Last One - 4'x6' Steel, Epoxy and ClearPath servo's

[1Jumper10]

Did a mock up to double check the riser height. The design goal is to reach the cutting bed with a 1" cutting bit. I decided to move the spindle up in it's bracket to shorten the gantry risers some. They seemed too long how I had them.

Sent from my Pixel XL using Tapatalk

[mactec54]

Thanks.
Just the one servo is all I have at the moment. I had to have a Nema 23 servo on the linear actuator. The ballscrew was there already and so those factors made it easy to spec. It's the high torque "D" series like I used before but this one is a 3 stack servo. My friends is a 2 stack.

Sent from my Pixel XL using Tapatalk

It's strange that you quote your motors as servo's having a stack value, this term ( stack ) is only used for stepper motors, you may have a Hybrid Stepper motor, but they are not servo motors, if they are made up of different stack numbers

[1Jumper10]

I dont know exact terminology and what term is properly used where so I'm sorry if I misspoke. The Teknic engineer I spoke with referred to their different versions of the same models as "stacks". In the pic, you can see the rotor of this model has two magnets. A 3 "stack" version of this motor would have 3 rotor magnets. The NEMA 23 servo I used on the Z axis of my friends router was a 2 magnet rotor version. This router Z-axis, I used the same version NEMA 23 servo but with 2 magnets on the rotor instead of 3.

[mactec54]

I dont know exact terminology and what term is properly used where so I'm sorry if I misspoke. The Teknic engineer I spoke with referred to their different versions of the same models as "stacks". In the pic, you can see the rotor of this model has two magnets. A 3 "stack" version of this motor would have 3 rotor magnets. The NEMA 23 servo I used on the Z axis of my friends router was a 2 magnet rotor version. This router Z-axis, I used the same version NEMA 23 servo but with 2 magnets on the rotor instead of 3.

No You only said what you where told, any motor like this, that has a closed loop can be called a servo, so these do have a closed loop so can be called a servo, the Teknic engineer is the one who was miss leading, to tell you that they have a 2 or 3 stack rotor, this is a term used for Stepper motors, do you know the pole count of these motors, they say in there spec's that they have a low pole count, compared to other servo's, it appears that they are in between, they are not a Stepper motor but don't compare to a normal Ac servo motor either

[louieatienza]

No You only said what you where told, any motor like this, that has a closed loop can be called a servo, so these do have a closed loop so can be called a servo, the Teknic engineer is the one who was miss leading, to tell you that they have a 2 or 3 stack rotor, this is a term used for Stepper motors, do you know the pole count of these motors, they say in there spec's that they have a low pole count, compared to other servo's, it appears that they are in between, they are not a Stepper motor but don't compare to a normal Ac servo motor either

Their literature and videos say specifically that their ClearPath servos come in single, double, and triple stack configurations.

[mactec54]

Their literature and videos say specifically that their ClearPath servos come in single, double, and triple stack configurations.

Making that statement, tells you that they are not real world servos, as real servo motors don't have stack configurations, this is a term only used for Stepper motors, the only part that makes them able to be called a servo is the encoder and having a closed loop

[louieatienza]

Making that statement, tells you that they are not real world servos, as real servo motors don't have stack configurations, this is a term only used for Stepper motors, the only part that makes them able to be called a servo is the encoder and having a closed loop

Well they do have a skewed stator similar to many servo offerings, so I suppose it could be more of a "hybrid" system than just a "step-servo" system....

[mactec54]

Well they do have a skewed stator similar to many servo offerings, so I suppose it could be more of a "hybrid" system than just a "step-servo" system....

Skewed stator or rotor is used with stepper motors and servo motors this helps with the cogging that you get with these kinds of motors, I found some more spec's of these motors and they are quite well made, what I found is that they are 8 pole so would need the skewed stator to help with cogging, most normal modern servos are 12 pole with most having the rotor skewed not the stator

[TeknicTom]

Hi, this is Tom from Teknic. I’ve seen some questions in this thread about ClearPath, so I figured I’d see if I could help clarify.

