Struggling with an awful o-ring groove

[JB7280]

6 minutes ago, crazy^millman said:

That is pretty much a given now a days. I am referring to the graduated folks with degrees that are worthless if this the kind of things they are not learning about manufacturing or willing to go investigate on their own. No one taught us this we applied some common sense and figured out that is I want an insert for a certain size bolt I need the extra room for that insert to fit into. Basic common sense to know this and design a part to be manufactured and not know that is just ignorant at levels it just blows my mind. You call out a seal what is the first thing a designer is supposed to do? Go look at the manufacturing specifications to machine the feature to work correctly in. Certain seals have certain tolerance requirements for machining the features they will work in. A Keensert is no different the designer needs look up what are the manufacturing specification for putting that into your design. Just lazy and entitled is what that screams to me.

Management in most companies not getting this I just roll my eyes and do what I need to get the work done. There is no excuse for any designer today to not know DFM and to get a degree some of these designers get and not be taught this is a disgrace to them and the places where they were taught.

Ron, I don't mean to get you wound up any more, but I should point out that this customer supposedly has over 900 engineers with PHD's!!  I have a friend and former coworker who is a mechanical engineer, who basically demanded he be able to spend time working in the shop, running parts, setting up machines, and went on to do some programming as well.  Because of that, I think he's one of the most talented engineers I've met.  We no longer work together, but he still asks me quite often "how much more difficult will it be to machine this part if I put this tolerance on it".  Or, "what changes would make this feature easier to machine".

 

And he has always said anything above a master degree in an engineering field is a waste.  There's a point when it's a bigger benefit to get out of the classroom, and into the real world.  

[crazy^millman]

11 minutes ago, JB7280 said:

Ron, I don't mean to get you wound up any more, but I should point out that this customer supposedly has over 900 engineers with PHD's!!  I have a friend and former coworker who is a mechanical engineer, who basically demanded he be able to spend time working in the shop, running parts, setting up machines, and went on to do some programming as well.  Because of that, I think he's one of the most talented engineers I've met.  We no longer work together, but he still asks me quite often "how much more difficult will it be to machine this part if I put this tolerance on it".  Or, "what changes would make this feature easier to machine".

 

And he has always said anything above a master degree in an engineering field is a waste.  There's a point when it's a bigger benefit to get out of the classroom, and into the real world.  

I am not wound up just what I see day in and day out and people keep complaining about the skills gap, but it is really an education gap. Your friend sounds some of the best engineers I have meet also.

[JB7280]

2 minutes ago, crazy^millman said:

I am not wound up just what I see day in and day out and people keep complaining about the skills gap, but it is really an education gap. Your friend sounds some of the best engineers I have meet also.

Oh I know you're not actually wound up.  I was just making a joke there.  This group of engineers is by far the most frustrating I've dealt with in that their first designs have been atrocious.  Half of these features that we've fought, and struggled with, we ask them if they can change it and they say "Oh, yea, sure, no problem."  THEN WHY WASNT IT THAT TO BEGIN WITH?!?!!?

[crazy^millman]

5 hours ago, JB7280 said:

Oh I know you're not actually wound up.  I was just making a joke there.  This group of engineers is by far the most frustrating I've dealt with in that their first designs have been atrocious.  Half of these features that we've fought, and struggled with, we ask them if they can change it and they say "Oh, yea, sure, no problem."  THEN WHY WASNT IT THAT TO BEGIN WITH?!?!!?

Well you're love this. Helping a company with parts. Every part that has gone on a tags for issues that inspection finds have been approved to be used as it for 71% and have not failed in testing or use. Every part that goes on table made by the machinist goes against them. Doesn't matter how much the machinist tells them it can't be done derangement doesn't want to hear it and says do it anyway. They make what they can make and then it gets tagged by inspection because it was .001 microns out of tolerance. That deviation goes on their employment record and counts against them come review time. They refuse to loosen their tolerance and make it more manufactureable. I have a part right now that is loosing 85% of the mass when we 1st started. It has a .2mm(.0078") profile tolerance everywhere that is a 3D surface. I have some tool well over 240mm long and told them upfront no way will i say I can hit those tolerances.

