4 Attachment(s)
Re: Problem: Losing Steps?
I've attached some photos.
I'll check voltage at the driver tonight when I head out to the shop.
Re: Problem: Losing Steps?
Hi,
Pacific Scientific make great quality motors.....but there seems to be a discrepency between two of your photos.
One shows the name plate of the motor which shows power of 71W whereas the table shows the axis motors having a power of 230W?? Maybe they are referring to the
combined power of the three motors??
Either way I would be guided by the motor nameplate. Having said that the nameplate says 1500rpm, which is very fast for a stepper, so maybe the Yanks have come up with
a BS number, they are not above BS!
The eight wires mean that the windings are arranged in two phases with two coils in each phase, ergo eight wires. This arrangement allows you tow wire them in bipolar series,
bipolar parallel or bipolar centre tapped. This arrangement allows the greatest flexibility and is favored by manufacturers.
In order to get the best possible torque and best possible speed you must wire your motors in bipolar parallel. Can you find any more info about the motors, particularly the
wire colour assignment.....that would be easiest, you can measure it and do it empirically but it would be hard to step you through it over the internet unless you are already familiar
with the process and have the gear, a signal generator and an oscilloscope?.
If you can confirm that the motors are wired in bipolar parallel then you have the max the motors can produce. If there is no mechanical or an issue with the drivers then you'd need
to replace the steppers with more powerful units. If however they are NOT wired in bipolar parallel but say bipolar series, then correcting that will increase the torque and speed by
a factor of four. Would that be enough?
Craig
Re: Problem: Losing Steps?
Thanks for the description. Maybe my issue is this axis is wired differently, hence the issue on one axis.
I'll dig around and verify a few things you all have suggested, driver voltage and the proper wiring diagram.
Thank you.
Also to elaborate, I may have made things confusing. The spec sheet is the original specs of the machine, it was converted to these steppers and drivers before I purchased it.
1 Attachment(s)
Re: Problem: Losing Steps?
Based on this diagram, the motor is wired in bipolar parallel. I'll verify the drive voltage when I can tonight.
I really appreciate the quick responses. This machine is a work horse for me when I receive orders. I don't want to keep it down for too long.
Also with regards to the 1500 rpm. I've wondered if perhaps I'm over spinning them. As there is a belt driven reduction built into the machine. The ball screw has a larger pulley than the motor.
I've lowered my acceleration and deceleration settings but that hasn't seemed to help. Unless maybe I didn't go low enough.
Re: Problem: Losing Steps?
Confirming I have 51.5V at the stepper driver, even while Y axis is homing. I'm unable to verify at a higher speed. Could that be a potential issue ?
Edited: had someone hold the y axis button, when the axis is moving at full speed voltage drops to 50.5v.
Maybe I should try reconnecting x axis and verifying voltage while it's moving ?
Drivers are pacific scientific 6410.
Re: Problem: Losing Steps?
Hi,
Quote:
had someone hold the y axis button, when the axis is moving at full speed voltage drops to 50.5v.
That voltage drop is inconsequential. By all means hook up the X axis and compare.
A 50V power supply surprises me, its lower, quite a bit lower than I would have expected. With the '1500rpm' on the label I would have expected the voltage to be north of 100VDC.
Good low inductance steppers (less than 1.5mH) could be expected to retain 35% of the holding torque at 1000rpm with a 60-80VDC driver. A lesser motor, say 4mH inductance,
would be lucky to have 25% of its holding torque at 500 rpm, let alone 1000 rpm, and at 1000rpm I would expect less than 10%.
Taking the two considerations in mind, ie high voltage equates to high speed and low inductance equates to highspeed leaves a bit confused about what steppers you have,
are they some 'super low inductance whizz bangs that defy physics' or what?????
I suspect you are right, this machine was equipped with servos and clearly steppers are never going to match them no matter how hard you try.
What I would recommend is calculate the gear reduction ans set the max velocity such that the stepper is doing 500rpm at max velocity. Set the acceleration to 10% of that.
Then, leaving acceleration alone, increase the max velocity until it stalls or starts missing steps. That is a better and more practical measure of what your steppers can achieve in your machine.
Once you have established that limit, reduce the max velocity to 75% of what you have found and start increasing the acceleration, until it too starts missing/stalling.
The aim is to have as high acceleration as possible, it promotes the best and fastest toolpath following, and is usually more important to cycle time than max velocity, strange as it
may sound. Most toolpaths spend more time accelerating than at max velocity, so acceleration is favoured over max velocity.
Craig
Re: Problem: Losing Steps?
Hi,
Quote:
Motor max rpm is 1500.
Dubious, especially with only 50VDC to urge them along, if you had 150VDC to push them then yes I would say 1500rpm sounds possible.
Craig
Re: Problem: Losing Steps?
Hi,
Quote:
From reading the pacific scientific literature the voltage for these drivers is 25-75VDC, same literature for motors shows torque charts with a max of 3k rpm.
I've just been reading/digesting the charts you have posted and your stepper P22NRFB-LNN-NS-00 has a torque of only 70oz.in at 5000 full steps/sec, ie 1500rpm, and
less than 40oz.in at 10000 full steps/sec. You steppers will miss steps or stall LONG before you get to those speeds.
Voltage applied to a stepper is about the driver, not the stepper.
I used 5 phase Vexta steppers on my previous mini-mill and matching Vexta drivers, and they applied over 150VDC and I ran them for years at 2400rpm,
and could tune then as high as 3000rpm, if somewhat less reliably. If you want steppers to go fast apply the highest voltage driver you can find.
If you want to go faster than that get AC servos. I use 750W Delta AC servos on my new mill, direct driving 5mm pitch screws to 5000rpm for 25m/min rapids. Its
scary fast, so I've deliberately detuned then to 15m/min. This machine weighs 800kg, so when it accelerates you bloody well know about it! Servos eat steppers
for breakfast.....I'll never go back to them.
Craig
Re: Problem: Losing Steps?
Hi,
if you are interested in servos look into Delta (Taiwanese made in China) and DMM (Canadian made in China). They are both good quality
with high resolution encoders, good backup, good documentation and most importantly free set-up and tuning software at very fair prices.
There are many hundreds of parameters that need to be programmed into a modern servo drive. If you try to program them by pushing buttons like a
microwave, you'll be there forever and STILL make a mistake.
https://www.fasttobuy.com/Supply-asdab2-series_c498
There are even cheaper Chinese made brands, Liuchan and ToAuto, come to mind. they are very well priced and attractive as a result, but the documentation is poor
and no set-up and tuning software. I have no reason to doubt their quality or performance, its just hard to set them up, especially if this is your first foray into servos.
Modern AC servos from all manufacturers tend to be very similar in terms of features and programming. If you've done one you've done them all. Delta and DMM
are typically 25% more than Liuchan and ToAuto and well worth the extra.
Many servo-newcomers favor Clearpath servos, and they are very good quality and industry leading after sales support but they are expensive for what you get.
Firstly Clearpaths have as standard a paltry 800 count/rev encoder, one, only one, digital output, no auxiliary outputs and no analog inputs.
A 400 W (actually 397W) Clearpath servo is $517USD while I paid $435 for a Delta B2 servo (160,000 count/rev)...that's twice the power for less.
Craig