What's new
What's new

G84 Form Tapping Aluminum Troubleshooting

Raymgriff

Plastic
Joined
Feb 8, 2018
Hi guys,

I'm running into an issue while tapping a 3/8" NC hole in some 3/8" thick aluminum plate.

Essentially what's happening is that the canned cycle is working properly while plunging into the threading. The problem is that while retracting, the tap isn't being fully cleared from the hole by the time the tool has stopped climbing in the Z. So, the tap is spinning in reverse while still sitting in the hole and it winds up reverse feeding the tap into the last 2 threads of the tapped hole. It's making them pretty chewy and I'm having to chase the threads from the other side of the hole. On 140 holes it's a bit annoying.

I'm running this in a floating tap chuck and I've played around with dwell times as well as retract heights and feed heights. I'm programming all of this in Inventor HSM and I've got the tapping cycle set as though it should be starting 1/4" above the part, which should be plenty of clearance for the spindle to allow the tap to be clear of the part. The retract height is also way overkill (.6" above the top height). Based on my setup to my knowledge anyways, the tap should be reverse feeding well beyond the top extent of the hole, but no matter what I do it doesn't seem to want to clear the tap from the hole... Am I missing something that should be extremely obvious?

Ray
 
Hi guys,

I'm running into an issue while tapping a 3/8" NC hole in some 3/8" thick aluminum plate.

Essentially what's happening is that the canned cycle is working properly while plunging into the threading. The problem is that while retracting, the tap isn't being fully cleared from the hole by the time the tool has stopped climbing in the Z. So, the tap is spinning in reverse while still sitting in the hole and it winds up reverse feeding the tap into the last 2 threads of the tapped hole. It's making them pretty chewy and I'm having to chase the threads from the other side of the hole. On 140 holes it's a bit annoying.

I'm running this in a floating tap chuck and I've played around with dwell times as well as retract heights and feed heights. I'm programming all of this in Inventor HSM and I've got the tapping cycle set as though it should be starting 1/4" above the part, which should be plenty of clearance for the spindle to allow the tap to be clear of the part. The retract height is also way overkill (.6" above the top height). Based on my setup to my knowledge anyways, the tap should be reverse feeding well beyond the top extent of the hole, but no matter what I do it doesn't seem to want to clear the tap from the hole... Am I missing something that should be extremely obvious?

Ray

Check your tap holder. You can adjust some of them for compression and pullout (extension I guess)? When pulling and pushing by hand does it have somewhat equal travel? IF not it needs adjustment (if you have that style). There should be a screw down inside where the tap adapter fits, at least ones I have used it was like that. But it might also be a machine problem. I use something like this -

N9780 T8 M6
N9790 G0 G90 G54 X-.73 Y.215 S600 M3
N9800 G43 H8 Z2.
N9810 M8
N9820 G98 G84 Z-.35 R.2 F15. (change R value here to .5 and see what happens)

more positions

N9920 G80
M9

If it is still not working with an exaggerated R value you probably have a machine problem like an encoder, or belt slipping, or..??
 
what kind of machine/ control? Sounds like your spindle isn't in sync with your Z. Can you watch your tap holder and see if it's compressing/ expanding?

Ist thing I'd do is check your spindle speed with a tach.
 
what kind of machine/ control? Sounds like your spindle isn't in sync with your Z. Can you watch your tap holder and see if it's compressing/ expanding?

Ist thing I'd do is check your spindle speed with a tach.

It's a Milltronics BR5100 gantry style machine. It's an almost new milltronics controller. These controllers run a modified fanuc as far as I know.

The tap holder is definitely expanding and as far as contraction goes it looks like it's spinning and for lack of a better word, "chattering" in the last two threads and is springing back into them with the compression of the tap holder. G84 being a rigid tap cycle, there should be almost no compression or expansion in the tap holder, correct?
 
