CSS Formula

[CarbideBob]

When reading this litany of posts (which method is "better") remember that the actual cutting speed for any material is really a best guess as to what speed won't burn up the tool, really nothing more. Years ago Valenite published a cutting speed chart we had posted in our shop with SFM values for mild steel varying from 250 to 2500sfm depending on the depth of cut and feedrate and also for maximum rate of production and maximum tool life.
So what is the point of calculating rpm to a hundred decimal places anyway?

Yup 4.0 works just as good as 3.82 and is easier.
The base problem is that people select some nice round number for SFM and have no idea if it is the best.
What makes 500SFM right for an application? Why not 482 or 515?
If your starting number a guess super accurate math is simply a waste of time.

If I'm cutting a 3 inch part at .050 deep which surface footage? Tool tip or out on the flank of the tool at stock diameter?
These are two different numbers in the one cut.

 

[EmGo]

Yup 4.0 works just as good as 3.82 and is easier

Not restricting yourself to some stupid formula is easiest yet

It really is just a trivial division problem. The distance you want to go in a minute divided by the distance around the part equals the number of times you have to turn the part per minute to get there.

If a mile is 1,760 steps and your step length is 3 feet, how many feet in a mile ? How many steps per minute do you need to walk to get there in time for dinner ? It's that simple.

The base problem is that people select some nice round number for SFM and have no idea if it is the best.
What makes 500SFM right for an application? Why not 482 or 515?
If your starting number a guess super accurate math is simply a waste of time.

If I'm cutting a 3 inch part at .050 deep which surface footage? Tool tip or out on the flank of the tool at stock diameter?
These are two different numbers in the one cut.

Experience. Everyone operates a little differently. For me, the way I run a lathe and the inserts and steel I generally use, 400 fpm and hefty cuts work best. And that's at the tip, for sure not worth being concerned about whether it's 400 or 415 or 387. Material hardness, chip breaking, all that will influence what you want to actually be at for any particular run. That's what the speed override dial is for

 

[gcodeguy]

Experience. Everyone operates a little differently. For me, the way I run a lathe and the inserts and steel I generally use, 400 fpm and hefty cuts work best. And that's at the tip, for sure not worth being concerned about whether it's 400 or 415 or 387. Material hardness, chip breaking, all that will influence what you want to actually be at for any particular run. That's what the speed override dial is for

What steel, how hefty and at what feed rates? I ask because I run SFM a lot faster than that. Are you saying if SFM is too fast for roughing, that you dial down the override for all operations? Or stand there and just dial it down for the roughing operation(s) only?

Seems to me it would be financially beneficial to modify the roughing operation(s) SFM and leave the override dials at 100% for all operations. Where I work, cycle time is king even when slowing down a bit results in more parts made at the end of the day. Personally don't understand that thought process, but I am just a lowly peon so what do I know.

Most of our lathes are barfeeds. Can't run too heavy of a cut or at too fast a feed rate (unless quite shallow DOC) or material pushes back. Especially when making several cuts to a shoulder. One cut isn't bad, but several cuts leave a staircase.

I'm confused why this thread has resulted in such a flurry over how to and what formulas to use. Don't most people use a Machinist Calc Pro to figure RPM these days? And if not that calculator, then Pi is on every scientific calculator I've ever used. All this arguing seems pointless.

I admit that I figure RPM to the nearest whole number, but is that really necessary (other than to satisfy me)? (I'm a little anal with my programming.) If you are using a G96, then the control is going to change its RPM based on the X-move programmed before the G96. How bad can it be if you figured 50 RPM off on the initial spindle speed? The G96 will correct that mistake.

 

[angelw]

Not restricting yourself to some stupid formula is easiest yet

It really is just a trivial division problem. The distance you want to go in a minute divided by the distance around the part equals the number of times you have to turn the part per minute to get there.

A formula is just a way of conveying the method to someone. When you multiply the diameter by 22 and divide by 7 to get the circumference, then convert the FPM to IPM by multiplying by 12 and finally dividing the IPM by the circumference, you're using a formula. RPM = (FPM x 12) / (D x 22 / 7) is your formula; you're just expressing it in words

Bob's reference to 4 is from the simplified formula RPM = FPM x 4 / D. Expressed in words it would be: FPM times 4, divided by the diameter. FPM x 4 / D can just as easily be done in your head, easier really, as (FPM x 12) / (D x 22 / 7)

 

[sinha]

A formula is just a way of conveying the method to someone. When you multiply the diameter by 22 and divide by 7 to get the circumference, then convert the FPM to IPM by multiplying by 12 and finally dividing the IPM by the circumference, you're using a formula. RPM = (FPM x 12) / (D x 22 / 7) is your formula; you're just expressing it in words

Bob's reference to 4 is from the simplified formula RPM = FPM x 4 / D. Expressed in works it would be: FPM times 4, divided by the diameter. FPM x 4 / D can just as easily be done in your head, easier really, as (FPM x 12) / (D x 22 / 7)

I have stopped arguing with him

 

[goooose]

I have stopped arguing with him

I'm with you on this one.
People like him, even when they realize they're 'wrong', they are too stubborn to admit it and will fight you on it till the end of time. Better off sharing your knowledge with someone who is happy to receive it.

 

[sinha]

I'm with you on this one.
People like him, even when they realize they're 'wrong', they are too stubborn to admit it and will fight you on it till the end of time. Better off sharing your knowledge with someone who is happy to receive it.

