[Mariss Freimanis]

Hi,

This is a progress report on new developments in the "unstallable stepper project".

The goal of the "unstallable stepper project" is very simple. Imagine you have built / bought a 4-foot by 8-foot gantry-style router. You are doing 3D (x,y,z) routing with it when your drunken brother-in-law is over and gets the idea to see what the ride is like if he were sitting on the gantry. The result would be ordinary steppers or servos would immediately stall / fault as soon as the sobriety-challanged one clambered aboard.

Now imagine the motors simply slowed down or even stopped due to the "idiot aboard" overload but they never leave the 3D path in progress. Dislodge the load and the motors pickup to their original speeds and finish the work as if nothing happened. Not a gouge or mark on the finished work afterwards.

So how to get from here to there? A lot of parts that all have to come together is how.

Part 1A, Stepper Servo: You see BLDC servos, AC servos, PM DC brush-motor servos. You don't see step motor servos (Vexta Alpha Step and the like, step drives with monitoring encoders don't count because they are half-assed solutions). Why? Because technically it is very hard to servo-tame a step motor. Very, very hard in fact.

Begs the question; why bother what with all the other servo choices? The answer is steppers have a unique speed-torque curve that makes them perfect for 2-mode applications. Lots of torque at low speed work feed-rates and only enough torque for high-speed rapids. Nothing is wasted.

Part 1B, Stepper Servo: Open-loop step motor systems have to be seriously derated. They are running open-loop after all. Closed-loop, a mild-mannered Clark Kent type motor becomes Superman. It literally jumps on the bench from acceleration reaction torque on its way to 15,000 RPM. This from a NEMA-23 2A motor with a 24VDC power supply. It takes 0.03 seconds to go from a standstill to 3,000 RPM. 0.42 seconds later and it's at 15,000 RPM.

The biggest differece? It sets its own optimal rate of acceleration, it cannot resonate, it always gives 100% but cannot cross the line and stall while trying to give 101%. That's built-in.

Part 2, The G-Rex factor: Your 300-lb brother-in-law on the gantry requires way past 100% from the motors. They won't deliver more than 100% so something has to yeild. What has to give is the step pulse rate and that's where the G-Rex enters the equation.

The G-Rex from the get go was designed as a vector based pulse engine. What that means is the G-Rex generates axis step pulse frequency ratios naturally. Now imagine 3 motor axis tracing out a 3D path; the motor speeds are at fixed ratios along a 3D line. One or more servo steppers report an impending overload to the G-Rex and it responds by slowing down the vector velocity step pulse rate.

The motors respond by backing up their speed-torque curves to find more "grunt"; what was a near overload at one speed becomes a manegable load at a lower speed. Only step motors have this beneficial speed-torque curve. This works all the way down to zero velocity.

No matter what your brother-in-law does, he cannot disturb the router from its programmed path. He may be able to stop it but he cannot move it off of the intended path. That is the goal for the unstallable step motor project.

Part 1 is now progressing rapidly and it's consuming the effort currently being applied. Part 2 should follow after part 1 is finished.

Mariss

[epineh]

Interesting project Mariss, wish you all the best with it. I have been thinking for a while of getting "servo-ish" behaviour from steppers. My thoughts were to use a microcontroller to act as a buffer between the step/dir source (PC) and the drives, with the encoders being read by the micro, and simply add in the missed steps within the motion profile as soon as possible, not perfect but possibly acceptable for routing jobs.

I simply don't have the time and abilities YET to achieve any more than the theory, and this is a far cry from real world application. That and the sheer depression from being told continuously that it simply can't be done with steppers (chair).

I look forward to the finished product.

And as for the drunken brother in law, he shouldn't be concerned if the router injures him, he is a dead man as soon as he gets off the machine...

Russell.

[Mariss Freimanis]

I wasted 2 years trying to solve the problem using Step and Direction inputs in conjuntion with encoder feedback. I finally realized it could never work so I took a different approach.

Mariss

[turmite]

Mariss do you have an idea as to a timeframe?

Mike

[Mariss Freimanis]

It will be much sooner now that have I abandoned the old approach. I avoid using timeframes on something like this because there are still many unknowns in the circuit design. Even one serious design "pothole" can take a long time to resolve.

Mariss

[pastera]

It will be much sooner now that have I abandoned the old approach. I avoid using timeframes on something like this because there are still many unknowns in the circuit design. Even one serious design "pothole" can take a long time to resolve.

Mariss

I just went through a fun one - found six errors in the data sheet for a new chip the hard way! Ever try to field mod a 0.4mm QFP.

All that and doing work for someone who thinks a new product only needs one shot at the PWB to be perfect.

The tech in my lab swears at me every time she looks at these boards and remembers it's her job to play with these things once they are in production.

Aaorn

[Richards]

Mariss,
It looks like your new design is going to be a winner; but, you're causing me some personal grief. Since my stepper motors won't need encoders, what can I mount to the rear motor shaft? I've been thinking of using the propellers from my Beanie Caps but I'm not sure my wife will let me wear a Beanie Cap in public without a propeller.