Check your tap holder. You can adjust some of them for compression and pullout (extension I guess)? When pulling and pushing by hand does it have somewhat equal travel? IF not it needs adjustment (if you have that style). There should be a screw down inside where the tap adapter fits, at least ones I have used it was like that. But it might also be a machine problem. I use something like this -

N9780 T8 M6
N9790 G0 G90 G54 X-.73 Y.215 S600 M3
N9800 G43 H8 Z2.
N9810 M8
N9820 G98 G84 Z-.35 R.2 F15. (change R value here to .5 and see what happens)

more positions

N9920 G80
M9

If it is still not working with an exaggerated R value you probably have a machine problem like an encoder, or belt slipping, or..??


Here is the program code. The R value is well above the part (almost a whole inch) with the top of the part being Z=0 I'm thinking it might be an issue of the spindle rotation being out of sync..

N70 M29 S1600
N71 G84 X17.6875 Y-0.5625 Z-0.475 R0.85 P1 F100
N72 X23.375
N73 Y-3.9688
N74 Y-8.0312
N75 Y-11.4375
N76 X17.6875

I know that there is an issue with one of the contacts that controls the high and low gear speeds of the spindle, where the spindle can only operate in high gear (the high gear contact is stuck closed so because it cant release, the low gear contact can't engage. New part is on order) I'm wondering if maybe the inability to shift gears is causing the issue?
 
Here is the program code. The R value is well above the part (almost a whole inch) with the top of the part being Z=0 I'm thinking it might be an issue of the spindle rotation being out of sync..

N70 M29 S1600
N71 G84 X17.6875 Y-0.5625 Z-0.475 R0.85 P1 F100
N72 X23.375
N73 Y-3.9688
N74 Y-8.0312
N75 Y-11.4375
N76 X17.6875

I know that there is an issue with one of the contacts that controls the high and low gear speeds of the spindle, where the spindle can only operate in high gear (the high gear contact is stuck closed so because it cant release, the low gear contact can't engage. New part is on order) I'm wondering if maybe the inability to shift gears is causing the issue?

Are the threads ok beyond the first couple (they are through yes?), if they are..hmm. Not sure what your P1 does in the second block? Not sure on your control, but I have seen G84 used even when machine did not have rigid tapping, that is what the compression style toolholder comps for. If you trust your eyes enough, try running the tap cycle above the part. Is the spindle instantly stopping and reversing, or does it kind of ramp down in rpm as the Z is decelerating? This *should* be an indicator if you have actual rigid tapping on your machine or not. If it does appear to be a instant stop/reverse cycle you could try your tap in a collet chuck and see what that does.*

* at your own discretion!! use caution!
 
if you have rigid tapping, why are you using a floating holder? I'm not sure why, but I've always gotten shit threads doing that.

It was mostly for convenience in a sense that it was a spare tool holder already devoted to tap holding. Already had the tap collet etc.

It's a thru hole and I've got plenty of clearance passed the actual hole bottom. So, even though with the floating tap holder it might not start the thread immediately and may compress a bit before the threads actually engage, it hasn't effected the majority of the threads. It's just on the retraction that's being problematic.
Are the threads ok beyond the first couple (they are through yes?), if they are..hmm. Not sure what your P1 does in the second block? Not sure on your control, but I have seen G84 used even when machine did not have rigid tapping, that is what the compression style toolholder comps for. If you trust your eyes enough, try running the tap cycle above the part. Is the spindle instantly stopping and reversing, or does it kind of ramp down in rpm as the Z is decelerating? This *should* be an indicator if you have actual rigid tapping on your machine or not. If it does appear to be a instant stop/reverse cycle you could try your tap in a collet chuck and see what that does.*

* at your own discretion!! use caution!

The threads passed the two starting threads are fine, it's just the chatter happening as the tap reverse feeds itself into the compression of the tap holder. If I were running it in a rigid holder, it leads me to think that the tap wouldn't just chatter, it would full on shear the threads right out.