Wound up is caring so guilty as charged.

[JB7280]

1 minute ago, crazy^millman said:

Well you love this. Helping a company with parts. Every part that has gone on a tags for issues that inspection finds. 71 have been approved to be used as it and have not failed in testing or use. Every part that goes on table made by the machinist goes against them. Doesn't matter how much the machinist tells them it can't be done derangement doesn't want to hear it and says do it anyway. They make what they can make and then it gets tagged by inspection because it was .001 microns out of tolerance. That deviation goes on their employment record and counts against them come review time. They refuse to loosen their tolerance and make it more manufacturable. I have a part right now that is loosing 85% of the mass when we 1st started. It has a .2mm(.0078") profile tolerance everywhere that is a 3D surface. I have some tool well over 240mm long and told them upfront no way will i say I can hit those tolerances.

Wound up is caring so guilty as charged.

Interesting on that profile tolerance.  I have the same thing.  It's on all of their parts.  Doesn't happen to be for a large semi-conductor company, does it?  Lucky my employer knows the difficulty of the parts we are making, and those things don't get counted against us, or at least it's looked at on a case by case basis, and as long as it's not a carelessness, or lack of common sense type of error, it's understood.  

 

Going back to the original topic for a moment, can feed mills be made to work well in aluminum?  I wonder how a custom endmill, with a feed-mill type geometry would work for this application, considering that the toolpath is similar to what would be used for a feed mill.  Tried ramping, and it's definitely deflecting quite a bit.  but it's weird, like a staircase.  Almost like it's deflecting less with each pass.  Trying a different ramp depth now.  

 

And yes, the shape of the o-ring groove really is that stupid.  Once I find a good method to cut with this long tool, I do plan to eliminate the areas where I'm able to use a shorter tool.  Just using it for testing at the moment.  

 

 

 

[crazy^millman]

5 minutes ago, JB7280 said:

Going back to the original topic for a moment, can feed mills be made to work well in aluminum?  I wonder how a custom endmill, with a feed-mill type geometry would work for this application, considering that the toolpath is similar to what would be used for a feed mill.  Tried ramping, and it's definitely deflecting quite a bit.  but it's weird, like a staircase.  Almost like it's deflecting less with each pass.  Trying a different ramp depth now.  

 

And yes, the shape of the o-ring groove really is that stupid.  Once I find a good method to cut with this long tool, I do plan to eliminate the areas where I'm able to use a shorter tool.  Just using it for testing at the moment.  

 

 

 

I assume one way since it is contour you are using and yes that makes sense the deeper the tool gets the more stable it will be. It has the side walls supporting the endmill the deeper you go and that support will dampen the tool. Are you going down the center with ramp then coming back to the walls? Trying to find a highfeed style this small that long might be difficult. Might have to have it custom made.

The part is for car racing I was referring to.

[JB7280]

4 minutes ago, crazy^millman said:

I assume one way since it is contour you are using and yes that makes sense the deeper the tool gets the more stable it will be. It has the side walls supporting the endmill the deeper you go and that support will dampen the tool. Are you going down the center with ramp then coming back to the walls? Trying to find a highfeed style this small that long might be difficult. Might have to have it custom made.

The part is for car racing I was referring to.

I am using a contour ramp, down the middle, running back and forth, ramping in both directions. Then planning to finish the walls and floors after.  Unfortunately, the deflection has blown past the walls on this part.

 

 The endmills I have now are custom made, I wouldn't be opposed to getting more custom made if they would make the job smoother.  Might have to make a phone call on Monday. 

[crazy^millman]

Well I see that burr being pushed up and have to wonder how the tools were made? Unless this is gymmy grade of aluminum that it looks like you are cutting I wouldn't expect to see that much burr on a good sharp tool.

[JB7280]

5 minutes ago, crazy^millman said:

Well I see that burr being pushed up and have to wonder how the tools were made? Unless this is gymmy grade of aluminum that it looks like you are cutting I wouldn't expect to see that much burr on a good sharp tool.

Yes, I thought the same thing about the burr.  It's a brand new tool from Garr.  But I plan to take a good look at the edge when I get back in on Monday.  The material is just 6061-T6, stress relieved after roughing.  

