Smart Steppers

[alexo]

Well, now that I find you work for them things are different. I won't compare with specific products anymore and I was getting ready to do just that. I think you have some idea how much people like me are being pestered by sales people trying to push product. Nothing wrong with that but for the fact they almost never seem to know what the product is used for by me. Most of my interest is 1 to 4Kw metal cutting machine tools. Wood chopping sometimes. That means most of the claimed features of new products are absolutely irrelevant. A CNC machine can work for many hours on an expensive mold and a single error could translate into a large loss. And errors happen all the time because the toolpath software is not foolproof. If you machine a mold for a say a car door then how fast the machine moves while cutting air or between tool changes is irrelevant. Reliability is everything. Dynamic performance is a big issue with lasers where forces and speeds are perfectly well known.

Elmo's idea to call their drives "Clarinet" , "Bassoon", Accordion and so on was stupid. My clients tend to fork out money for Yaskawa or Mitsubishi and at a push might be ok with products from a TV making company. Like Panasonic. This is reality. For me is counterproductive to push product from lesser known manufacturers and it'd take me a page to explain why in detail. One example : check how much 2nd hand stuff from Elmo is available "out there". Now compare with Panasonic or Yaskawa. That matters big time for me and many like me.

I couldn't discover Elmo's sales office in SA but anyhow, I would definitely not have the time to test their products. Nor would I be prepared to put up with the expense. The couple of figures you kindly supplied and some others I was forwarded by somebody in the US do not warrant the trouble. The fact they do not have product manuals freely available and sufficiently detailed is an alarm signal for me. Seen that before once or twice... It's a very bad policy - I'm not going to "discuss my needs" with their almost incompetent sales rep. It'll be a waste of time and I don't want to have my clients shortcircuited.   

The 22g and 4Kw drive is an impressive piece of engineering but what would I need it for The demo you kindly uploaded shows it works somehow. Not knowing what the peak power and duty cycle are there that might as well be a 400W drive. Might be 50W...

Powering one motor from two drives is interesting but again, nothing I could use it for. Never had an issue with the available space for a drive. This is a real niche application and given that a specially wound motor is needed I'd be hard pushed to even start imagining where this could be useful. And anything over 10Kw is completely out of my league.

Haha, no. I don't work for them...

Frankly they supply a lot to military applications and don't cater to small volume applications like retrofitting due to support requirements. Don't flatter yourself in thinking I'm some kind of salesman targeting you.

I provide technical support and do setup / integration work across 4-5 different servo drive manufacturers.

I'm not trying to sell it here, I think you got me wrong. I admit it sounds like I'm pushing the product a bit, but this entire conversation is more about -servo drives- then anything else. I'm only pointing out Elmo as they are damn impressive in performance and reliability.

For your information, we do see a lot of business in the twitters. They are used often in drones, robot arms and other compact machines.

Again, most of this stuff isn't even targeted at you, because there isn't much business in that area, but it is still useful when other drives simply don't give the performance available.

That said, I do agree they should put more of the information out on the public.

Drones and electric vehicles are a good example for a specially wound motor and need for extreme compactness. You want as much space as possible for extra batteries and want to drop weight a lot too.

If I actually was trying to push this product, I'd be going about it a very different way. Its amusing how hostile you seem to new technology. I take it you still use exclusively analogue scopes and resolvers? 

Well, I'm surely glad you're not working for them. But I won't tell you why.

I'm not at all hostile to knew technology BUT if you've been long into this ( since the '80s ... ) you must've seen the deluge of "New Technology" we were inundated with. Where is most of it ? Nowhere. In the early 2000s we saw a lot of new and often very exciting products just to watch them disappearing within a couple of years. Big manufacturer does not like competition. How do you think Fanuc survives ? And then, there is something else : a lot of the new technology did not pan out on the factory floor. Just not enough performance gain for the trouble/cost and VERY often associated with other poorly researched new technologies. On paper, some 5 axes milling machines looked wondrous. In practice however agile their drive systems were the spindles lasting only a couple of hours was a downer. Couple of years ago client of mine bought two laser machines from a very reputable German manufacturer. One had linear motors and was substantially more expensive. Which machine do you think worked flawlessly 24/7 and which one broke down all the time ? Which machine do you think could be repaired with non OEM parts for cheap and for sure ? Specialized application can benefit a lot from the latest. Machining is a different game - down time costs real money. My impression is that a lot of times new technology is purely sales driven - there is no real, quantifiable need but we need our product to seem different in the market.

