Threading M64x6 on QT15N

[tsabo]

Hey all.

I'm attempting to thread M64x6 threads on my QT15n T32B using top notch NT4R widia 6025 coating.
Material is 4150 prehard.

I stalled the spindle at both 160sfm and 200sfm .004 DOC. I am a little stumped as to where to go from here. It looked like it was taking a large width chip when the spindle stalled. Would a lower SFM help this because it lowers the feed?
There is the option for constant area threading, but I do not know much about it. I looked at the verification and seemed like it was going to take 100 passes Maybe someone could enlighten me on this.

Side question.
I have another part that gets a 10-24 thread with a .031 max undercut/relief before the shoulder. I thought you need at least the distance of one pitch to ensure a good last thread. I probably misunderstood what that meant. I also couldn't find a top notch that would thread that close to a shoulder. I looked at ER11 size lay down inserts and it only gets .04" away from the shoulder.

I'm still a bit new to threading, so I appreciate the help

Tyler

 

[gmc1724]

Angle should be 60 Deg for metric thread.

 

[hvnlymachining]

.031 is likely the depth not length of relief, full 60 degree depth of cut would be .0361" but since there should be a radius at the bottom of the thread .031" would be very close. ( I don't have that measurement handy) if you were given a drawing double check that you should have been given a depth and length of relief ( and usually shape, radius, square etc.) If that information was not given than you may have to go back to the customer and find out what they want. Your insert, if designed for 24 tpi should be able to thread to the center of the relief, if your relief is the correct width.

It is also possible that an engineer made a simple math error in calculating thread relief width.

 

[tsabo]

Angle should be 60 Deg for metric thread.

I saw this in the programming manual, but it doesn't make sense to me why it would matter. I'll give it a try anyways. I threaded some M20 x 1.5 and M16 x 1.5 with angle 55 and it was okay.
I watched Phil's mazatrol video on youtube, and he uses 60 for inch even though the manual says to use 55. The settings and what they do is slightly confusing to me.

 

[tsabo]

.031 is likely the depth not length of relief, full 60 degree depth of cut would be .0361" but since there should be a radius at the bottom of the thread .031" would be very close. ( I don't have that measurement handy) if you were given a drawing double check that you should have been given a depth and length of relief ( and usually shape, radius, square etc.) If that information was not given than you may have to go back to the customer and find out what they want. Your insert, if designed for 24 tpi should be able to thread to the center of the relief, if your relief is the correct width.

It is also possible that an engineer made a simple math error in calculating thread relief width.

The print specs .125"-.135" on the diameter of the relief and .032" width maximum. I never looked at the full profile inserts before, but it does appear that they can get closer to the shoulder. Thank you for this info.

 

[tsabo]

I tried a few things here. I changed the angle to 60 and lowered the sfm to 140 which ran at about 50ipm. I went to the constant area cutting setting with a first DOC of .0164. Stalled the spindle on the 2nd to last pass.

Changed to .014" first DOC and kept everything else the same. Stalled the spindle with probably 10-20 passes remaining.

Things I noticed when the machine was running:
1) Sometimes the tool would pause for a split second in front of the part before running the next pass. Other times, the tool would go right into the cut with no hesitation.

2) The spindle load was typically around 10-20% but would jump to 35-45% occasionally.

3) On the part that almost fully threaded. The crests look like they vary in width. Note: I am only threading about 5 threads because these are scrap pieces of 4150 I have laying around. The actual part is about 13 threads long.

Could there be an issue with the spindle going "out of sync" with the thread? If so, how would I be able to check/fix this? I think this is more of a machine issue than a program issue. Maybe a parameter?

-Tyler

 

[Philabuster]

Perhaps the encoder is dying and the machine is loosing sync with the thread which is causing the machine to engage the thread at a different Z position? That would explain the varying spindle loads.

Have you checked the condition of the encoder belt or electrical connectors to see if it is something loose causing false signal?

How about the condition of the spindle drive belts? Are they loose?

 

[tsabo]

Perhaps the encoder is dying and the machine is loosing sync with the thread which is causing the machine to engage the thread at a different Z position? That would explain the varying spindle loads.

Have you checked the condition of the encoder belt or electrical connectors to see if it is something loose causing false signal?

How about the condition of the spindle drive belts? Are they loose?

The spindle load would be fine for a few passes then randomly in the middle of a pass spike up unusually.

The encoder belt looks to be in good condition. 6 of the 7 spindle belts look good and the one closest to the spindle looks a little worn. They feel like they have correct tension on them

The connector that goes into the encoder has a tad wiggle to it, but it felt like it was seated all the way in.

There is oil everywhere in the back of the machine. The encoder belt is dry and the spindle belts are a little oily on the outside of them.

-Tyler

 

[Philabuster]

The spindle load would be fine for a few passes then randomly in the middle of a pass spike up unusually.

Again, this is due to the machine loosing sync with the thread and taking a heavy bite randomly. The spindle load should slowly creep up as the thread gets deeper, but not spike up.

In another thread on PM recently, someone had a similar issue on another brand lathe and it was the connector to the spindle encoder that was the problem. They removed the connector and cleaned it and it worked after that. Try that first.

 

[tsabo]

In another thread on PM recently, someone had a similar issue on another brand lathe and it was the connector to the spindle encoder that was the problem. They removed the connector and cleaned it and it worked after that. Try that first.

I unthreaded the connection and everything is clean and looks good. I tried another test piece and I'm still getting spindle load spikes and varying thread crests. I ran a test in aluminum just to compare and the threads look better, and it didn't seem to have the same load spikes. Still stumped.

 

[Philabuster]

Have you tried higher SFM yet? Perhaps 300 or so? I'm out of ideas.

 

[tsabo]

Have you tried higher SFM yet? Perhaps 300 or so? I'm out of ideas.

Just tried one this morning @300sfm. It looked promising but, about half way through it started spiking and then stalled. I broke the shim seat for my tool, so that's about all the testing i can do at the moment.

Would you still be leaning towards a faulty encoder?

Thank you for your time on this
-Tyler

 

[Philabuster]

That would be my guess but I am not a maintenance tech.

 

[RoundHouse M&F]

93 sqt18ms in the shop i worked at ( a c axis), the encoder board on the spindle unit controlled the thread timing and c axis positioning. it started going out and would show only on the c axis when it would get hot but would thread just fine, our tech said it was a wonder it would still thread at all. as the board got hot it would start losing degrees in the positioning. after the boss figured out just the new board cost $6000 USD (was not rebuildable because it had some kind of capacitor/battery pack built in, cant rember what it was), a fan was bought instead to aim at the spindle and help keep it cool in a non AC shop. surprisingly this worked for 3-4 years i was still there, and my buddy who still works there says its still working.

Im not sure how your encoder is wired, but you could have a board getting hot and going out of sync.

 

[EnderDRM]

been there done that