From a construction perspective, ClearPath is an 8-pole, permanent magnet, 3-phase brushless motor with a built-in drive that uses AC (sinewave) commutation and vector torque control. The motor part of the ClearPath integrated system is also often referred to as a PMAC motor. Brushless, 3-phase, PMAC motors are currently the most common motors used in servo control applications, so they are often called servo motors even though they can be used in non-servo applications (e.g., they could be run synchonously without a feedback device), in which case it would be not correct to call them servo motors. Conversely, most types of motors (including stepper motors) can be run as servo motors, and could properly be called servo motors in those cases.

The word servo just refers to a system that continually compares the actual value of a variable you’d like to control (e.g., velocity, position, etc.), with the desired value, and changes one or more inputs to the system to make the actual value be closer to the desired value.

A stepper motor is also a permanent magnet, brushless motor, and in most industrial steppers, typically is a 2-phase motor with 50 north-magnetized rotor “teeth” and 50 south-magnetized teeth, and is typically run open-loop (i.e., not as a servo). That said, you can run a stepper motor as a servo if you, 1) put a feedback device on it (e.g., an encoder), and 2) you use a stepper drive that can measure the actual torque, velocity and position, and do a reasonable job of making the actual values of these variables closely follow the desired values for these variables. It would be proper to call this a servo-controlled stepper motor system. The special “toothed” design of stepper motors which allows them to be used open-loop for position control, also however makes them less than ideal for servo-controlled applications compared to a standard PMAC motor.

The number of poles in a particular motor design, 8 in the case of ClearPath, is a balancing act of many different objectives. Higher pole-count motors have more inductive and iron losses as speed increases, so brushless PMAC motors (very often called servo motors, as described above) most commonly have 4 to 8 poles, although higher pole counts are not unheard of. (Relatively few commercial PMAC designs use 12 rotor poles, although many have 12 stator slots. Sometimes these are confused when looking at the internals of a motor.)

The term stack length is, I believe, a term that originated with stepper motors, and so is often associated with steppers. But it just refers to how many rotor magnet assemblies are “stacked” on one another to get increased torque for a particular frame size. ClearPath motors use the “stacking” of rotor magnet assemblies, so it seems reasonable to use the term “stack length”, especially because many ClearPath customers are used to buying steppers, and understand what the term means. (The cut-away drawing of ClearPath in this thread illustrates the stacking design.)

In my personal opinion, I think it might be better if the industry simply specified the length in inches or millimeters, and eliminated the ambiguous stack length terminology. But old habits die hard.

Finally, on the question of skewing stators versus magnets: Either technique can be used to reduce the detent torque produced by the non-uniform attraction (as a function of rotor angle) of the permanent magnets to the stator iron. Skewing the magnets requires the use of molded magnets that are magnetized after they are molded and mounted on the rotor. In ClearPath, we use sintered (i.e., pure) neodymium-iron-boron (NdFeB) magnets, not molded magnets.

Molded magnets have the advantage that they can be magnetized after installing them on the rotor, and thereby magnetized in a skewed pattern, but because they are not pure magnetic material (they are made with a mix of a nylon polymer or polyamide, and magnetic powder), they have much lower magnetic energy than sintered magnets. The higher energy per unit volume of the sintered magnets provides much more motor power per unit volume, but they can’t be magnetized with a skew, so we have to skew the stator laminations instead to reduce detent torque. (As an aside, Teknic also maps the residual detent torque pattern of each ClearPath motor when it’s built, and the ClearPath firmware uses a proprietary algorithm to reduce the effect of this detent torque while the motor is running.)

Sorry for the extremely long post. These topics are very technical and so there is often a lot of confusion even among very smart and experienced engineers. I hope I clarified at least a few points and didn’t bore anyone with all the technical details.

Best of luck with your build! It looks great so far. Please contact Teknic if you have any questions.

