PDA

View Full Version : GH1440 Lathe Bed Conversion



Valfar
30-01-2017, 02:56 PM
Hello,

New guy here and glad to be part of this forum.

I'm in possession of a GH1440 Lathe Bed + Headstock + Tailstock (a quite rigid base I would say) and I'm willing to convert it into a CNC lathe + 4th axis Router/Mill. Will be used for Al alloy, steel and perhaps some plastic, for hobby and prototyping (max 50mm O/D parts, but occasionally bigger O/D parts as well). Ideally, I'm willing to achieve 0.01mm tolerances.

I've got a lot of unanswered questions as you can imagine, but at the moment I cannot decide how to proceed with the lathe spindle, which will be the 4th axis as well. To make it simpler, I was thinking to design it to be able to mechanically switch between the lathe and the 4th axis mill. So the first question is what motor should I use for the 4th axis / lathe spindle? I was thinking to go for a high torque NEMA34 hybrid stepper (servo stepper) but I calculated it wouldn't give me much RPM for the lathe spindle (about 300RPM). How much RPM would I need anyway, given the parameters? Would I need 2 different motors instead, one dedicated for the 4th axis and another one dedicated for the lathe spindle?

Anybody who finds a free minute is more than welcome to suggest/advise anything.

Many thanks!

Valfar
22-02-2017, 09:46 PM
Hi, I know nobody answered, sometimes I ask stupid questions.

But do you find the speeds for the lathe fairly accurate in the *.pdf bellow?

http://gradstudentshop.usc.edu/assets/001/64641.pdf

LE. I was thinking to use this servo stepper as a lathe spindle motor. Do you think it would suit the application?
http://www.ebay.co.uk/itm/EU-STOCK-Nema34-12N-m-Closed-Loop-Servo-Motor-154mm-HSS86-Hybrid-Driver-CNC-Kit-/162398824086?hash=item25cfb96696:g:MVwAAOSwKOJYIDq I

m_c
22-02-2017, 11:09 PM
For that size lathe, I wouldn't personally got for anything less than 2HP for the spindle for turning.

For a 4th axis, that's where things get tricky. Even with a large servo, by the time you gear it to get 2-3000rpm, you end up with very little holding torque for doing accurate positioning work.

And not only that large servos are expensive. This is something I've often though about, and for ideal use, you need some way of changing servo:spindle ratios, so you can have one high speed low torque ratio, and one low speed high torque ratio. The problem is how to achieve that while minimising backlash.

Valfar
22-02-2017, 11:33 PM
Thank you for the reply m_c. Good points.

My thoughts were to use 2 separate motors, and couple them both to the spindle: one geared up and the other geared down. But only one would be powered on and running at a given time. At the same given time, the other one would still be rotating, but driven by the spindle (as it won't be powered) - the spindle would rotate due to the other one.

I would create a lathe profile (in Mach3 for instance) and run the machinery in lathe mode, with the high RPM / low torque servo power on (and the other servo off).

When milling is needed, I would shut down everything, create another profile (mill profile) and run the machinery in milling mode only, but this time with the low RPM / high torque servo powered on (and the lathe spindle servo off, which would still be rotating due to the other motor), configured for the 4th axis role.

The lathe and mill would share the X and Y axis steppers.

I hope it makes sense.

Do you think I'm going on the right track, or talking bananas?

komatias
23-02-2017, 10:25 AM
Hi Valfar

Bananas for sure.

Get a mill for milling and a lathe for turning. Even a small benchtop mill is better than what you are trying to do.

Also, for a lathe , there is no need for a servo spindle, just install a indexing pulse to tie every thing together. Then with a nice 3phase motor on a VFD you can do all you need to do.

Regards

Valfar
23-02-2017, 02:10 PM
Hi Komatias, thank you for the input.

I'm sure there are lots of things better than what I'm trying to achieve, but in my view it all goes down to what one's able to compromise, like every single bit in this life. I think everybody agree a lathe/mill combo is not a new concept. That being said, I don's see a problem with the design idea. The challenge would be how it's implemented. I now need to understand what's wrong with my approach.

I understand a VFD + 3 phase motor would work similar say to a 3kW spindle found on the ebay, but also add an indexing pulse into the equation. Is that correct? Is there any VFD/motor you would recommend for a DIY application, or does any reasonable priced ones bought over the ebay would do the job?

Cheers!

komatias
23-02-2017, 05:04 PM
Valfar,

Are you looking to make money from your machine or are you just doing it for a hobby? If you are planning to make money, do not be cheap.

