Acceptable machining times

tex707 · 03-03-2004, 05:22 PM

What do you consider as an acceptable machining time for a waterblock? A few months ago I have been criticized…

…about a design of a hard-to-be-machined waterblock on this forum. I’ve just made some calculations…it takes around 13 minutes to machine the center (and most complicated) section of the waterblock from the screenshot attached with a .0394” flat endmill. Comments?

Cathar · 03-03-2004, 05:32 PM

It all depends on economics as they apply to your situation. How much does machine time cost for you? How much do you want to sell your product for?

One block I'm making has about 2hrs of total CNC machine time, with 1.5hrs of that CNC time spent on a single plate. I don't intend to mass-produce the blocks though.

tex707 · 03-03-2004, 05:38 PM

Quote:

Originally Posted by Cathar

It all depends on economics as they apply to your situation. How much does machine time cost for you? How much do you want to sell your product for?

One block I'm making has about 2hrs of total CNC machine time, with 1.5hrs of that CNC time spent on a single plate. I don't intend to mass-produce the blocks though.

Actually, I'm not talking about commercial production, but prototypes only...

I've still not designed the block that takes 2hrs total to be machined yet...

...will try, though....

JFettig · 03-03-2004, 06:13 PM

Typical machine time costs about a dollar a minute plus programing and setup. What are you basing your machine time on? what kind of feeds and depths of cuts?

There is no way that would take 13 minutes. Im thinking close to 1hr with a 1mm EM

Jon

tex707 · 03-03-2004, 06:37 PM

Quote:

Originally Posted by JFettig

Typical machine time costs about a dollar a minute plus programing and setup. What are you basing your machine time on? what kind of feeds and depths of cuts?

There is no way that would take 13 minutes. Im thinking close to 1hr with a 1mm EM

Jon

I am not quite sure what you mean by programming, but if one provides a correct G-code for a given CNC machine, I suppose that there is no much programming left to be done, right?

Feedrates and depths...OK...depths are less than .02" (since the endmill is .0394”), .0197" to be exact. Feedrates are 40, 30, 80 and 20 (for approach, machining, retract and finishing respectively) with 50 slowdown rate and 70 feedrate reduction in corners...spindle output is set to 6000 RPM...

CATIA and Mastrecam simulation shows almost identical time of 13 minutes so I consider this close to real life....or I am very wrong about this…

-J- · 03-03-2004, 06:59 PM

just an idea, but i think that the most cost efective way of doing such a block, would be using a simple rotor method for the center of the block, and just milling a channel around it.

the pins wont be hexagonal, but it will be far more simpler and i dont know but ill guess that more efficient too.

JFettig · 03-03-2004, 06:59 PM

by programming I mean making the G-code, For some machines it would need code for toolchanges and whatnots and woudl probably have to be reprogrammed.

The depths sound right but the feed rates are about 10x too high. I would say around 3-6IPM for a 6000rpm spindle. I still have to experiment, but thats about as fast as youlll get. The rapids on my machine are about that fast, theres no way I could cut something with any tool using rapids

Jon

Cathar · 03-03-2004, 07:46 PM

Quote:

Originally Posted by tex707

Actually, I'm not talking about commercial production, but prototypes only...

I've still not designed the block that takes 2hrs total to be machined yet...

...will try, though....

Not sure what you're asking then. Acceptable time is therefore dependent on whatever you're prepared to pay to get it done.

Machininist are making up that long-time block tomorrow morning. Over 1M lines of G-code.

JFettig · 03-03-2004, 10:05 PM

holy bawls Cathar, over 1million lines of G-code! the longest I have done is around 500 lines

I supose it could be that I mostly hand wrote mine, but still! wow! your gonna post pictures right?

Jon

Cathar · 03-03-2004, 10:21 PM

Quote:

Originally Posted by JFettig

holy bawls Cathar, over 1million lines of G-code! the longest I have done is around 500 lines

I supose it could be that I mostly hand wrote mine, but still! wow! your gonna post pictures right?

149 jet tubes that are each individually carved out as proper circular cylinders will do it...

Yeah - will post pictures.

tex707 · 03-17-2004, 04:48 PM

Quote:

Originally Posted by JFettig

by programming I mean making the G-code, For some machines it would need code for toolchanges and whatnots and woudl probably have to be reprogrammed.

The depths sound right but the feed rates are about 10x too high. I would say around 3-6IPM for a 6000rpm spindle. I still have to experiment, but thats about as fast as youlll get. The rapids on my machine are about that fast, theres no way I could cut something with any tool using rapids

Jon

It seems that you are right...and I see no reason not to believe your experience...

Anyway...I have this small HAS Speeds/Feedrates calculator....only variable there is a Surface Speed FPM...unfortunately, there is no recommended value for copper. What do you suggest?

jaydee · 03-17-2004, 08:19 PM

Quote:

Originally Posted by tex707

It seems that you are right...and I see no reason not to believe your experience...