[1Jumper10]

Thanks Tom. I appreciate the detailed explanation from someone who knows about the product. ClearPath servo's are working great on the machine I just built and they should work just as well on this machine.

[NIC 77]

@ Tecnic Tom

Would you say that your products are similar to an mdrive stepper/servo? If not, what are the differences?

Personally, I would like to see a product that has more of a flat RPM vs torque curve, not entirely flat, but more flat, a lower initial torque and a higher final torque. I like the simplicity of being able to run your servos with the same kinds of inputs for stepper motors. For me, a true closed loop is not important.

Your products are expensive but the driver is built into the motor, is that right? So there is value in that. I think you have great products, and if your price point was lower, you'd take over the stepper market. Out of my price range at the moment, and I'd consider going to true closed loop servos for that price, not because of the closed loop per se, but because of the flat torque speed to get me better acceleration at higher speeds.

It does look like you have a very nice product though! If I could trade the steppers I have for a similar model of your servos I would do so in a heartbeat.

Is there a particular model you would recommend for 1Jumper10's build?

@1Jumper10

ClearPath servo's are working great on the machine I just built and they should work just as well on this machine.

Your last machine used CNCRP Pro belt reducers? 3:1? Also the input gear inertia on a belt driven gear system is low.

You are using 10:1 reducers on this build. Also, I was looking at the spreadsheet I did using inertia effects, and for your build, I think I used a much smaller value of gear inertia than I should have. I just found a gear and used that value, but I have since been rethinking this. I don't think that gear is what you have as it has low input and output torque values. Is there any way you can find out the specs of your 10:1 reducers, in particular, the inertia?

If you're thinking that you can use the same motors you did on your last build and get the same performance, well, this may be true if your acceleration is low, but a bit more math is needed IMO before should buy the rest of your motors.

Have you seen any other rack and pinion machines with 10:1 gearing that gave good performance using stepper motors or clearpath servos? Something with a similar torque / speed curve?

[louieatienza]

@ Tecnic Tom

Would you say that your products are similar to an mdrive stepper/servo? If not, what are the differences?

Schneider m-drive is a stepper with integrated control/drive

Personally, I would like to see a product that has more of a flat RPM vs torque curve, not entirely flat, but more flat, a lower initial torque and a higher final torque. I like the simplicity of being able to run your servos with the same kinds of inputs for stepper motors. For me, a true closed loop is not important.

That would be the SD-HP series...

[TeknicTom]

Thanks Nic 77 for your questions. Louie is correct. MDrives are steppers which means that they are less efficient and are limited on torque at high speed. They also have some limitations when it comes to acceleration and when microstepping. The ClearPath motors are true closed loop servos in every sense of the term.

When comparing steppers and servos you should always de-rate the stepper motor torque rating by 50%. This is stated on many stepper motor websites but not all. This helps to ensure that you don't push the motor too hard and lose steps.

It should also be noted that there are three different winding choices for the ClearPath motors in order to optimize the motor characteristics to your application objectives.The "P" and "D" windings provide higher power at faster speeds while the "S" winding provides higher torque at relatively lower speeds. The correct winding selection can provide the characteristics that you mentioned with the flat torque-speed curve out to an appreciable speed.

The servo drive is indeed built into the enclosure housing with the motor. This eliminates the cables between the drive and motor and makes a compact more cost effective design. While the motors may seem expensive when comparing them to off-shore steppers, it is all about value and quality. We could have made them "cheaper" but not with their current level of performance and quality. They are made to be industrial grade and last for many years.

As far as motor selection for 1Jumper10, I am happy to help select the motors but I would need more application data. I can run a simulation in order to optimize the selection for price and performance.

[NIC 77]

That would be the SD-HP series...

you are correct. I should have clarified. I meant in the less expensive series.

When comparing steppers and servos you should always de-rate the stepper motor torque rating by 50%.