Any lathe has been designed to take cutting forces in the direction that causes the saddle to compress against the bed ways. Even the machines that have a milling head on the top of them take forces in that direction. Using the lathe spindle to mill causes a torque which compresses the one side of the saddle while lifting the other side. Not very good for the saddle clamps and will undoubtedly make milling a pain in the rectum.

Myford users get away with it because they only take light cuts. With the 1440, it would be a waste really.

Yes, you are correct in the watercooled spindle comparison, only difference is that the lathe spindle motor can be 2.2kw and have lower RPM, thus higher torque ;)

m_c
23-02-2017, 05:56 PM
Komatiad, my understanding is the spindle would become an A axis, with a separate milling head bolted on elsewhere.




My thoughts were to use 2 separate motors, and couple them both to the spindle: one geared up and the other geared down. But only one would be powered on and running at a given time. At the same given time, the other one would still be rotating, but driven by the spindle (as it won't be powered) - the spindle would rotate due to the other one.

That wouldn't work. Say you have the low speed with a 10:1 ratio, and the high speed direct drive, that would mean at high speed the low speed motor will be getting driven with a 1:10 ratio, so at 3000rpm, it would be doing 30'000rpm.


I would create a lathe profile (in Mach3 for instance) and run the machinery in lathe mode, with the high RPM / low torque servo power on (and the other servo off).

When milling is needed, I would shut down everything, create another profile (mill profile) and run the machinery in milling mode only, but this time with the low RPM / high torque servo powered on (and the lathe spindle servo off, which would still be rotating due to the other motor), configured for the 4th axis role.

The lathe and mill would share the X and Y axis steppers.

There is a Swapaxis function within Mach3 that you probably want to look at.
If you've not already seen it, you may want to have a read of this thread - http://www.machsupport.com/forum/index.php/topic,11422.0.html

A possible option, depending on if you don't need the 4th axis to move while machining, is to add a brake, as then you don't need lots of torque to hold the spindle. You just need enough torque to locate the spindle, then use the brake to hold the spindle.

Valfar
24-02-2017, 10:08 AM
Morning!

The only major compromises I'm making is the compact size of the machinery, and the lathe + 4th axis mill combo. Have already gotten the bed + headsock + tailstock. From here, will definitely won't go cheap. It will be used mainly for hobby, not production, but you never know. I need to bring to life a few design ideas I have.

M_c is right, the lathe's spindle won't be used as a milling spindle. The lathe's spindle will be used as a 4th axis when in milling mode. Another spindle (perhaps an ordinary 3kW water cooled spindle off ebay or a better one - no idea yet) will do the XYZ operations, similar to what a conventional milling machine does. But it will all be much clearer when I'll have a 3D model ready to show you.

I'm not yet familiar with Mach3 and with the low level programming. At work I'm using an industrial 3 axis router, and software like Vectric / ArtCAM (you get the idea). Will definitely look into the 'Swapaxis' function, thank you for bringing it into discussion.

I was aware of the brake system design for a 4th axis. But my thoughts were to use the 4th axis simultaneously. The brake could be added as an option though, when only one face would need machining. But it adds to the complexity, and I'm trying to keep it as simple as possible. Wouldn't a geared 12Nm Nema34 do the job without a break? I don't need high speed & feeds. I will consider a brake if it doesn't.

Also, good point m_c about the low speed motor driven by the other one at extremely high RPM. Haven't thought about it. That means the lathe motor (the 2.2kW low RPM komatias has been suggested) can be permanently attached to the spindle, and a servo (hybrid) stepper would need somehow to be detached when in lathe mode.

I see this guy nailed it. He made a lathe and a 4th axis mill run simultaneously. I'm sure you are well aware about his project.
https://www.youtube.com/watch?v=b2-Kdud7eiA

And with a single motor attached to the spindle. What's the catch?

Plenty of homework for the week-end.
Thank you for your time. You guys are great!

hanermo2
24-02-2017, 04:42 PM
I am perhaps the only one with experience, who has done this, so...

At this time, after 3-4 trials, I have a 2.5 kW AC brushless servo as the spindle drive, at 1:3 belt drive, via HTD 8-30 belts, on a very strong, sturdy, rigid frame of approx 60 kg mass.
So I get 0-1000 rpm, with a C axis or "live cnc spindle".