I'll run a dry run and see how it looks anyways. I've got the retract set a full inch above the part and it hasn't changed anything about how the tap interacts with those first two threads...
 
It was mostly for convenience in a sense that it was a spare tool holder already devoted to tap holding. Already had the tap collet etc.

It's a thru hole and I've got plenty of clearance passed the actual hole bottom. So, even though with the floating tap holder it might not start the thread immediately and may compress a bit before the threads actually engage, it hasn't effected the majority of the threads. It's just on the retraction that's being problematic.


The threads passed the two starting threads are fine, it's just the chatter happening as the tap reverse feeds itself into the compression of the tap holder. If I were running it in a rigid holder, it leads me to think that the tap wouldn't just chatter, it would full on shear the threads right out.

I'll run a dry run and see how it looks anyways. I've got the retract set a full inch above the part and it hasn't changed anything about how the tap interacts with those first two threads...

Another question. After the tap spins a couple times at the top of the hole on the retract, does the machine/tip of tap then get to your R plane? If it does, it sounds like your holder is screwed up. IE, it is making the retract moves ok, but spring pressure is keeping the tap 'engaged' for a couple more revs.
 
Here is the program code. The R value is well above the part (almost a whole inch) with the top of the part being Z=0 I'm thinking it might be an issue of the spindle rotation being out of sync..

N70 M29 S1600
N71 G84 X17.6875 Y-0.5625 Z-0.475 R0.85 P1 F100
N72 X23.375
N73 Y-3.9688
N74 Y-8.0312
N75 Y-11.4375
N76 X17.6875

I know that there is an issue with one of the contacts that controls the high and low gear speeds of the spindle, where the spindle can only operate in high gear (the high gear contact is stuck closed so because it cant release, the low gear contact can't engage. New part is on order) I'm wondering if maybe the inability to shift gears is causing the issue?
G84 expects the feedrate to be in TPI. You have it programmed at 100. I'm assuming it is not a 3/8-100 tap correct?

Try correcting that and I bet things will get better.

Teryk



Sent from my XT1710-02 using Tapatalk
 
Another question. After the tap spins a couple times at the top of the hole on the retract, does the machine/tip of tap then get to your R plane? If it does, it sounds like your holder is screwed up. IE, it is making the retract moves ok, but spring pressure is keeping the tap 'engaged' for a couple more revs.

So I've done a bit more digging and I think I may have found the issue. Unlike every other CNC programming language I've come across, Milltronics has taken the liberty to be different and actually uses G84 for soft tapping which explains the lack of sync... So I need to change the post processor for G88 in order to do rigid tapping. I'll give it a go and report back. Hopefully it's not too big a pita to modify the post processor.

Ray
 
Here is what I think is happening..

Form taps don't start easy.. Coarse form taps are even worse..

Your T/C holder is compressing a ton trying to get the tap to start,
and it *might* even be bottoming out.. So now when you come out
of the hole, the T/C holder expands and pushes your tap down and
screws up the first thread of your hole...

2 things you could do.. Tap it twice in the machine.. The second time
the tap will follow your already existing threads and clean them up,
that way you don't have to do it by hand...

The other thing.. Feed in SLOWER than the pitch to allow the compression
to come out of the holder, and feed out FASTER than the pitch, for the
same reason..

I don't know Fanuc or Milltronic specific codes to save my life. But G84 on a Fadal
automatically feeds out 5% faster than it feeds in.. Floating tap cycle. That
retract feed can be altered with the P.. Basically the same thing you have..
Try a higher P #.

You should be able to find a sweet spot where your T/C holder is at a neutral point
on the feed out.
 
G84 expects the feedrate to be in TPI. You have it programmed at 100. I'm assuming it is not a 3/8-100 tap correct?

Try correcting that and I bet things will get better.