Perhaps this is the way some manufacturers can account for low quality drivers. As a blanket statement though, I don't believe it. If I look at a torque vs speed graph for a stepper that is based on tests and the same kind of driver, and power supply I expect to be within 10%. If you look at a high quality stepper like an mdrive, I really don't think it's fair to say that you should reduce the torque by 50% for that motor, but consider yours at 100% when doing calculations. Of course, I am looking at the torque vs speed graph along the entire range, not certain what you are referring to by torque rating.

You may be correct, I just don't see your opinion as unbiased, and I don't believe it at this point in my understanding.

Don't get me wrong, I do think based on what I've seen that you have good quality products, and yes, I would still gladly trade my steppers.

2nd last question for you. What % of time should you be able to run at peak torque vs continuous torque? Can you use the peak torque values for calculating acceleration on a cnc machine that is doing fast intricate work, or if a compromise exists, what value should be used between peak and continuous torque in order to calculate acceleration?

It should also be noted that there are three different winding choices for the ClearPath motors in order to optimize the motor characteristics to your application objectives.The "P" and "D" windings provide higher power at faster speeds while the "S" winding provides higher torque at relatively lower speeds. The correct winding selection can provide the characteristics that you mentioned with the flat torque-speed curve out to an appreciable speed.

Ah, Ok, for some reason I always ended up looking at just this page:

https://www.teknic.com/products/clea.../clearpath-sd/

Are these the only ones you can control with a regular BOB that you might use with steppers? Are the other ones you mentioned relevant for DIY CNC purposes? If so, is anything more needed to drive them?

[1Jumper10]

27

Nic77 - The mass moment of inertia of the Gam 10:1 gearbox I'm using is .318lb-in^2. But I think your getting way too hung up on inertia matching servo's to loads. While I agree that gross mis-matching would have a negative effect, I think that for the most part, the CP servo's auto tune themselves to a wide range of loads. And probably only at the extremes of performance would you see the difference.

I just checked the SDSK product line on Teknic's web page and there are 27 different models in the NEMA 23 and 34 size range. Its that broad range of models and their capabilities that allow a novice like me to get good performance from a home brew CNC machine. I say this because I have first hand experience using these servo's on my friends machine. During the tune-up stage when I was first getting it moving. I was getting scary fast accelerations and top speeds at somewhere near 30% of max current! I think I ended up limiting them at ~15% (?) just to be safe. At 1200 IPM rapids and 400 IPM cuttng speeds, if the status lights weren't flashing I wouldn't even know they were on. There is no noise, no heat, nothing. And the only thing I did to size them was talk with Teknic, decide on my desired performance and then AUto tune. This is why I think your getting too concerned with inertia matching. And its why I'm using them again for this machine. They have a model that will work well and when I get my mechanicals finished I'll choose it. Matth's machine is going to be heavier than mine and he's using NEMA 23 steppers geared 5:1 and is shooting for crazy fast performance! I'm just saying choosing a drive motor isn't critically important nor hard to do.

[ger21]

Matth's machine is going to be heavier than mine and he's using NEMA 23 steppers geared 5:1 and is shooting for crazy fast performance!

I believe that he's using 400w AC servos, if I'm thinking of the same machine.

[1Jumper10]

Yes. Gerry I believe you are correct and I misspoke. I was right about the NEMA 23 part though.

Sent from my Pixel XL using Tapatalk

[louieatienza]

you are correct. I should have clarified. I meant in the less expensive series.

That's why they're less expensive. But they're meant to be higher-power stepper replacements, which is why the torque curves of the SK series more closely resemble that of steppers. If you sort the list by price high to low, you 'll see the higher priced motors just happen to be the ones with the higher power outputs.

Perhaps this is the way some manufacturers can account for low quality drivers. As a blanket statement though, I don't believe it. If I look at a torque vs speed graph for a stepper tis based on tests and the same kind of driver, and power supply I expect to be within 10%. If you look at a high quality stepper like an mdrive, I really don't think it's fair to say that you should reduce the torque by 50% for that motor, but consider yours at 100% when doing calculations. Of course, I am looking at the torque vs speed graph along the entire range, not certain what you are referring to by torque rating.