Servo is 10.000 counts at 1:3 belt drive for 1:30.000 counts at spindle.
This is not rigid enough for milling.
And my 600€ belt drive 30 mm wide, 8 mm profile, taperlock pulleys, 24:72 teeth at spindle, setup is very, very industrial, so...

The setup is excellent for turning and lathe use.
I get 90 Nm at spindle.
Spindle motor is 10 Nm cont, 30 Nm peak (3 secs) x 1:3 = 90 Nm at all rpm.

I will use it as-is, until I switch to a secondary, likely a (double) worm drive or planetary at 1:10, at some point.

My experiences, working at this for 10+ years.
Steppers are no-where near accurate enough.
Even the best belt drives are not accurate enough.

HAAS uses a gear on the spindle, for C axis use.
But they use a class 9 ultra-precision gear for the C axis, only, switch it in-out, and the gear grinder cost 9M$, I have met the guy who sold it.

So my system is better/similar to simpson of cubestudio, and not good enough in rigidity for C axis.
I use a bigger/more accurate servo, and similar/bigger belts (his are also HTD-8, I think mine were wider).
And my lathe is a 12x, fwiw..
Hth.

m_c
24-02-2017, 04:48 PM
I am perhaps the only one with experience, who has done this, so...


Yet nobody has ever seen any prove that you have, despite being asked numerous times...

hanermo2
24-02-2017, 05:34 PM
I try to provide data.
Any data that seems incorrect, please comment on.

I have numerous pics, data, stuff from numerous clients, and none of it may be shared without prior approval from clients, and usually client they do not want to.
I try to help people, thats all.


Yet nobody has ever seen any prove that you have, despite being asked numerous times...

Valfar
24-02-2017, 09:25 PM
Thank you Hanermo2 for the valued information.

I read your concentrated post at least 10x times to understand what's in there, but I think I got there in the end.

I understand steppers driven by belts are not accurate enough, and steppers driven by gears are rigid, suitable for the application, but the problem with these is the backlash. Is that correct? It does make sense, as the belt at the end of the day is made out of flexible material, unlike gears which are made out of steel.

When you say you'll be switching to secondary, do you mean you'll add another servo stepper for the C axis, but driven by a worm gear or a gearbox?

In my headstock I still got there a secondary spindle with spur gears all over it. There's also a clutch mechanism on the front panel which can engage the secondary spindle with the main spindle. So I might be able to direct connect a servo stepper for the C axis to that secondary spindle. But the trouble would be the backlash between the spur gears, and probably less accurate than a belt drive.

Today I grabbed a 2.2kW motor (that was quick!), but don't know many details about it yet. It was free of charge and it's blue :) . Next I'll need to invest in some VFD and play with it to understand the process.

I'll keep reading here in there.

Cheers!

LE Have found this 'CNC Lathe Spindle Encoder' design http://ve7it.cowlug.org/spindle-encoder.html

m_c
25-02-2017, 12:16 AM
I was aware of the brake system design for a 4th axis. But my thoughts were to use the 4th axis simultaneously. The brake could be added as an option though, when only one face would need machining. But it adds to the complexity, and I'm trying to keep it as simple as possible. Wouldn't a geared 12Nm Nema34 do the job without a break? I don't need high speed & feeds. I will consider a brake if it doesn't.

You would need to run some figures using the torque your milling spindle could produce, and the diameter of your work piece, to calculate the torque required at the 4th axis.
I.e. using some random figures, say your milling spindle can produce 1Nm, and you have a 5mm radius (10mm diameter) cutter.
At the very edge of that cutter, you potentially have 200N of force (1Nm/(0.005m/2)).
Now apply that 200N to the edge of a 25mm radius work piece, you get 5Nm of torque produced (200N*0.025).

That means under perfect transfer of torque conditions (very unlikely), you need 5Nm to hold the spindle. You're not likely to get perfect transfer of torque, however the torque applied isn't constant, and that is compounded by should the cutter snag, you have the inertia of the spindle/cutter which will produce peaks far higher than 5Nm.
However that is assuming you're using some form of side cutting cutter, cutting at a tangent to the workpiece, and at 90deg to the 4th axis axis. Things like drilling holes around the circumference, will result in minimal rotational torque being applied to the 4th axis.

If you want to run figures, most of this can be worked out to a reasonable accuracy.



Also, good point m_c about the low speed motor driven by the other one at extremely high RPM. Haven't thought about it. That means the lathe motor (the 2.2kW low RPM komatias has been suggested) can be permanently attached to the spindle, and a servo (hybrid) stepper would need somehow to be detached when in lathe mode.