[crazy^millman]

2 hours ago, JB7280 said:

Yes, I thought the same thing about the burr.  It's a brand new tool from Garr.  But I plan to take a good look at the edge when I get back in on Monday.  The material is just 6061-T6, stress relieved after roughing.  

There is you answer. Stress relieved if the material was stressed relived wrong then it became softer than T6 it might be down to a T4 or even a T3 and that makes all the difference in the world unless it was cold stress relieved than that wouldn't effect the temper to make it softer. I have had T6 get softer when heated up too much and get gummy on me.

[JB7280]

1 hour ago, crazy^millman said:

There is you answer. Stress relieved if the material was stressed relived wrong then it became softer than T6 it might be down to a T4 or even a T3 and that makes all the difference in the world unless it was cold stress relieved than that wouldn't effect the temper to make it softer. I have had T6 get softer when heated up too much and get gummy on me.

interesting.  Ill have to look at the heat charts.  Unfortunately the stress relieve spec is part of the order from the customer, so that can't be changed.  

 

I do have some pieces of the stock though, that haven't been stress relieved.  Ill try cutting the slot on one of those, with the same toolpath.

[Leon82]

I made a vibe fixture from t6 and they put in in the oven to weld it and when I put the finish profile on it it was a gummy mess. 

[CADCAM3D5AXIS]

Hey Guys on a lot of this type of work sometimes its in the guy making the parts interest  to have a sit down   table top review of the machine parts that they would like to make or attempt to make  and then go over the prints /models and pick out the trouble spots  and point out the errors of there ways in regards to DFM and materials and other options like hardware inserts and out processes like plating , heat treating  and anything else  and then ask them Do You really need the things that look like trouble and time consuming to machine   when it may just be a hole for a wire or hose to go thru and its only clearance not a bearing fit . and if they are receptive to change to make it a better part to build and less costly its in there interest    ie $$$$$ and also the machinist's interest less headaches then maybe they may just see the light . Hopefully . But for the most part if they are Professional Licensed Engineers and its on them  they may just be open to a lot of Good ideas to build his products in a timely manner that can be repeated without custom tools and lots of money and time down the drain. Then Like the One guy that was working with you  after a while they move on or go elsewhere if things keep being a PITA. 

 But Good Luck   some times they just need to have someone point out the error of there ways in a Friendly Manner so they don't Get Upset . 

[Newbeeee™]

Haven't read all....but holders.

I had a feature similar dia to length ratio (although a 5mm cutter) and couldn't stop it ringing like a church bell. My mate said use a ER25 holder.

I told him he was having a laugh, as we tried hydraulic shcunk, showa microchuck (mechanical), and just tried a shrink and was now outta ideas.

ER25 was gravy. Harmonics are a funny thing....

[cncappsjames]

Get the CS Contour and Orbital Milling option on your machine while you're at it then you can basically scrape it.

 

Google "Hale Milling". A certain machine tool builder likes to say it is their own proprietary function blah, blah, blah... (bovine excrement) it may be their name that is proprietary, but the function has been around for YEARS! I've been recommending the function for certain applications since the early 2000's. No telling how long before that it was created. 

[Matthew Hajicek - Singularity]

Sink EDM it.

[AMCNitro]

On 1/29/2022 at 10:29 AM, JB7280 said:

Ron, I don't mean to get you wound up any more, but I should point out that this customer supposedly has over 900 engineers with PHD's!!  I have a friend and former coworker who is a mechanical engineer, who basically demanded he be able to spend time working in the shop, running parts, setting up machines, and went on to do some programming as well.  Because of that, I think he's one of the most talented engineers I've met.  We no longer work together, but he still asks me quite often "how much more difficult will it be to machine this part if I put this tolerance on it".  Or, "what changes would make this feature easier to machine".

 

And he has always said anything above a master degree in an engineering field is a waste.  There's a point when it's a bigger benefit to get out of the classroom, and into the real world.  