Teryk



Sent from my XT1710-02 using Tapatalk

I'm going to assume your phone may have done some autocorrect. ;)

3/8-16 tap @ 160rpm=10ipm feed

3/8-16 tap @ 1600rpm=100ipm feed. His feed is correct, but I think the spindle is a little fast IMO, but should work if everything is good and tight..
 
You might want to try increasing your feed by 2-5% to actually compress the holder.
I bet the short time spent reversing at the bottom of the hole is straining the tap and holder.

Also, I don't think you need to modify your post. Just change the 4 to an 8 in the control.;)
 
I’m tooling guy and it sounds like you figured out the problem in the programming. FWIW you should be using a collet chuck for form tapping or a rigid tap adapter.
 
I'm going to assume your phone may have done some autocorrect. ;)

3/8-16 tap @ 160rpm=10ipm feed

3/8-16 tap @ 1600rpm=100ipm feed. His feed is correct, but I think the spindle is a little fast IMO, but should work if everything is good and tight..
No, I meant what I posted, only I could have been more clear. I agree with you that, on the machines I have run, the feedrate for tapping is calculated as you wrote. My understanding was that, on Fanuc, the feedrate is specified in threads per inch unless preceded by a G95 in which it would be inches per rev.

Rigid Tapping G84 Canned Cycle - CNC Training Centre

Admittedly, the example on the above page is poor in that it uses a 20tpi thread and a 400 rpm spindle speed :(
So perhaps his understanding of the GCode is incorrect and he just got lucky. Looking elsewhere I can't see that statement repeated so probably a mistake.


Teryk
 
His issue is that he's using G84, which on a Milltronics is for tension and compression tapping. If he wants to rigid tap, it's G88 and the F is the pitch, .0625. He also had dwell in there, P1, which you don't need with rigid tapping.
 
His issue is that he's using G84, which on a Milltronics is for tension and compression tapping. If he wants to rigid tap, it's G88 and the F is the pitch, .0625. He also had dwell in there, P1, which you don't need with rigid tapping.

Brian is correct. I've got the rigid tapping working correctly and I've now removed the dwell. As far as getting the cam software to post properly I've gotten it to post with the G88 but I have yet to determine how to get it to post with the proper F equaling to the thread pitch. Until then I can just manually alter the program.

Oddly, the post processor is set up for default G84 being perceived and variabled as rigid tapping. So that's why when it did run the compression/tension tapping cycle, it wasn't changing its behaviour when I'd alter the retract parameters in the cam software. Now that the software is properly calling up the G88 it's referencing all the retract heights as intended. I just need to sort out the parameters for spitting out the F value in the post processor code which is turning out to be a bit tricky. From one issue on to another it seems.
 
There is no G98 or G99 called out, it really should be. It sounds to me like it is your machine if the Z stops feeding and the spindle keeps spinning. Is your tap holder centered in the movement or does it only compensate in one direction? Mine uses a spring to pull the tap all the way up so I feed at 95%. The changes you need to your Fusion post are dirt simple.

Since it is a form tap try tapping at 300 rpm with nothing in the way and gently hold the tap with something to tap against as it taps to check sync. If it's not obviously out checking it this way the tap holder will compensate fine.
 
In G84 (on a Milltronics machine) the spindle is not synchronized to the Z feed. So you call out a Z-.5 for example and you are spinning 1000 rpm then it takes maybe .8 of a second to reverse the spindle, that's where the Tension and Compression holder springs out, and the dwell is necessary for the spindle to stop and get reversed before you call out a Z.3 move to get clear of the hole. Yes. he should be using G99 (stay down at the specified R level, I like to use .3 when tapping) and G98 on the last line to pull back up to safe level.

Unfortunately Milltronics software is kind of bass ackwards from everybody else and uses G88 for rigid and G84 for T&C. You will need to mess about with your post to get the proper output. I believe someone on the cnc dot zone forum for Milltronics machines, another user posted a processor that works for Rigid tapping on the Milltronics controls with Fusion.
 








 
Back
Top