I see this guy nailed it. He made a lathe and a 4th axis mill run simultaneously. I'm sure you are well aware about his project.
https://www.youtube.com/watch?v=b2-Kdud7eiA

That is a video of the original design from that thread I posted. It's worth having a read of that thread, as Simpson explained quite a lot about how he designed and built the inturn's.


And with a single motor attached to the spindle. What's the catch?

Getting the required torque.


Plenty of homework for the week-end.
Thank you for your time. You guys are great!
Spreadsheets are your friend for working out some ball park figures for the required torque.

m_c
25-02-2017, 12:23 AM
I try to provide data.
Any data that seems incorrect, please comment on.

I have numerous pics, data, stuff from numerous clients, and none of it may be shared without prior approval from clients, and usually they never do not want to.
I am not engaged in a popularity contest, and if/when I want to show publicity stuff I will do so.

I try to help people, thats all.

But I thought you had built this as a personal machine?
At least that's what you seem to claim until anybody asks you for prove.

And just so you know Valfar, Hanu has posted about this mythical machine for many years on various forums/email lists, and although his theory is reasonable, nobody has ever seen any kind of prove that such a machine actually exists. Whenever pushed for evidence, the excuses I've seen are he has no camera, he can't because of confidentiality, or the toys simply get launched out the pram, so make up your own mind.

Valfar
26-02-2017, 02:46 PM
Thank you m_c! It does make sense, and I've also found some spreadsheets for calculating stepper torque values. I also believe they are a good start before deciding to buy.

Also, an excellent tutorial (from my point of view) can be found bellow:
http://www.mycncuk.com/threads/1524-What-size-stepper-motor-do-I-need

But now Hanermo2 wrote:


My experiences, working at this for 10+ years.
Steppers are no-where near accurate enough.
Even the best belt drives are not accurate enough.


Now from my understanding that means stepper motors driven by belts are not accurate enough (what's the exact value of this? 10 thou? 100?), and the way to go would be either a worm gear or a gear box. But the problem with these is the backlash they introduce, and a good quality one is out of the question for a DIY budget. Is that correct?

That leads me to believe is the belt itself causing the problems, being flexible, and not the stepper. Is that correct?

And of course, a brake system applied on the C axis would work, but that means the same issues are encountered on the other axis (XYZ) as well?



Getting the required torque.


I understand the torque value, but this contradicts with the high RPM value needed for the lathe operations?

Cheers!

Valfar
26-02-2017, 04:25 PM
This is the AC motor I have, similar to the one in the link bellow.

http://www.brookcrompton.com/upload/files/products/20113E_W_CI_IE2.pdf

Brook Crompton
kW: 2.2
V: 380-415
A: 5.0-4.8
RPM: 1415
Phase: 3

I shall use it for the lathe operations.

Is there a good VFD you would recommend, or should I just go for a reasonable one over the e-bay?

I will have to also design a lathe spindle encoder similar to the one in the link bellow, is that right?

http://ve7it.cowlug.org/spindle-encoder.html

Cheers!

Neale
26-02-2017, 04:46 PM
Now from my understanding that means stepper motors driven by belts are not accurate enough (what's the exact value of this? 10 thou? 100?), and the way to go would be either a worm gear or a gear box. But the problem with these is the backlash they introduce, and a good quality one is out of the question for a DIY budget. Is that correct?

One data point based on my own recent experience. My setup might or might be applicable to your situation but it might give a little bit of insight. I've been working on a Z height setting macro on my router. I'm winding a 2.2KW watercooled spindle up and down using a 1605 ballscrew driven by a 3Nm stepper via a 1-1 HTD5 belt and pulleys. Max speed 2500mm/min but can't remember current acceleration settings off-hand - think it's about 400mm/s/s. I can touch off the setting plate, wind the Z up and down 50 times at full acceleration/rapid speed, and retouch. I get repeatability within one or two microsteps, time after time. That corresponds to 3-6 microns (0.003-0.006mm) variation. Ok, the machine is very new so the belt has hardly had time to stretch, and I don't claim that the machine has long-term accuracy of anything like this. Similarly, I can set one tool to the work surface, change tools, retouch off a fixed setting plate, and the new tool will be just touching the surface of the work. That is testing various aspects of the drive system, and I'm pretty happy with it. You will be needing very high precision gearing to get anything close to that, and/or very careful attention to fitting and adjusting and maintaining, and/or a lot of money to buy the components. Although my own experience is with a router, I would be happy to get that performance out of a vertical mill.