The new stepper drivers and G100 combo will certainly change the way we think about stepper motors.

-Mike Richards

[sdantonio]

Mariss,
It looks like your new design is going to be a winner; but, you're causing me some personal grief. Since my stepper motors won't need encoders, what can I mount to the rear motor shaft? I've been thinking of using the propellers from my Beanie Caps but I'm not sure my wife will let me wear a Beanie Cap in public without a propeller.

The new stepper drivers and G100 combo will certainly change the way we think about stepper motors.

-Mike Richards

Hi Mike,

Sounds like you onto something here. self cooling steppers. That solves the problem of them overheating. If I were you I would run to the nearest patent office and patent the beanie cap propeller

[turmite]

Hi Mariss,

Now don't take this the wrong way since it isn't meant to be bad. I was asking for a general time frame, i.e. months, years etc. I've been patiently waiting since you first mentioned this concept some time ago, a couple of years I think, so it's not like I'm going anywhere! Trust me when I tell you I know about the potholes, just a different road for me than you.

Mike

It will be much sooner now that have I abandoned the old approach. I avoid using timeframes on something like this because there are still many unknowns in the circuit design. Even one serious design "pothole" can take a long time to resolve.

Mariss

[kreutz]

... I avoid using timeframes on something like this because there are still many unknowns in the circuit design. Even one serious design "pothole" can take a long time to resolve.

Mariss

Agree 100%

Regards,

Kreutz.

[Mariss Freimanis]

No offence taken.:-) The servo is in the design phase. The step motor is running in a self-commutating mode right now which means it is an uncontrolled (non-sevoed) high pole-count brushless DC motor and drive. The test circuit performance reveals what the potential closed-loop performance will be and it has me salivating.

Like Pavlov's dogs, that concentrates my attention entirely on developing this new circuit because I can see what it will be when it's done. It will be a very smooth and trouble-free project if it's finished in 6 to 12 months from now.

When it's done, I expect the new drive to be the same size and also cost about the same as a G203V.

Mariss

[turmite]

Mariss will it be designed specifically to run with a G100? Ought to make things smoooooooth!

[salukikev]

I wasted 2 years trying to solve the problem using Step and Direction inputs in conjuntion with encoder feedback. I finally realized it could never work so I took a different approach.

Mariss

Quick (noob?) question: Could you summarize the conclusion that you arrived at there? Why would it never work? Do you mean only in the context of the geckodrive controller-- so it could potentially work via software accomodations?

I guess since I just retrofit my mill with steppers, and my mill already has a DRO, I was, in the back of my head thinking it would be nice to simply plug in the encoder data already available and have the steppers be aware of their position through software.

Thanks for any clarity on that one.
-kevin

[rodzilla]

What is the status of this project any further breakthru's i can't wait for this to hit the market i will be first in line to purchase

[Mariss Freimanis]

It could never work as a servo if only step and direction are the only available control variables. Encoder phase to winding excitation timing is everything.

Progress is being made. Just recently (a) motor current is now a function of load just like in any servomotor. No motor load equals no motor current equals no motor heating. (b), Servo stiffness is set by the encoder (+/- 0.09 degrees) versus a step motor's natural +/- 1.8 degrees stiffness. Servo stiffness is the angular rotation of the motor as load goes from zero to maximum torque.

Mariss

[David_Web]

You could plot a vector from pulse/dir data but that would mean a deley between sending out the pulses and reciveing a vector. If the system needs to slow down you will quickly run out of buffer unless you have the entire program buffered. And at that point it's better to spit out vector data in the first place.

@salukikev: Depending on the type of encoder and DRO you might be able to input that to the G100.

This means that I might have use for some of the steppers that I have that is already fitted with an encoder.

Will this system work with mach? And how does it affect jogging?
Homeing? Probing?

Will it work with any driver that takes step/dir or is this a new driver that you are creating?

[DonW]

Wow... from my ignorant point of view, (and I mean that I understand almost nothing about motors) it sounds like you're coming up with a new drive that will allow servo-esque performance from a stepper? That would be a great thing, eh?
I'm about to invest in a bunch of G203V's and a G100 for the cnc router I'm building, and now you're saying I might want to wait 6 months to a year and get the new drives instead? Are there going to be special motors made for this too, or just the drives? Yeah, I know I sound dumber than a fence post...

Sounds really great!
Er, except that I'll have spent $500 on the 203's already....will you take trade-ins?

[Eson]

So there will be a new drive for all these features? But will these features also be inplanted in the g-rex? I dont quite understand

[motomitch1]

It's not worth the effort to dress a sparrow up in peackock's plumage. If you want a peacock, start with one.

[ger21]

He's trying to create a "super peacock". Read the goals in the first post. This system will keep on cutting where a servo drive would have faulted.