[alexo]

There is an entirely underserved market for people with some experience with motor controllers: car parking systems.

In the building next to ours they installed a parking robot. 6 months it stood barren since a turnstile (on the robot) would not stop within the allowed angle. In discussing this with their support team they insisted that cars were "heavy" and therefore the servo was "unable" to position the cars correctly. They insisted the motor/drive pair was good and well specified and configured (yeah right).

They ended up welding a spring (and I assume some form of damping) that the turnstile would stop against.

Crappy systems.

YES. Welcome to the game : most time and money are spent fixing the design. Most designs are crap. Look at cars : one would've thought that in 120 years they have it figured out.

[CM800]

My impression is that a lot of times new technology is purely sales driven - there is no real, quantifiable need but we need our product to seem different in the market.

Haha, yes. It's hard to argue that.

The marketing wank is indeed strong these days.

How long until we get our first Bluetooth & wifi enabled wireless IoT servodrive?

It scares me that some people thing wireless controlled servo drives would be a good idea... Nothing like someone hiding a cigerette packet sized jammer in a warehouse to stop production for a month....

That said. I would personally rather like a drive that can be configured via bluetooth from a phone app  as an option (switch on side of drive) Though with the advent of fieldbus being so common, it doesn't hold as much appeal these days.

One of the big things I like with elmo is that -it just fucking works-

I wire the drive up, I connect to it, go through the setup wizard, spend half-hour turning it and you get damn impressive performance for the average application. You don't even need to enter the motor specs, it runs some 'experiments' on the motor to work everything out - which actually works.

That said I'm seeing a lot of new companies coming out with these features.

Elmo's Gold range is pretty reliable from all I've seen, rarely get any back from the ones I've helped install, and when I do it's almost certainly a user error (Elmo burn the drives in for 24 hours before sending them out, heck, their 'Extreme Environment' range, they do full vibrational testing under full load too.

I've done a good bit of work for military applications, one thing they absolutely hate is change. So many still use VME64, Resolvers etc. Yet, I'm seeing a lot of them begin to use Elmo. I think that says something.


I really do agree with your point on how many products out there are badly designed these days and 'look the shit' yet turn out so bad. It's a symptom of modern business operation where the management don't care what they ship out, as long as it makes them money. Career CEOs and Managing directors Etc. Ones who just hop between completely different industries with no expertese in the actual product they are in charge of.

[Assafl]

Resolvers? Aka Singer / Kearfott / Selsyn?

Are those still around? Talk about infinite resolution.

[CM800]

Resolvers? Aka Singer / Kearfott / Selsyn?

Are those still around? Talk about infinite resolution.

To be honest, Resolvers are very common in Military and Nuclear, great way of getting feedback if silicon can't be trusted in the environment (Ionizing radiation)

[691175002]

Every "Closed Loop Stepper" system I have seen is just a rebranding of a brushless servo.  Most of them have skewed windings, and more than half actually use three phase brushless motors!

They borrow the stepper name purely as a marketing label.  A "stepper" system is generally viewed as being easier to use, providing higher torque per dollar, and operating at lower speeds than the average servo.  Closed loop "steppers" use high pole count motors and more forgiving control algorithms to achieve those objectives.

The defining characteristic of a stepper motor is that position is controlled by energizing the windings in order.  If you are commutating with position feedback its a servo.  (And if you are going to go through the trouble and cost of commutating with position feedback you might as well use a motor appropriate for the job, which is why "closed loop stepper" never actually use stepper motors.)

[CM800]

Every "Closed Loop Stepper" system I have seen is just a rebranding of a brushless servo.  Most of them have skewed windings, and more than half actually use three phase brushless motors!

They borrow the stepper name purely as a marketing label.  A "stepper" system is generally viewed as being easier to use, providing higher torque per dollar, and operating at lower speeds than the average servo.  Closed loop "steppers" use high pole count motors and more forgiving control algorithms to achieve those objectives.

The defining characteristic of a stepper motor is that position is controlled by energizing the windings in order.  If you are commutating with position feedback its a servo.  (And if you are going to go through the trouble and cost of commutating with position feedback you might as well use a motor appropriate for the job, which is why "closed loop stepper" never actually use stepper motors.)

I know a few closed loop steppers that -do- use stepper motors in them, operating with FOC,

Schneider Electric's Lexium MDrives are one. Evodrives (lesser known, it's owned by a friend of mine but he has put the company on hold for the moment due to other engagements)

I believe there to be a few others too.