Belt drive might not be perfect but it beats any gearing available at an affordable price, I suspect. And while a stepper might not have the resolution of a high-count servo encoder, how often will you need that level of resolution?

m_c
26-02-2017, 07:48 PM
You need speed for turning. To give an example, I run my lathe at 2000rpm for 20mm aluminium, and that is nowhere near the recommended speed. The reason I limit to that, is to stop the stock whipping in the spindle, as it's not got any spindle liners to support material.

Running some basic figures for 20mm bar through HSMadvisor, for 6061 series aluminium recommended speed is 10222rpm, low carbon steel is around 1900rpm, and 316SS is 724rpm.

With stepper motors, the rated torque is holding torque. As soon as the motor starts spinning, the actual produced torque at the shaft quickly drops off.
Also, you can only guarantee a stepper motors position to one full step (1.8 degrees for a standard 200 step motor). The main purpose of microstepping is to improve smoothness at low RPM, not to improve positioning. Microstepping may provide fractional steps in movement, however due to friction/stiction within the system, the exact distance cannot be assured.

Valfar
27-02-2017, 05:32 PM
What you guys are saying makes perfect sense to me, but here's one other thing I cannot understand without actually testing a stepper.

Say a (standard 1.8 degrees/200 steps) stepper motor is rated at 10Nm holding torque (never mind the micro stepping), and we apply a tangent force. With no rotation, just holding the spindle in place, what does this actually mean in practice?

Will the 10Nm stepper keep the spindle dead locked until the 10Nm value is reached, and then it starts skipping in the direction of the force?

Ok, say not until 10Nm value is reached, but will it keep the spindle dead locked until at least 5Nm is reached and than it start's vibrating within the 1.7 degrees interval, and when it reaches 1.8 degrees (10Nm torque) value, it skips in the direction of the force?

Or will it vibrate anyway, no matter the applied force? So for 1Nm it'll vibrate 0.2 degrees, for 2Nm -> 0.4 degrees and so on (is it a linear or logarithmic graph?). If yes, I can understand why a brake is needed in this case.

I hope it makes sense.

Cheers!

hanermo2
01-03-2017, 02:58 PM
Depends.
At the full-step point, it will hold at 10 Nm, until the torque is over 10 Nm, and then it will move one step.
Because the holding torque is very much stronger at full-steps only.

So at less than 10 Nm stress, at the full-step position, it will not rotate at all.

At 1/10 microsteps the torque is == 10% iirc.

So, if the stop point is at 9 microsteps off a full step, it will "bend" or comply until it gets to the full step point.
It does not lose steps, but acts as a spring.
When torque is removed, it goes back to the position it was meant to be in.

Servos are different.
A servo "knows" where it is supposed to be, ie what the encoder count must be to stop.
If I rotate the chuck the servo drive led shows how many counts off it is.
It tries to use peak torque, 3x of rated torque, to get back to the position it needs to be in.

So, the 2.5 kW servo, at 10Nm torque and 30 Nm peak torque, tries to use 30 Nm torque x 1:3 belt drive = 90 Nm to get back to the position it "wants" to be in.
The peak torque applies for upto 3 secs, and then goes to sustained rated torque, 10 Nm in my case (2.5 kW).

The effect is very obvious and intuitive.
You can see on the servo drive led, the error by encoder count, in real time.
If you use so much force that you overpower the servo max-error setting, D iirc, it faults.

This takes less than 1 ms, or 0.001 secs.
Typical servo loops are 12 kHz, or 0.12 ms.

Mostly, the spindle can be off by == 0.1 mm at outer edge of 12" chuck before fault.
With very very loose (poor) servo tuning (factory default).

Servos do not have "microsteps" and cannot loose steps.
Servos position perfectly at rest, and then lock (modern ac servos).
When in use, they lag a bit, and this depends on tuning.
The lag is shown on the led at the drive.

The faster the servo runs (or more acceleration), the more lag it has.

Valfar
01-03-2017, 04:34 PM
It makes sense, what I understand is when a 200 steps stepper motor is used for the 4th axis, it stays dead locked until the rated torque is reached. So basically I could machine a stock on 200 faces with the ultimate precision (which is a 3 to 6 microns variations as stated by Neale - more than enough for me, as I'm not planing to machine air bearing for instance) dictated by the belt (which is the weakest link in the chain). The stepper will never act like a spring unless micro stepping comes into the equation. To me it sounds like a good option to drive the C axis.

