Hello Chaps,
I am having a lot of trouble getting repeatable sizes on my lathe x axis.

I dont know how to tell if steps are being lost but i suspect that this may be the
cause of the problem.

If steps are being lost can it be heard or how can i tell. When I do the steploss
check ( backwards and forwards 50 times to return to a zero point) it seems to be
ok - but i do not know what happens when the machine is taking a cut.

I think I might have found a possible cause.

when I mounted the stepper motors i didnt understand the torque characteristics
of the stepper. In my wisdom I decided to gear down by means of toothed belts
by a ratio of about 1 to 3 - the smaller pulley being on the motor. In theory i thought
that the gearing would prevent overloading of the stepper.
But now it has been pointed out to me that the stepper is at its most powerful
at very low revs and there is a decided reduction in torgue as very speed increases.

Am I correct in assuming that if I reduce the gearing and bring it 1 to 1 that i will
be better utilizing the torque characteristic of the motor?
My steps per mm will be much less and the motor would then be running at lower
revs at which the motor is delivering its maximum torque?.
At low revs  is there any chance of overheating the motor?

Best regards and thank you

Fred Evans



  

Fred,
As the stepper rpm increases the torque decreases. What you want is power which is speed x torque.
So the power part of the curve is somewhat subjective / over a usefull or wanted range of rpm.
The current needs to be limited to the stepper as too much current will overheat the motor and can cause damage.
Increasing voltage can increase stepper rpm. All steppers are not the same in rating and preformance.
Usualy a curve is supplied with the stepper which shows a graph of torque available to rpm at some voltage. Very seldom do you see multiple graphs at different voltage ratings. The curve gives you a starting range for tuning, but, one still needs to adjust settings for max reliability. ( just a comment....if you spin the steppers shaft at various speeds it will generate a voltage
and interesting enough that voltage at some point will come to a max no matter how fast you turn it)

On my Sherline 400 approx in oz are used and on the small Atlas lathe 500 in oz are used. The 500 's have that rating at 3A & 36 volts. They can be on / idle  all night and never get hot and never get hot when in use, warn in use , but not hot.

I have mine such that the gear raito is about 2.5:1 with the smaller pulley on the stepper. So the stepper speed is increased
at a loss of some torque but the torque delivered to the screw is increased by the pulley ratio. So the axis delivered power is somewhat max. Yet for the X axis it is directly driven, but, the screw a good pitch such that adequate resolution
is avialable. It's all a tradeoff in the end.........

I would suggest you go to the Gecko site as they some very good info on steppers. Also you may want to look at the
thread I did  about the Atlas lathe conversion. If I recall correctly, info was posted about motor torque and anticipated preformance of the lathe.  

A stepper can skip steps and you may never know it, but, usualy you can hear the skipped steps,  if loss of stepping is significant the axis behavoir will be  ........one of where it goes nobody knows...... control is lost or return to a position is no longer correct.

Thus for a stepper system RELIABLITY is important. Better to run the axis slower and accurately with repeatablity than
high speed  / feedrate.

So, you the operator, learns what YOUR lathe is capable fo doing for different maching tasks, etc.

How much torque do you need.........well , enough to power the axis , some spare torque for reliabily, enough for
a resonable  acceleration, and finaly enough to do the machining task. There are programs that can calculate all that,
but, stiil experince and judgement will be used by the  manufacturer in the final sizing.

Done rambling.....hope this helped,

RICH

 

Thank you rich for this info-- My steppers are Kelling i know not what but i will try and
identify then first then i will go through your advice in detail.
I have not made any changes to the electronics so i presume that i am still using
the voltage that the breakout board is putting out. I am presuming that this voltage
will be a constant?? Can i take a voltage reading by connecting a meter  to the stepper
wires and which wires must i use ?

Hi Fred I had a similar problem a couple of years ago and not being an electronic man I reverted to my old mechanical fitting skills. This was on a CNC mill but the same principal applies.
Replace the tool with a dial indicator, write a program with a several repeatable square loop set of moves and include a rapid move.
Reference a tool with a zero on the dial indicator against a fixed point (the component or any fixed point on the machine) run the program and note the dial indicator reference against the fixed point every time the cycle touches that point. if the figure is the same you are not loosing any steps. If they are not then run the program again with a slower feed rate, keep repeating until the set reference point maintains its zero.
I found I was running my rapid travers to fast so loosing steps, slowed down the rapid and problem solved.
Just for reference one of my slide oil pipes was blocked which made the X axis tight, another way of loosing steps.
I know this is a mechanical engineer method of checking but at least it should rule out missed steps, then its over to the rest of the forum to make suggestions.
Also the load on the tool when cutting could force the program to loose steps, but mainly its the rapid traverse that causes lost steps
Hope this helps
Good luck
Jim

thanks Jim-- i have done a step loss  program but didnt mix g0 and g1 like you suggested so i will try that

You mention "slowing down the rapid" > Am i correct in assuming that this is done by slowing down
the velocity slider in motor tuning dropdown??

thanks

fred
 

Hi Fred, try both however my steps were lost on the acceleration time to get to the G00 rapid feed rate. its something that make little difference to me as I'm not doing production work so time doesn't matter. The motor tries to get to max feed rate instantly so looses step, use the slider in the motor set up to give it more time. We are only talking milli-seconds each time you make a move.
I was asking the machine slides hence the motors  to move quicker than the motors could cope with. Does you manual give a maximum rapid (G00) as MACH will let you put in whatever you like so it could be over the machines recommended limit.
Hope this helps.
Jim

I've been following this discussion as a newbie to CNC and have experienced everything in this thread. As I try to dial in the ideal mix of speed and torque, I find myself vacillating between different micro-step ratings on my controller to drive my NEMA 17 steppers (its a home-built micro-mill) with varying results. So I have to ask, is it always better to drive the steppers at the highest micro-stepping setting to maximize torque?

For example, I can set the microstepping to 1/8 or 1/16 giving me 1600 steps per rev or 3200 steps per rev respectively. But I also am trying to minimize the stepping voltage setting so I don't burn up the motors. At 25% voltage, it seems to be happy running 1/8 micro stepping, but at 1/16 micro stepping, it just buzzes and I have to switch the voltage to 50% to get it to move. I thought the reverse would be true i.e. 1/8 would require the 50% voltage and the 1/16 would run fine at 25%. Anyway, it sounds like you have to experiment to see what works best. But with an array of 4-5 variables to work with, the combinations of such is exhausting! Any advice or observations welcomed! THX

Greater than 1/10 microsteps is somewhat of a waste. SO would say 1/8 should be fine. Microstepping provides for smoother stepping but is drive dependant. As the steps increase you get get better resolution but just assume that you will actualy have that resolution, manny times unlikely. You start by finding where you start to skip skips at some velocity, then say,decrease velocity by 30%. Then increase acceleration and see what happens and again reduce when it starts to skip.

It's not exhausting, just need to do it in an orderly fashion.

RICH

Thanks for the sanity check. I'll work with the 1/8 setting to see what kind of results I can get. As of today, I was able to get all three axis dialed in and at least drawing (pen instead of cutting head) in units that match my design. However, I noticed that when it encounters curves, like a radius, the X-axis goes limp and skips steps like crazy while the Y axis is fine even though they are setuo identically so the curve looks more like a straight line! When I increase the voltage at the controller for the X axis, the thing hums (even at rest) and runs hotter than the former setting, but it does complete the arcs as expected. The motors are rated at 1.7A 12VDC and the controller has a setting of 50% of max 3A which would mean I'm only running these things at 1.5A. Should they be making that kind of noise and running that hot? THX