Actual spindle RPM is very steady, no issues there I can tell.  Is it easy to go backwards to v62?  Would I be better off migrating to Mach4?  I don't want to make a science project out of this, but I DO want to get the threading working again.

Mach4 would definetly be a science project at this point. Do a simple search for the best ver for turn and threading it may NOT be V062 it is probably an older version.

(;-) TP

Can't help you if you don't want to do as requested since
I will need to dupicate what you are doing asking some simple
questions based on eliminating posibilities to begin with.

G96 is constant surface speed, G46 sets the max speed when in G96,
and G97 turns CSS off ( from my notes / not in the manual, and I
don't use them ).

If you did read the Threading on the Lathe Manual, page 4,
How It Works, then would understand a little why I request using the
generic screen set. There may be problems when using CSS as it may
be fighting the threading which is reading true time rpm value and adjusting the next pass based on the monitored rpm during the threading.
If it does not see the index then the treading cycle won't start. It waites for  an index and changes from G94 to G95.

For Now,
RICH

Can't help you if you don't want to do as requested since
I will need to dupicate what you are doing asking some simple
questions based on eliminating posibilities to begin with.

G96 is constant surface speed, G46 sets the max speed when in G96,
and G97 turns CSS off ( from my notes / not in the manual, and I
don't use them ).

If you did read the Threading on the Lathe Manual, page 4,
How It Works, then would understand a little why I request using the
generic screen set. There may be problems when using CSS as it may
be fighting the threading which is reading true time rpm value and adjusting the next pass based on the monitored rpm during the threading.
If it does not see the index then the treading cycle won't start. It waites for  an index and changes from G94 to G95.

For Now,
RICH

So, I tried using the standard Marc3Turn screen and that doesn't work at all - it was never configured for my machine so it just throws out a bunch of errors. 

I tried my threading test program again (using my custom screen) and it still does the same thing.  It is exactly the same behavior as when using a free license.  I'm hoping someone has a clue why it would start acting up all of a sudden after working for a long time.

Be of good cheer. Threading does work very well. I use it all the time.

I am running W7, M3...062, with an ESS, and standard lathe screen set.
You need a good clean Index pulse, and to enable spindle averaging.

Cheers
Roger

Be of good cheer. Threading does work very well. I use it all the time.

I am running W7, M3...062, with an ESS, and standard lathe screen set.
You need a good clean Index pulse, and to enable spindle averaging.

Cheers
Roger

I also use it all of the time....or should I say USED to use it all of the time.  I have a bunch of production parts that have threading cycles which have worked great for a long time.

I'm really disappointed that nobody can help understand why it stopped working.  Everything else is checking out ok (I double checked the indexing and that looks fine).  It is really frustrating because a CNC lathe that cannot make a thread is not very useful.  Kind of running out of ideas here!

Hi Jamie

OK, so it used to work but now it does not. So, something has changed. There are three possibilities:

1) The Mach XML file has become corrupted and needs fixing. Sadly, that seems to happen a fair bit. But it can be fixed by going through every (damn) screen and checking every detail. Many of us have done that.

2) The hardware has suffered a failure somewhere. This may only be a partial failure. only operating some of the time. It may only happen at speed: Rc filtering perhaps. For this you will probably need an oscilloscope. Sorry about that.

3) The 3rd possibility is a funny one: an interaction between HW and SW. I found that my displayed spindle speed was fine up to ~2200 RPM, then it went haywire. But the HW signals going into the ESS were fine at all speeds. It turned out that at 2200 RPM the Index pulse width dropped below the Mach SW filtering threshold, and so Mach was not seeing the Index pulses any more. I cancelled the filtering.

Cheers

Hi Jamie

OK, so it used to work but now it does not. So, something has changed. There are three possibilities:

1) The Mach XML file has become corrupted and needs fixing. Sadly, that seems to happen a fair bit. But it can be fixed by going through every (damn) screen and checking every detail. Many of us have done that.

2) The hardware has suffered a failure somewhere. This may only be a partial failure. only operating some of the time. It may only happen at speed: Rc filtering perhaps. For this you will probably need an oscilloscope. Sorry about that.

3) The 3rd possibility is a funny one: an interaction between HW and SW. I found that my displayed spindle speed was fine up to ~2200 RPM, then it went haywire. But the HW signals going into the ESS were fine at all speeds. It turned out that at 2200 RPM the Index pulse width dropped below the Mach SW filtering threshold, and so Mach was not seeing the Index pulses any more. I cancelled the filtering.

Cheers

Thanks, I'll take another look and see what happens.

I have not worked out why, but it seems that having anything to do with a CNC means either that you pay large amounts to a service technician OR you have to have a DVM and an oscilloscope, at a minimum. That's life.

But there is some satisfaction when you find the bug and fix it. :-)

Cheers

I have not worked out why, but it seems that having anything to do with a CNC means either that you pay large amounts to a service technician OR you have to have a DVM and an oscilloscope, at a minimum. That's life.

But there is some satisfaction when you find the bug and fix it. :-)

Cheers

I have a scope so that is a not a problem.  What I can't figure out is why everything seems ok - one day it worked and the next it didn't without any apparent cause.  I fear that I'm going to have to dig really deep which could take a lot of time.