Roger,

Now, translate that to the hobbyist who has a small 500 W DC motor on the spindle and some sort of simple power supply to drive it. Can he make a nice clean M45x0.5 mm thread on that? (Close to a camera filter thread I think.)

YES, and can probably do a 0-80 thread which is even finer, ( on my wimpy punny Sherline for example)
BUT  
1. How accurate is the lathe system?
2. Is setup appropriate for what will be done?
3. Is practical knowledge and epxerience being applied to the task?

So the controller can do it's function but the user in many cases ( plug and play comes to mind)
just lacks 1,2,3 above and you get "repeated complaints that threading is not working". BTW,
people don't read any more ( we live in a visual world), it takes too long, require instant gratification, want for free, etc ,etc.  Most folks are just happy that the nut goes on. Other folks or folks in industry  that specify threads and do analysis in design of things have a different understanding of threading appication.

A walk down memory lane is appropriate:
Threading was "broke / didn't work some 10 years ago" and it got fixed. A lot of effort went into
fixing the single point threading it has not changed since then. So it is what it is and not going to get updated to provide for electronic gearing, encoder use  as the internal code must be changed and
developement procedures will be required. The lathe has taken a back seat and will probably continue
as such.

As the PP continues to go away you have the changes in hardware and external motion controllers addressing the issue. If it's so easy to implement how come it's taken 10 years? Just note that
we need to keep it all in perspective ie; high end controllers etc. compared to low end stuff.
Mach4 is surely going to provide for a higher control end and the burdon falls on the external motion
developers to implement the appropriate functionality. Only time will tell what happens in development
and suggest one dosen't hold their breath.

The above said, if  there are users that can help make a step change that would be great.
The pay is great just don't spend it all in one place!

Enough rambling ........ keep up the effort guys,

RICH

Hi RICH,

people don't read any more

ain't that the truth!

Craig

Hi Roger,
RICH raises an excellent point about people wanting instant gratification without any work.

Just think about the number of posts on this forum alone where someone has bought some gear, spent
half an hour assigning pins etc on a 'by guess or by God' basis and then straight onto the forum asking
questions...grumpily. To suggest that they read and then reread the installation manual is ignored 90%
of the time.

Any manufacturer of technical software, Artsoft and Warp9 being the two in focus in this thread, are
really 'sticking their neck in a noose' because so many customers will complain/whinge/demand satisfaction
when it doesn't work straight out of the box and forget reading the manual, the manual is just a source
of paper for wrapping rubbish!

Craig

Hey Graig,
I don't read anymore ..... just write or type what the boss tells me. 
I leave the hard part to my better half.   


RICH

We have all been newbies and have an understanding about what a broad spectrum CNC covers.
Fortunately the members in here need to be applauded for assisting users.

A friend hates to ask me questions, he always starts by saying  I know, I know....RTFM. 

I believe he can read but can't write.  Me come from Mars....him Jupiter! 

RICH

How true about manuals, sigh. Except that they can't be used for wrapping paper these days: they are just electrons.

Back in the days of a certain mainframe, the reply to RTFM was WFM: Which Fine Manual? The shelf of grey 3-ring binders was over a metre long!

Actually, my thread example was a very poor one. An M45x0.5 thread is not that hard to do, because the tip engagement is never large. Try instead an M30x1.5 thread, where the tip engagement can suddenly go from 0.2 mm length to 1.0 mm length. In case you are wondering, yes, recent experiences! The results were horrible.

No, we cannot expect any changes inside Mach3, and I don't think Mach4 can do any better. The PID control has to go into the external engine. Many older-generation servo drives come with that embedded: you bought the motor AND the controller as a package, and there were two thick cables between them (power and encoder). But those were incompatible between companies and hence very expensive. We are now transitioning to much more modern and lower cost versions where the functions have been separated and there is price competition. The digital Step/Dir + PWM interface has become the 'Standard'.

My vision is that the code in the existing ESS could be upgraded to take encoder signals from the spindle motor and to PID control the motor to the accuracy of the encoder (via PWM). That would be a pure firmware upgrade in the ESS (you load the firmwarte every time Mach starts), plus the allocation of 2 pins on, say, port 2, for the encoder. This means you would have to fit some sort of encoder to the spindle motor, but that is not that hard to do. At the very least, you already have a primiitve encoder there to give the Index pulse. Just add slots!

Technically, this is possible. The Q is whether the FPGA in the ESS has enough spare gates to implement the high-speed code to do it. That is what Andy at Warp9 is working on.

Could the average novice handle this? I think so - because the ESS could come set up with some conservative default values for the PID control. He would need to enter ONE number into the ESS Config: the number of slots in the encoder.  If he can't do that ... he should try gardening.

Cheers
Roger


Cheers
Roger

I tried gardening Roger, which is why I changed to something easy like CNC instead. Thanks Craig for your input re the encoder pulses from the servo drive, but I am now down the path of making an improved version of my old spindle based encoder, so I'll keep going....
The thread has devolved into an interesting discussion on the possibilities based on a hardware encoder  on the spindle of the lathe, and I am enjoying it immensely. I bought the servos that I am using about 6 years ago and it was seeing a you-tube video showing someone cutting a hex onto a spinning shaft with their lathe, using parametric Gcode that made me extract my digit.
Cheers
Dave

Hi David,
yes Roger, RICH and I have rather hijacked your thread but as you say it is interesting and  least loosely related to your post.

Cutting a hex into a spinning shaft is very impressive, would require untold acceleration in X axis.

Under that circumstance the spindle is no longer a spindle according to some definitions but becomes a C axis. Like any axis to be worthy of its name
it has to be capable of co-ordinated motion. Given that you already have step/dir control of your spindle you can claim 'axis' status for it and so you could
as it stands replicate what you have seen.

As I've said before you have made a substantial investment in your spindle drive and the capability it confers is impressive and exceeds a lot of pro gear
and 99.9% of hobby gear. Kick back and enjoy.

Craig

Hijack away. Here is my goal. "https://www.youtube.com/watch?v=DajrTlI8IWA" Imust admit I have no idea how I will acheive it, but you never know.
Cheers
Dave

Hi DMBGO

Dead easy if you use a rotary table as your chuck. You could do it 'lathe' style, or you could just mill it. You could even do it on the mill table with the rod pointing up if it is not too long.
Been there, done that, several ways, already.

Cheers
Roger