The worst part is that even his language is rough, using words like stupid, nonsense etc.

 

[CarbideBob]

How does one know the proper or best SFM to program at?
If facing should not the SFM change higher as the front heel clear changes with diameter with a given tool?
(the angle is the same in CAD, front heel clearance in the cut is not)
Or is chip velocity across the tool face a bigger concern for tool life?
The catalogs and insert packages give a range but often this range is huge.
How do you pick a number? Do you pick a nice round number?

I have no problem at all with EG. I have been called so stupid many, many times. If he dickslaps me there is likely a good reason.
In my book he has many chips in his shoes and ignore that at your own peril.

For sure we need good cad jockeys here.
Sinha and anglew know a huge amount of things about the cnc side.
If the talk is tough from others I do have to wonder about life on the floor where we do make millions of parts.

Many points of view. How is your's brought about? Why are others 180?
Bob

 

[EmGo]

A formula is just a way of conveying the method to someone.

I'm obviously not making myself clear. In fact I left this rathole because jaguar36 is a dishonest piece of shit but it's 4:00 a.m. and I'm sick of being locked up, even tired of watching hot girls do exercise routines on wechat but the wifi is good ...

No, what I do is not a formula. All I do is take the desired speed in distance over time, divide by distance over one revolution (same as multiply by the inverse) to get revolutions over time. If this is a formula, then so is filling your water glass 3/4 full at the sink. Or taking a recipe for an 8" pie and multiplying it by relative volumes for a 9" crust. It's just simple logic.

Why care ? In fact, people are welcome to do whatever they like. They don't actually have to know a damn thing about machine tools these days to make parts. They can just look up a recipe in sinha's book and now they are a machinist.

However there are also other ways. I was "inspired" by the fact that this thread got to be pages long with ridiculous descriptions of a simple problem taking up a whole page from a certain party who couldn't even turn on a damn lathe. Then along comes a guy who thinks that 22/7 is a troll and throwback to the ancient Egyptians, when it comes out with the exact same answer as his "accurate method" and better than his quick and dirty. But, you know, he's cool and everyone else is a troll or fool or "wrong".

My position is simple. On a mill there's three axes of motion and complicated contours and all the hassles of programming by hand justify crappy programs. Just watch a complicated program some day and the 20% air time will have an actual machinist cringing. But it's worth it because the programming is too complex to do in a reasonable amount of time by hand.

Lathes are not generally like that. The op's example should have been answered in three posts. "Take your desired speed, divide it by the distance per rev, and that's your answer." Simple. Instead we got pages of fanuc-reasoning.

If people want to use canned cycles and cookbook recipes for blowing their nose, it's up to them. But there is a better way. Sinha's way sucks if you've ever actually run a lathe. Same part, let him program it and me program it and my parts will be better and come out faster and be easier on the machine. For two or three pieces, fine, do it the quick and shitty way. But if you make 50 or more, you're losing time and quality. There is a better way.

 

[sinha]

At least the language has improved.

 

[goooose]

Can't wait till we get to see how he figures out feedrate

 

[EmGo]

Can't wait till we get to see how he figures out feedrate

That one's easy. You figure out how many lineal inches of roughing you have, then how long it takes for a shit. Multiply the inches by the reciprocal of your shitting time and that's your G93 inverse time feedrate.

 

[guythatbrews]

Use what works for you.

CNC manually programmed or CAM post processed all my programs use the RPM=3.82*SFM/dia explicitly expressed in the program. I let the CNC calculate the RPM. Max RPM aside, I don't care about RPM and want to see SFM in the program.

When standing in front of a manual machine I recall 100 rpm is close enough to 100 SFM on a 4" diameter and mental math from there.

 

[Rogue_Machinist]

Wow didnt expect to open this can or worms. But yeah I got it figured out and had to manipulate a few numbers. But got it working. Dang. LOL.

 

[CarbideBob]

Can't wait till we get to see how he figures out feedrate

Effective chip thickness, the thing that counts most, is almost never feed per rev.
How do you do it?
Tool radius, lead angle, ... Here is where cad really helps.
Think about doing such in the days before cad. Nightmare math on the drawing board with pencil and paper.

I do most certainly remember the days of thinking that this was simple-simple so can give no fault of anyone here.
Bob

 

[CarbideBob]

Wow didnt expect to open this can or worms. But yeah I got it figured out and had to manipulate a few numbers. But got it working. Dang. LOL.

Maybe opening the can of worms is a good thing.....Even if some "feelings" get hurt a tiny bit.
I so wish I could master this stuff but there is always something that should and does not work.
I fail a lot of times on speeds/feeds and I am not shy to say so.
Bob

 

[Booze Daily]

Push it til it breaks, then back off a little.

 

[angelw]

Use what works for you.

Exactly. Who gives a toss how you get there, just so long as what you use works. Goldstein seems to be the only one making an issue out of something that is simple, whichever method that's used. It's either his way or the highway.

When standing in front of a manual machine I recall 100 rpm is close enough to 100 SFM on a 4" diameter and mental math from there.

and do you know why that combination works? Its because the closest Integer is "4" when 12 (used to multiply SFM to convert to SIM), is divided by Pi and is how the simplified algorithm (better not use the word Formula) RPM = SFM x 4 / Diameter, is derived.

Regards,

Bill