Anyway...I have this small HAS Speeds/Feedrates calculator....only variable there is a Surface Speed FPM...unfortunately, there is no recommended value for copper. What do you suggest?

I suggest you find someone with some hands on experience with the particlular mill and the particular material. Charts are not going to be to accurate. The real test is running one out on the machine it is to be done on. The real problem I see though is that block will not perform as well as it should for the cost it will take to make it (Which is the real reason no one has tried duplicating the Cascade).

What would make that block more economical and easier to make would be to Rotor style it. Drill holes and connect. A good mill should be able to pop them holes in in seconds each and then a few minutes cross connecting holes.

Thats what I did here:

Drilling the holes with a 1/8" drill bit.

----------------
Milling the ends for water to merge/spread before entering and exiting through the barbs.

----------------
milling the O-ring groove.

If I were to mill the pins it would have taken 5 times as long. It took about 45 minutes on my tiny mill to make that base.

tex707 · 03-18-2004, 01:18 AM

Quote:

Originally Posted by jaydee116

I suggest you find someone with some hands on experience with the particlular mill and the particular material. Charts are not going to be to accurate. The real test is running one out on the machine it is to be done on. The real problem I see though is that block will not perform as well as it should for the cost it will take to make it (Which is the real reason no one has tried duplicating the Cascade).

What would make that block more economical and easier to make would be to Rotor style it. Drill holes and connect. A good mill should be able to pop them holes in in seconds each and then a few minutes cross connecting holes.

If I were to mill the pins it would have taken 5 times as long. It took about 45 minutes on my tiny mill to make that base.

Thank you for the response....I see that you are referring to the CAD model of the block I've attached and you are definitely right. However, I was thinking about machining copper in general...what surface speed would be recommended...I see 300FPM for brass and 75 for plain steel.

Are you machining the O-ring groove in a single pass only?

jaydee · 03-18-2004, 09:28 AM

Quote:

Originally Posted by tex707

Thank you for the response....I see that you are referring to the CAD model of the block I've attached and you are definitely right. However, I was thinking about machining copper in general...what surface speed would be recommended...I see 300FPM for brass and 75 for plain steel.

Are you machining the O-ring groove in a single pass only?

Brass mills like aluminum (butter

). Thats what brass is made for easy milling with the stength of copper but it looses coppers good thermal properties as a trade off.

300FPM? That dosn't sound right for any material. I don't even think a laser can move that fast. Thats the length of a foot ball feild in 1 minute! Providing FPM stands for Feet Per Minute. Maybe 300IPM (Inch Per Minute). Thats is still 25FPM though.

Something is wrong with whatever chart your using. Machines just don't move that fast even when not cutting anything?

EDIT: And yes, 1 pass for the Oring groove. It is only 1/16" deep.

tex707 · 03-18-2004, 11:00 AM

Quote:

Originally Posted by jaydee116

Brass mills like aluminum (butter

). Thats what brass is made for easy milling with the stength of copper but it looses coppers good thermal properties as a trade off.

300FPM? That dosn't sound right for any material. I don't even think a laser can move that fast. Thats the length of a foot ball feild in 1 minute! Providing FPM stands for Feet Per Minute. Maybe 300IPM (Inch Per Minute). Thats is still 25FPM though.

Something is wrong with whatever chart your using. Machines just don't move that fast even when not cutting anything?

EDIT: And yes, 1 pass for the Oring groove. It is only 1/16" deep.

I know that brass is perfect for milling, that's the reason I've mentioned the 300FPM surface speed. I think that you are misunderstanding this...it is not the feedrate, but machining surface speed...please see the attachment.

jaydee · 03-18-2004, 11:07 AM

Quote:

Originally Posted by tex707

I know that brass is perfect for milling, that's the reason I've mentioned the 300FPM surface speed. I think that you are misunderstanding this...it is not the feedrate, but machining surface speed...please see the attachment.

Yup, I am lost. Non of that looks right to me.

JFettig can probably have a better answer.

JFettig · 03-18-2004, 04:23 PM

Well, actually I dont usually use fpm, are you sure it isnt sfm? surface feet per minute, that would be kinda like the area of the cutout. at least I think thats what it is, it has to do with the half of the tool that would be doing the cutting in. Thats some stuff I wish I knew and would be interested in learining.

Jon

tex707 · 03-18-2004, 04:34 PM

Quote:

Originally Posted by JFettig

Well, actually I dont usually use fpm, are you sure it isnt sfm? surface feet per minute, that would be kinda like the area of the cutout. at least I think thats what it is, it has to do with the half of the tool that would be doing the cutting in. Thats some stuff I wish I knew and would be interested in learining.

