Messages

4811

General Mach Discussion / Re: Can you wire a proximity sensor directly to a motion control board like a pokeys

« on: April 13, 2018, 05:43:34 PM »

Hi,
most of the commonly available proximity sensors have a voltage input requirement of 6-36V.

The limit/home switch inputs of your 57CNC are 5V and PoKeys specify a maximum of 5.5V.

I would want to be 100% that the proximity switches didn't blow the 57CNC. At the very least I would have a current limiting resistor in series with the 57CNC input pin and the proximity sensor and a 4.7V Zener diode to earth
and thereby limit the voltage presented to the input.

Another alternative would be to use a transistor, either a BJT or a MOSFET, either would be a few cents each and would provide some measure of isolation between the proximity sensor and the input pin.

Craig

4812

G-Code, CAD, and CAM discussions / Re: fusion 360 cam gcode to mach3 scale change

« on: April 13, 2018, 05:02:58 PM »

Hi,
could it be that your machine is interpreting the Gcode as inches whereas the code is actually in mm. This would result in the part being 25.4 times to big.

Read up on G20/G21. They cause Mach to interpret Gcode in inches or mm respectively.

Craig

4813

General Mach Discussion / Re: Setting new cnc router build to achieve max servo speed

« on: April 13, 2018, 04:56:39 PM »

Hi,
could that specification be a misprint?

https://www.automationtechnologiesinc.com/docs/manuals/KL34-180-90.pdf

Craig

4814

General Mach Discussion / Re: Setting new cnc router build to achieve max servo speed

« on: April 13, 2018, 04:34:20 PM »

Hi,
I just Googled your servo and found the KL34-180-90 and it has a Kv 0f 18.95 V/100 rpm. Can you confirm?

To make it run at 3000 rpm would require a servo drive with an open circuit voltage of:

18.95 X (3000/100)=568.5 V

Without PFC boost or a transformer that would require three phase power of 415V. Do you have three phase?
Do you have a servo drive capable of that voltage?

Do you actually own these servos or are you going to buy them?
If you don't I urge you consider AC servos, they are a more modern solution and work out cheaper than old style DC servos.

Craig

4815

General Mach Discussion / Re: Spindle servo on lathe is 10% fast

« on: April 13, 2018, 03:07:22 AM »

Hi,
that story rather puts CNCing as a hobby or a job into perspective. I hope your recovery progresses, it sounds that while you have come a long way it is still a long road.

I play Bridge recreationally and its a competetive game and I have a phrase, 'Bridge, its only a game...its not life or death....its more important than that!' While the humour is fun
its not in fact 'life or death'.

Using a modern AC servo as a spindle  puts you at the leading edge performance. You will, once you get it sorted and provided you remain within the torque limits of the servo
have a spindle with speed/position stability the envy of many.

Craig

4816

General Mach Discussion / Re: Spindle servo on lathe is 10% fast

« on: April 13, 2018, 12:48:40 AM »

Hi,
kool. Next question is why you pissed around fitting an index pulse generator when you could use a digital ouput based on the inbuilt
encoder?

Craig

4817

General Mach Discussion / Re: Spindle servo on lathe is 10% fast

« on: April 13, 2018, 12:40:18 AM »

Hi,
another thought is that maybe the 'electronic gearing' of the servo is not 1:1.

What is the current value of parameters PA12 and PA13?

Craig

4818

General Mach Discussion / Re: Spindle servo on lathe is 10% fast

« on: April 13, 2018, 12:23:00 AM »

Hi,
well that should mean that the speed is spot on, that is to say that every time the ESS pulses the servo should advance
one encoder count. The ESS should be producing pulses at exactly (within 0.1%) of what is commanded by Mach.

Do you have an oscilloscope or a frequency counter? I would be prepared to guess that Mach an the ESS are working
correctly but your index pulse generator is somehow wrong.

I would issue a speed command in Mach such that the step pulse to the driver was 5kHz and measure it as accurately as you
can.

Craig

4819

Mach4 General Discussion / Re: My File is running real slow

« on: April 12, 2018, 07:50:57 PM »

Hi,
when I start calculating things I always revert to radians. Like wise I tend to do all engineering and scientific
calculations in metric (MKS), even though I tend to present the result of the calculations in imperial like lbf
and pounds per square inch.

Craig

4820

Mach4 General Discussion / Re: My File is running real slow

« on: April 12, 2018, 07:41:34 PM »

Hi,
now get real and use radians per second like a proper physicist...

6.282 radians/sec =60 rpm LOL

Jokes aside, radians have value when you consider that the ratio of the circumference of the arc traversed to the cricumference
of the entire  circle is the angle in RADIANS.

Thus for OPs problem 1 inch on the surface of his 4 inch diameter job:
1/4 x pi=0.07959 radians which at 57 degree per radian=4.5 degrees.

So radians, an angular measurement has a direct correlation to surface circumference, its just the numbers look a bit weird.
Like 360 degrees = 2 x pi, 90 degrees= pi/2 and so on.

Craig