Hello Guest it is March 28, 2024, 10:46:31 AM

Author Topic: lathe will not thread  (Read 5013 times)

0 Members and 1 Guest are viewing this topic.

Offline fast89

*
  •  154 154
    • View Profile
lathe will not thread
« on: August 18, 2017, 07:30:05 PM »
Yeah i've read post after post. Been trying to get this lathe to work for 2 days now. This is my third lathe, the other two have probably threaded 100,000 parts. Def. not new to threading. Here's the problem, tool comes into position, rpm is stable within 1 and the actual feed-rate on the g76 line is .001" . It should be .05"  for the pitch. I've tried everything, g95 is in effect, different rpm, more debounce, 0 debounce, index active low, active high. The only difference between the three lathes is this one has a servo spindle with step/dir control. The other two are PWM control with  a vfd. I tried mach version .066, .062 (currently) and .022 (wouldn't load).   This lathe has the same mach motion apollo 1 breakout board the other two have. Pressure is on, i have alot of parts to run and everything works good except threading.




Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: lathe will not thread
« Reply #1 on: August 19, 2017, 06:58:05 AM »
Could be due to a number of things.
Will post later today.........

RICH

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: lathe will not thread
« Reply #2 on: August 19, 2017, 06:46:45 PM »
Don't think the internal threading of Mach3 was ever changed in years or after version 3.042.034 and I use version .062 and have had no problems.

Info from Threading on the Lathe Manual:

A gcode file defining the axis moves along with related thread information is read by  Mach3. The index pulse provides Mach with rpm data and the program controls the Z axis to a move appropriately from a dead start, accelerate to a defined distance, and then maintain a feedrate such that the cutting tool produces a spiral cut along a cylinder representing the lead of the thread. The start of the Z motion happens when and if a timing pulse is seen. If no pulse is seen the threading will not begin or continue. The timing pulse synchronizes the z axis location to the spindle rotation the same as closing of the half nuts on a manual lathe would do. So, the threading is activated with an index pulse.


The P in G76 should be the pitch and YOU define it. I will assume that you are using the original G76 command and not some altered one. If you use the wizard you should use the Calc Number of Passes button to check if you motor setting is appropriate for the rpm and pitch entered. Remember that with a stepper as the velocity is increased the torque decreases / there is less power available and you want some “headroom” left for  cutting the  thread. Also you have a choice in what code will be posted and that is covered in section 7.2 of the manual.
-------------------------------------------------------------------------------------------------------------------
NOTE:
If you are generating gcode from some program then that is different matter as the post processor needs to be correct for MACH3 LATHE.

-------------------------------------------------------------------------------------------------------------------

So from the manual info above, it seems that the index which triggers threading, is not seen by Mach3. Test that the index is working ( how to do is in the manual). Not familar with the Appolo 1 break out board, use of their plug in, or their screen set.

Do you have the index configured in Mach3? How?
Are you setting spindle rpm manually? ( don't want the vfd fighting the index info to Mach as that can  become problematic)

FWIW,

RICH

Offline fast89

*
  •  154 154
    • View Profile
Re: lathe will not thread
« Reply #3 on: August 19, 2017, 08:09:22 PM »
Mach sets the spindle speed via step/dir motor control as it is a servo. It is over 2kw servo spindle motor and the axis motor is 750w. No problem on power. It does begin the thread so i believe it is seeing the index pulse. i can rotate the spindle and see the led light up. With any kind of speed though the led is very erratic or doesn't light at all. I thought this could be the issue but the other two lathes are the same. It is seeing the rpm  good all the way to 4k rpm. It's almost like mach is not seeing the true rpm even though it is displayed correctly. I even switched back to the stock turn screen but no joy. I have also disabled a axis since it uses the same port/pin as the spindle. still no joy.

Offline fast89

*
  •  154 154
    • View Profile
Re: lathe will not thread
« Reply #4 on: August 20, 2017, 06:16:41 AM »
The index signal is from an optek and goes to the breakout board. Just a port an pin assignment in mach3 active high. I have a slotted disc on the spindle shaft itself, the slot is about 1/2" wide, Started out at .300" wide but i thought maybe the pulse was too short so i opened it up to 1/2".

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: lathe will not thread
« Reply #5 on: August 20, 2017, 08:57:04 AM »
Don't want jerk you around.
If I think of something concrete will reply as I don't like to guess.

RICH
 


 

Offline fast89

*
  •  154 154
    • View Profile
Re: lathe will not thread
« Reply #6 on: August 20, 2017, 04:20:29 PM »
Did some more testing today. If i uncheck use spindle feed back in sync mode, it runs thru the thread at proper speed. Of course it won't make a good thread because it doesn't know when to start though. Made me think  it was a sensor or disc issue. I changed the optical pickup, fixed the bracket for better alignment, cut the slot wider and painted the shiny disc with black paint. Put it all back together and same thing, it wants to thread at .001" pitch  instead of .05"

Offline fast89

*
  •  154 154
    • View Profile
Re: lathe will not thread
« Reply #7 on: August 20, 2017, 07:12:14 PM »
Ok, i might be on to something here. Because this is a step/dir servo spindle, i have electronic gearing in the servo drive to speed up the motor without mach having to put out a ridiculous pulse rate. I just have the pulley ratio set to 1 and the rpm reads correctly and stable. Now, if i change the pulley ratio to 13 (approximately what is in the servo drive), i get incorrect rpm but the threading cycle will feed at much closer to the programmed .050" pitch.
Re: lathe will not thread
« Reply #8 on: August 20, 2017, 11:03:08 PM »
Hi fast89,
if you didn't use electronic gearing what pulse frequency would be required to drive your spindle at threading
speed?

To max out Mach it would have to be spinning really fast unless you are using one of those 'over the top'  high resolution
encoders.

Craig
'I enjoy sex at 73.....I live at 71 so its not too far to walk.'

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: lathe will not thread
« Reply #9 on: August 20, 2017, 11:04:10 PM »
fast89,
What I am struggling with is that you are using a different screen set, and may have different macros in use, an index that's
runs through a bob, and a bunch of things that requires setup via plug-in.
SO
Maybe simplify the whole thing by just testing with basics.

INDEX
I see  nothing wrong with your reply about your physical index. It turns on and off per test, you can make the single slot much wider, even a lot wider, than what is calculated per the Mach3 Turn Manual. Wider is better than too narrow. As long as you are getting correct rpm range in DRO all is good. The rpm is based on the index and the index also implements the movement to start threading when mode changes from G94 to G95.

In the simplest way of index configuration the index pulse just goes directly to pp port pin of the computer. Nothing to influence
it's info to Mach, such as going through a bob or whatever. One can actually make a timer circuit  ( or use a pulse generator instrument ) for  providing different pulses and check the DRO with out a motor running ( click the spindle button to turn on a faked motor which is actually the timer and Mach thinks the motor is turning the spindle / index providing for a dro readout and also threading implementation. It's been years since we tested and fixed the threading cycle.

So what the above would do is isolate the index to setup and confirm, then can be run through the bob or what ever and see if all is the same, and then add some stuff and define  the culprit.

Now I know that there have been modified versions of the M1076 done and posted. I would think we want to make sure the original is used as it's been around for years and frankly can't recall any problems  in  a looooooooong time of it's use.  Heck don't even anything and just create some  repeating code based on G32 ( which is nothing more than a G01 at a feed with the addition of when to start axis motion )

RICH