Jon

It is FPM...please take a look at the screenshot...

I wish I knew that too....as the matter of fact, I used to, but I had that course some 20 years ago....

prb123 · 03-18-2004, 04:40 PM

It states 55 IPM for the Feedrate. I have no idea what Surface Speed FPM stands for.

Butcher · 03-18-2004, 05:28 PM

300 FPM isn't that fast for a laser jaydee, it's only 3.4 mph.

It;s damn fast for any sort of tool though.

pauldenton · 03-18-2004, 05:30 PM

Quote:

Originally Posted by prb123

It states 55 IPM for the Feedrate. I have no idea what Surface Speed FPM stands for.

i believe it is just an (older?) terminology and is the surface speed in (surface) feet per minute...

that attachment shows a 1/2" diameter tool at 4583.7rpm...
using rpm=cutting speedx4/diameter of cutter, that's a cutting speed of 573 feet per minute....

tex707 · 03-18-2004, 05:41 PM

Quote:

Originally Posted by pauldenton

i believe it is just an (older?) terminology and is the surface speed in (surface) feet per minute...

that attachment shows a 1/2" diameter tool at 4583.7rpm...
using rpm=cutting speedx4/diameter of cutter, that's a cutting speed of 573 feet per minute....

OK, than....600 FPM is a rounded figure...

superart · 03-18-2004, 06:52 PM

Tex77, r/u using a blue PROlight mill? That setup screen looks vaugly farmiliar.

tex707 · 03-19-2004, 01:33 AM

Quote:

Originally Posted by superart

Tex77, r/u using a blue PROlight mill? That setup screen looks vaugly farmiliar.

I'm not quite sure what you mean....if you are referring to the calculator, it is a HAAS CNC Quick Code 4.0.2....the W/B model attached at the start of the thread is made with CATIA V5.

pauldenton · 03-19-2004, 04:29 PM

Quote:

Originally Posted by tex707

OK, than....600 FPM is a rounded figure...

actually it's exact i think - the 573 is approximate (based on Pi=3 i think....)

03-18-2004, 05:28 PM	#20
Butcher Thermophile Join Date: May 2001 Location: UK Posts: 1,064	300 FPM isn't that fast for a laser jaydee, it's only 3.4 mph. It;s damn fast for any sort of tool though. __________________ Once upon a time, in a land far far away...

03-18-2004, 06:52 PM	#23
superart Cooling Savant Join Date: Mar 2002 Location: FL Posts: 787	Tex77, r/u using a blue PROlight mill? That setup screen looks vaugly farmiliar. __________________ When you do things right, people won't be sure youv'e done anything at all. Looking to buy/trade for used Deep Fryer and Vacume Pack Sealer.

03-03-2004, 05:32 PM	#2
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	It all depends on economics as they apply to your situation. How much does machine time cost for you? How much do you want to sell your product for? One block I'm making has about 2hrs of total CNC machine time, with 1.5hrs of that CNC time spent on a single plate. I don't intend to mass-produce the blocks though.

03-03-2004, 06:13 PM	#4
JFettig Cooling Savant Join Date: Feb 2003 Location: Willmar MN/Fargo ND Posts: 504	Typical machine time costs about a dollar a minute plus programing and setup. What are you basing your machine time on? what kind of feeds and depths of cuts? There is no way that would take 13 minutes. Im thinking close to 1hr with a 1mm EM Jon

03-03-2004, 06:59 PM	#6
-J- Registered User Join Date: Dec 2002 Location: Buenos Aires, Argentina Posts: 217	just an idea, but i think that the most cost efective way of doing such a block, would be using a simple rotor method for the center of the block, and just milling a channel around it. the pins wont be hexagonal, but it will be far more simpler and i dont know but ill guess that more efficient too.

03-03-2004, 10:05 PM	#9
JFettig Cooling Savant Join Date: Feb 2003 Location: Willmar MN/Fargo ND Posts: 504	holy bawls Cathar, over 1million lines of G-code! the longest I have done is around 500 lines I supose it could be that I mostly hand wrote mine, but still! wow! your gonna post pictures right? Jon

03-18-2004, 04:23 PM	#17
JFettig Cooling Savant Join Date: Feb 2003 Location: Willmar MN/Fargo ND Posts: 504	Well, actually I dont usually use fpm, are you sure it isnt sfm? surface feet per minute, that would be kinda like the area of the cutout. at least I think thats what it is, it has to do with the half of the tool that would be doing the cutting in. Thats some stuff I wish I knew and would be interested in learining. Jon

03-18-2004, 04:40 PM	#19
prb123 Cooling Neophyte Join Date: May 2002 Location: Michigan Posts: 51	It states 55 IPM for the Feedrate. I have no idea what Surface Speed FPM stands for.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)