Then by this reasoning all stepper drive manufacturers make inferior products, which of course is untrue. They are what they are, and the reason why you de-rate the torque is to ensure that you don't miss steps - which is vital because there is no position feedback. And yes, you de-rate it along the entire curve. Of course, as DIYer's we're looking for the best bang-for-the-buck in terms of performance - we're hot-rodders at heart and sometimes we run our machines at the brink. But there's always going to be increased risk of missed steps with getting every last oz-in of performance from a stepper. Maybe the DIY crowd is more tolerant of this. If you were a mechanical engineer, however, and you had to rely on a stepper system being as fail-safe as possible, you would have to figure about a 30-50% derating of the torque. It's different with a servo - it will pump as much juice to the motor as needed to maintain position, at least until you reach a fault condition. Servos and drives can be pretty tolerant of a 10:1 inertia mismatch and more. With steppers, I'd try for as low an inertia mismatch as possible.

Ironically, in Schneider's own website they have an article on this: Sizing: Load to rotor inertia matching

Are these the only ones you can control with a regular BOB that you might use with steppers? Are the other ones you mentioned relevant for DIY CNC purposes? If so, is anything more needed to drive them?

I believe the SD and SC series take step and direction inputs. The inputs are opto-isolated, so you potentially don't even need a breakout board, though I believe the inputs need at least 5V at 8mA (and I believe can receive 24V signals from industrial controls), so depending on your setup you may need a buffered breakout board. I believe most motion control boards out there now are capable of this. You'd also need 18-75V PSU.

[TeknicTom]

Again, Louie and 1Jumper10 have raised some good points. Oriental Motors clearly states on their website to de-rate their own stepper motor specs by 50% and Schneider states that you should not go above 10:1 inertia mismatch (and lower is better). Perhaps a good source of information on this topic is this video:

The SD family is the step and direction family and is most appropriate for CNC type machines. These motors offer variable accelerations, velocities and distances up to the motor's max capability. All you need from a controls perspective are step and direction signals between 5VDC and 24VDC (with 0VDC being the low state).

Peak torque can be used for up to 3 seconds and continuous torque can be used 24/7. For inertial loads, peak torque is most often used for less than 1 second because once you reach full speed it takes much less torque to keep the load moving than it needed to accelerate the load.

[NIC 77]

Nic77 - The mass moment of inertia of the Gam 10:1 gearbox I'm using is .318lb-in^2.

OK, thanks, I'll have a look at it when I get a chance.

This is why I think your getting too concerned with inertia matching.

Servos and drives can be pretty tolerant of a 10:1 inertia mismatch and more. With steppers, I'd try for as low an inertia mismatch as possible.

No idea why you guys are talking about inertia matching. Not what I'm talking about at all. You need the inertia to calculate the torque required by the motor during acceleration. Inertia matching is to avoid overshoot or undershoot. Not looking to calculate an optimum gear ratio here, just to see what performance can be expected.

Again, Louie and 1Jumper10 have raised some good points.

You make me sad Teknic Tom. So have I.

Peak torque can be used for up to 3 seconds and continuous torque can be used 24/7. For inertial loads, peak torque is most often used for less than 1 second because once you reach full speed it takes much less torque to keep the load moving than it needed to accelerate the load.

Not at all what I was looking for. The acceleration will take a fraction of a second, but then you may need to decelerate again after another fraction of a second. More, I was looking for a duty cycle, as in percentage of time that you can run at peak, or if there was a compromise between continuous and peak that can be used for these situations. What happens if you use peak too much, will the motor and controller automatically tone down the movements or do you just get a fault?

Matth's machine is going to be heavier than mine and he's using NEMA 23 steppers geared 5:1 and is shooting for crazy fast performance!

No. He's not. He's shooting for very fast top end speeds with very mediocre acceleration, and doesn't really know what he will get at the end of the day. I think he will be OK.