Having a PHD doesn't mean that they have real world experience.  They can design anything they want, but if what they design cant be made, are they any good?  Not only that, they need to understand production as well, if your groove adds 30 minutes of runtime to the part they're costing the company money.  IF I was the CEO they could shove their PHD where the sun don't shine.  Problem is most CEOs know jack s**t about manufacturing and the first thing they do is blame the machinist/programmer.  I'd be a millionaire if I had a dollar for every time I said "this designer needs to spend a year in the machine shop".   I make molds these days(amongst other things) and I get tired of having to program .01 radiuses where a .015 could have done...

[JB7280]

On 1/29/2022 at 7:51 PM, CADCAM3D5AXIS said:

Hey Guys on a lot of this type of work sometimes its in the guy making the parts interest  to have a sit down   table top review of the machine parts that they would like to make or attempt to make  and then go over the prints /models and pick out the trouble spots  and point out the errors of there ways in regards to DFM and materials and other options like hardware inserts and out processes like plating , heat treating  and anything else  and then ask them Do You really need the things that look like trouble and time consuming to machine   when it may just be a hole for a wire or hose to go thru and its only clearance not a bearing fit . and if they are receptive to change to make it a better part to build and less costly its in there interest    ie $$$$$ and also the machinist's interest less headaches then maybe they may just see the light . Hopefully . But for the most part if they are Professional Licensed Engineers and its on them  they may just be open to a lot of Good ideas to build his products in a timely manner that can be repeated without custom tools and lots of money and time down the drain. Then Like the One guy that was working with you  after a while they move on or go elsewhere if things keep being a PITA. 

 But Good Luck   some times they just need to have someone point out the error of there ways in a Friendly Manner so they don't Get Upset . 

We've actually done that, and changed a lot of features on this part.  The O-ring groove, they say nothing can be done to change it.  I basically told them don't put another one in a part.  

23 hours ago, cncappsjames said:

Get the CS Contour and Orbital Milling option on your machine while you're at it then you can basically scrape it.

 

Google "Hale Milling". A certain machine tool builder likes to say it is their own proprietary function blah, blah, blah... (bovine excrement) it may be their name that is proprietary, but the function has been around for YEARS! I've been recommending the function for certain applications since the early 2000's. No telling how long before that it was created. 

James, I googled it and found a MAM72 video that looks similar to the turn-cut option we had on an Okuma, but more flexible.  I could see this groove being done with a long lathe grooving tool, in a holder which is supported on one side of it, with clearance on the other.  Am I thinking about this correctly?   

[Aaron Eberhard]

On 1/29/2022 at 11:42 AM, JB7280 said:

I do.  Unfortunately it was pretty far off on this one.  HSM Advisor reported .0019" deflection at the parameters I'm currently running.  What I'm seeing in reality is more like .012"-.013".

Just pointing out the obvious thing to keep in mind:  Although HSMAdvisor does a great job calculating mechanical deflection (e.g., how hard is this cut going to push my tool out of position), it really can't account for harmonics & vibration, which is what really causes chatter that is overcutting.  Remember that with this long of a stickout, you're going to be making a large spring every time you pull the flutes "into" the cut.  That's why you're overcutting, because it's bouncing back at least that hard. 

I like the suggestion of trying to do a conventional cut here.  

Another thing is if you can get a custom-grind on your tool shaft, even a small amount of taper will radically increase the stiffness of the tool.  We used to run into similar things when porting small cylinder heads and such, and the difference between a 2" vs 4" taper length is astounding.  It's mostly not due to rigidity, but to harmonics and damping.  Looking at the size of the tool, all of that taper happens over .2" or so, spread that back as far as you can.  You're trying to keep the tool from turning into a spring, and a cone is always a worse spring than a cylinder.

[Matthew Hajicek - Singularity]

[JB7280]

2 hours ago, Aaron Eberhard said:

Just pointing out the obvious thing to keep in mind:  Although HSMAdvisor does a great job calculating mechanical deflection (e.g., how hard is this cut going to push my tool out of position), it really can't account for harmonics & vibration, which is what really causes chatter that is overcutting.  Remember that with this long of a stickout, you're going to be making a large spring every time you pull the flutes "into" the cut.  That's why you're overcutting, because it's bouncing back at least that hard. 

