What's new
What's new

CNC is producing a taper in my tapped hole

dvice

Hot Rolled
Joined
Mar 12, 2004
Location
dutchess ny
Im drilling and tapping a 4-40 hole in 440 C stainless steel. Im drilling .093 hole( slightly large) on purpose. I'm using an EXOtap coated with TiCN , plug tap 4-40 G3 2 flute . Ive drilled .550" and am tapping only.350" deep. The lead on the tap is about .170. Ive told the program it is only .050.(didnt think it matters). Im starting .100" above the surface, to tap, and returning to that spot with the program, after I tap. The hole is true and round before the tap cycle. after the tap cycle, a .104 guage pin will start into the tapped hole, and a .099 goes about half of the taper that results from the tap operation... you can wiggle the pin a lot... When you look at the wall, there is barely a mark from the tap until you get about .05 or more down into the hole. The program rpm is 400, it gives me a feed of .030. ive left the feed at 100%on the control(dont think that matters) I took a part out before tapping, and tapped it by hand and I do not have this taper!!! why is the program causing the large taper in the top of the tapped hole?....is it that I programed that the tap only had a .050 lead, when in fact it has a .170 lead??? Im going to go change it now and check, but i thought id ask here anyway
 
i saw something very similar recently while troubleshooting tapping small waspaloy nuts on a lathe. the problem ended up being the operator slapping a new tap in every once in a while without inspecting the collet and collet holder for chips.

Shouldn't the feed for a 4-40 be .025 per rev?
 
i saw something very similar recently while troubleshooting tapping small waspaloy nuts on a lathe. the problem ended up being the operator slapping a new tap in every once in a while without inspecting the collet and collet holder for chips.

Shouldn't the feed for a 4-40 be .025 per rev?

the .030 is the feed GIBBS CAM gave.... I installed the info as 40 TPI.... and the rpm is 400.... the program figures the feed..... based on the rpm
tapped 100 of thousands of holes with the same program.... I changed the lead on the tap to .170 in the description of the tap in GIBBS, but it didnt make a different
 
the .030 is the feed GIBBS CAM gave.

That's an excuse for being lazy and not double checking the numbers. If my employee gave me that excuse his ass would probably be sent home for the day.

40 threads per inch = .025" per rev.
400 rpms at .025" per rev is 10 inches per minute.

Math doesn't lie.
 
the .030 is the feed GIBBS CAM gave.... I installed the info as 40 TPI.... and the rpm is 400.... the program figures the feed..... based on the rpm
tapped 100 of thousands of holes with the same program.... I changed the lead on the tap to .170 in the description of the tap in GIBBS, but it didnt make a different

So after asking what is wrong, and being shown what is wrong, you come back with this. I think you are correct, it must be something else. Good luck.
 
That's an excuse for being lazy and not double checking the numbers. If my employee gave me that excuse his ass would probably be sent home for the day.

40 threads per inch = .025" per rev.
400 rpms at .025" per rev is 10 inches per minute.

Math doesn't lie.

wow Im being lazy? I do something ive done for 20 years..... plug in the feilds on my cam package and it kicks out the wrong info...and its my fault! unbelievable.... I went into the DMG and altered the feed to .025 and the threads look nice... I just never had an issue like this before.... and it is amazing the tap didnt break....in gibbs there isnt a spot to PLUG IN THE FEED FOR RIDGID TAPPING!!! you fill the TPI field.... and plug in an RPM...the program does the rest.....im blown away that I came here and i get the massive guilt trip because I didnt do it the old fashioned way...I went ahead and used the CAM PROGRAM!!!!!!!!!!!!!!
 
Calm down. Learn from this that you have to check outputs. You're not going to "win" arguing with us about whether the code was wrong or not, it's on you, the programmer, to make sure things are correct.

Especially something so easy to find as incorrect pitch to RPM.

I do some 5 axis jobs, and it gets kinda complicated, and I have cut swept shapes and the program is VERY dynamic...especially if it is cutting spline type info... and feeds are changing all the time.... espcially if the file came in and was poorly made. there is no possible way anyone would sit down and try to figure what all the outputs should REALLY be... they just go to RENDER and look for issues or problems... and or go to the machine and run operation by operation on the first part!!!!!
 
wow Im being lazy? I do something ive done for 20 years..... plug in the feilds on my cam package and it kicks out the wrong info...and its my fault! unbelievable.... I went into the DMG and altered the feed to .025 and the threads look nice... I just never had an issue like this before.... and it is amazing the tap didnt break....in gibbs there isnt a spot to PLUG IN THE FEED FOR RIDGID TAPPING!!! you fill the TPI field.... and plug in an RPM...the program does the rest.....im blown away that I came here and i get the massive guilt trip because I didnt do it the old fashioned way...I went ahead and used the CAM PROGRAM!!!!!!!!!!!!!!

Software isn't infallible, it can become corrupt for reasons I cannot explain. Sometimes you just have to go old school to fix a problem. I think that gives us older guys an edge, we aren't helpless when the computer or internet go down. No reason for you to get defensive. I am shocked at people's inability to do math in their head or on a piece of scratch paper these days. How can someone not know quickly 1/40 = .025.

The defensiveness of people these days boggles my mind, doesn't anyone own their mistakes any more?
 
Yes, as the owner, I've got a few 5 figure doorstops to keep reminding me.

I always say if a person learns by their mistakes I should make Einstein look stupid by now. Hope those 5 figure screw ups were in the low 5 figures. Self employment can be a roller coaster ride for many, I would not have it any other way. I would rather die broke working for myself than well off working for others, and I would rather die on my feet than live on my knees. To each his own.
 
You should never just run whatever code your CAM spits out when it comes to threading without first looking at it to make sure it is correct.

Take this as lesson learned and stop being so dramatic. It's your job to make sure everything is correct.
 
I do some 5 axis jobs, and it gets kinda complicated, and I have cut swept shapes and the program is VERY dynamic...especially if it is cutting spline type info... and feeds are changing all the time.... espcially if the file came in and was poorly made. there is no possible way anyone would sit down and try to figure what all the outputs should REALLY be... they just go to RENDER and look for issues or problems... and or go to the machine and run operation by operation on the first part!!!!!

Cool story, bro. (JK)

CAM software isn't infallible, as you've just discovered. As the programmer, it's part of your job to ensure the code you send out to the shop floor is up to speed. It's one thing to use a verify tool to check a 40meg program, it's another thing to look at five lines of code for tapping a hole. Hopefully you never do that as a contract programmer sending code via e-mail to a customer. It won't win you any friends.
 
wow Im being lazy? I do something ive done for 20 years..... plug in the feilds on my cam package and it kicks out the wrong info...and its my fault! unbelievable....

YES IT IS YOUR FAULT!
Who entered the info into the computer = YOU
Who hit cycle start = YOU
Who didn't make sure the program was correct = YOU
 








 
Back
Top