropping into a pocket and starting to rough out kicks out huge loads and breaks tools often.
Video idea, what about what tap to use when? I can’t explain to our old school guys that there are aluminum taps, steels, stainless, etc. they keep throwing way wrong taps into way wrong material, like a green band osg spiral flute tap going at 1000 rpm into 304 stainless and then being confused on “what happened”. Especially with modern, fast, purpose driven tooling
You can use "feed optimization" in fusion and adjust the "reduced feed radius" to be slightly bigger than the tool radius and that will reduce the feedrate after the helical entry.
On deep pockets, relative to the tool diameter, I set mine to 1.5 times the tool radius so the first couple of turns are done at the reduced feed rate, about 2/3 in my case.
Now keep in mind that this will reduce the feedrate in other parts of the toolpath that fit the criteria.
Just got my Saunders fixture surface for my Shapeoko 3 and some of your vises. The threads are cleaner on your work than any other item I have ever owned.
nice video. I like the semi-pro machine content. More accessible to a bigger audience.
Needed some threads in a 30mm hole and needed to make a plug that screws into it. Only had a 3/8 x 16 threadmill (1.587mm Thread Pitch) handy so I used your sheet to calculate the sizes.
First try.. perfect. I don't mean OK, I mean perfect!
Thank you!!!
When will NYC CNC be making the spreadsheet to cover metric?
I second this request
Hi, I am a beginner in CNC machining and never (yet) milled thread myself, and I have no doubt that your (and others) experience shows that greatly reducing the feedrate leads to less tools breaking (especialy on less rigid machines). And initally the explanation seems to make a lot of sense. But I did some calculations and some drawing in a 2Dsketch in CAD and I have to say that the justification is actually wrong. The actual feed per tooth will never be greater (and most of the time it will be less) than the one specified for calculation. It is easier to explain with a drawing but the main point is that a chip is not uniform in thickness (conventionnal vs climb milling figures often show the chip shape very well), and even though the material effectively advances faster at the maximum hole diameter, the maximum chip thickness depends only on the tangential feed relative to the tool center's motion (which is the one calculated by Fusion). The chip thickness at the maximum diameter is always zero (cheap breakage if climb milling), exactly like in linear motion.
That said, I may be more of a theoric person but if experienced people say that I am gonna break less tools with this formula, I am gonna use it. Even if the justification does not make sense.
I wouldn't cry if you just found a expression to put in on the hole ø, or something... so it doesn't matter what you put it...
Juuuuuuuuust saying... Fusion can help us there..!
edit:
Never mind... i'm blind... you just did! Thanks !
I've been doing CNC as a hobbyist for almost a decade, now, and it never occurred to me that you need to adjust the chip load on tightly curved paths. Now that you point it out, it's obvious.
MR-1 in your shop from Langmuir anytime soon?
I sure hope so
Any thoughts on a double lead thread milling operation? I have a part that requires it and we have always just tapped. Quality of finish usually leaves much to be desired.
Would you recommend a 3020 cnc for aluminum? Or would It be a waste of money
Why aren’t you using in control compensation? Surely this would make the federate correct at the cutting edge as the toolpath coordinates would then be cutter edge not cutter centre? This also allows you to compensate for the correct thread fit.
Many of the hobby machines do not have that option and machines that do, can usually rigid tap. You are correct though, but one thing to add is that it can complicate few things including lead in/outs.
how to do manual programming of thread milling ???.... if possible can you make video??
Titans of CNC did a video on that topic, ua-cam.com/video/9J4uJs3VPEo/v-deo.html
I would put the threads in with a form tap or a regular tap first then mill the end. Faster then thread milling and cheaper.
Nice. I don't make any toolpaths yet (been working in the shop since almost 2 months, 2/3 days a week, next to mech eng uni course) and we don't even use Fusion, but this is useful knowledge anyways
plus we do everything metric 😅
This can still apply to you big time though. Use this formula when using endmills to bore holes and it will give you PERFECT finishes and no more push off than you normally would get. Even more critical when using long tools.
This formula is not unit dependent so as long as you keep your unit the same you're good.
But you were at like 400 sfm, where did the 1200 go?
What happened to the recommended SFM? Why not kick up the rpm to 24000?
Do forming next!
Does the machine need a spindle encoder to do thread milling?
no, thats the beauty of thread milling is that it can be done on just about any machine, unlike tapping which requires spindle encoders or special tool heads
Thank you for your dedication to the trade, you produce quality videos.
Quick question: is there a central dispatch for cnc work bidding? Iam new to this business, thanks.
Why not just use g95 and program in ipr? Number of flutes times chip load. I have never understood programming in ipm. There really is no logical reason to program ipm. Ever.