I like the suggestion of trying to do a conventional cut here.  

Another thing is if you can get a custom-grind on your tool shaft, even a small amount of taper will radically increase the stiffness of the tool.  We used to run into similar things when porting small cylinder heads and such, and the difference between a 2" vs 4" taper length is astounding.  It's mostly not due to rigidity, but to harmonics and damping.  Looking at the size of the tool, all of that taper happens over .2" or so, spread that back as far as you can.  You're trying to keep the tool from turning into a spring, and a cone is always a worse spring than a cylinder.

So you're saying, have the taper ground more gradually, basically?  I plan to try conventional cutting when I get into the finish passes, however, I have no choice but to full slot, into the groove is roughed.  

[Guest]

On 1/29/2022 at 9:49 AM, JB7280 said:

I'm cutting an o-ring groove and I've tried a number of things without tons of luck.  I'm pretty sure in the design process, their goal was the opposite of machinability.  The o-ring groove is .135" wide, .078" deep, and .040" away from a 3" high wall!!!

 

I have a couple of custom endmills from Garr.  .1875" shank, .125" flute diameter, .100" flute length.  I wanted to have the biggest shank I can.  It's also got their diamond coating.  amorphous diamond i think.  Running it in a Lyndex, Cat50 VC6 holder.  Tool is 5.5" OAL, with 2.375" stuffed in the collet.   I thought the VC6 might be a good option because the collet is piloted at the end, so there is 2 points of contact.  I thought that might help a little with keeping the tool rigid.  I've tried all sorts of methods.   Full depth just wandered everywhere.  I expected that, but figured I'd try.  I've tried RPM from 3000 to 12000.  HSMAdvisor recommends .007DOC, 11300RPM, .0008ipt.  That was pretty noisy as well.  Does anyone have any suggestions to make this slot go a bit smoother?

 

The one thing I have going for me, is I don't have to ramp.  There is an area where I'm able to enter from the outside.  

 

Have you tried dropping the rpm down to say 500, you might even have to go less than that... Those spindle speeds are waaaay too fast

[JB7280]

13 minutes ago, byte said:

Have you tried dropping the rpm down to say 500, you might even have to go less than that... Those spindle speeds are waaaay too fast

I've gone as low as 1000.  My best results so far seem to be 2000rpm, 5ipm, and .008 ramp, not sure what angle that equates to.  I skipped it for now to finish the rest of the part, but i'll try lower when the shrink holder comes in.

 

I only gave the 11K a shot because I remember another job where I had an .050" endmill, about the same L:D ratio, ramping in 4140 and max rpm did the trick, so I figured I'd give it a whirl here.  

[Guest]

11 minutes ago, JB7280 said:

24 minutes ago, byte said:

Have you tried dropping the rpm down to say 500, you might even have to go less than that... Those spindle speeds are waaaay too fast

I've gone as low as 1000.  My best results so far seem to be 2000rpm, 5ipm, and .008 ramp, not sure what angle that equates to.  I skipped it for now to finish the rest of the part, but i'll try lower when the shrink holder comes in.

 

I only gave the 11K a shot because I remember another job where I had an .050" endmill, about the same L:D ratio, ramping in 4140 and max rpm did the trick, so I figured I'd give it a whirl here.  

.125 is a long way from .050, I would just scale your formula back till you are turning very slow, then work your way up, if needed..

How do u calculate your feeds and speeds? What I would do was make a table with formulas in google sheets and i would input my tool parameters there..

[Aaron Eberhard]

15 hours ago, JB7280 said:

So you're saying, have the taper ground more gradually, basically?  I plan to try conventional cutting when I get into the finish passes, however, I have no choice but to full slot, into the groove is roughed.   

Yes, that's what I'm saying about the taper.   Assuming everything else was the same, we got significantly better surface finish and material removal with the top profile vs. the bottom 7" long tools (note, this is going back like 10 or 12 years, so might I might not have the exact dimensions) :

Sorry, I missed that this was the slotting cut that did that.... At least half of the tool was conventional cutting then!  

Oh, and we were always chasing the RPM vs. Feed to get the perfect amount of material removal.