Regarding your particular project and the servo motors, I understand your not happy with how it performs when the spindle takes the C axis role. Is that because the servos tend to act like a spring all the time, no mater the position, due to the lack of steps?

Also, there are these hybrid servo (servo stepper) motors. Do they also come with 200 steps + microsteps + all the other benefits of a servo motor? Or are they just as described by you, and 'hybrid' is just marketing? It's a bit confusing for me. If just marketing, I understand a 200 steps stepper would still be the best option for the C axis, provided there is no brake involved, and provided one must never ever go beyond the rated torque? Also, I'm sure there should be some electrics which could shut down everything in case a step is skipped (but this is not important right now).

I want to connect 2 motors to the headstock spindle. One for turning and one for the C axis. For turning I've got a 2.2kW AC. I'll invest in some 3kW VFD and make an indexer (not sure yet how). But at least I know the direction I should investigate. Now I'm looking for a good compromise for the C axis. The machinery will act either as a lathe, or as a 4 axis milling machine, but NOT at the same time. The AC motor will be permanently engaged (but powered off while in milling mode). I need to find a good way to disconnect the C axis motor while in lathe mode (even if I have to manually disengage the belt), so there would't be any backlash involved. The lathe and the mill would share the X Y axis motors. A BT30 mill would be fitted on the Z axis, to take care of the milling operations. I hope it makes sense.

Please don't hesitate to write down anything you think it may help (or not :) ).

Cheers!

Valfar
02-03-2017, 03:56 PM
Servos position perfectly at rest, and then lock (modern ac servos).


Do these modern servos you've mentioned come with an integrated brake of some sort? Similar to what m_c has been suggesting, but integrated?

komatias
02-03-2017, 04:00 PM
Do these modern servos you've mentioned come with an integrated brake of some sort? Similar to what m_c has been suggesting, but integrated?

Some do. I have one on the Z axis of my milling machine. Requires a 24V supply to release it.

Valfar
02-03-2017, 04:06 PM
Hmm... do you think these servos with integrated brakes would be more suitable for a C axis than a standard stepper? Can these brakes operate fast enough when machining a sphere for instance, with a ball nose end mill?

hanermo2
05-03-2017, 07:11 PM
Any brushless ac servo will be vastly better than any similar sized stepper.
A break will work with a stepper, and any servo, just fine.
Numerous examples.

All my servos (9+) and suppliers (3) offered brakes as options.
A break signal is a std output from the servo connector.

The stepper torque depends on where it was stopped.
E. 3Nm stepper, 200 steps, 10 ustep driver (M542 ( gecko 251)).
== same size Nema 23 1.3 Nm / 3000 rpm / 3.9 Nm peak torque servo.

Your position at-rest vs twist depends on how much torque You have, but in terms of accuracy at 1/10 microsteps you only get the 0.3 Nm from the stepper.

The servo positions to 5.000 points (+/- 0-1-2 counts depending. Often zero error).
In each of 5.000 points you have 3 x rated torque, so a 1.3 Nm servo has 3x 1.3 Nm = 3.9 Nm to bring it back to the desired position.

So the servo has about 13 times more torque best case, and 6 times more torque for positioning typical case.

Real world.
The servo positions to 2000 positions and the stepper to 400 positions.

Servo is 5 times more accurate.

Real world.
Servo accelerates to 3000 rpm in 0.1 - 0.2 secs, real world.
Stepper in 0.5 secs to 600-1200 rpm.
Servo is 5-25 times faster in acceleration.
Depending on application and how you measure.

Real world.
Servo runs 5-2.5 times faster top speed.
Depending on application.


Hmm... do you think these servos with integrated brakes would be more suitable for a C axis than a standard stepper? Can these brakes operate fast enough when machining a sphere for instance, with a ball nose end mill?

m_c
07-03-2017, 06:19 PM
I've been busy for a few days, so have not had a chance to reply.

Regarding the cutting forces, I'm not sure if tangent was the correct term (it's been a few years since I've had to use such terms!), but imagine you have a bit round bar mounted on the 4th axis, and you want to machine a flat across the top of the bar using a vertical cutter. At the point where the cutter is directly over the centreline and at maximum cutting depth, is where you're going to get maximum torque working against the 4th axis.

You can calculate things to a reasonable accuracy, if you know the angles involved, and how the cutter torque will be getting applied to the workpiece/4th axis.


Regarding brakes, they're generally used for where you need to lock an axis in use (i.e. where you need to move to a set position and lock solidly), or when powered off (i.e. to stop a vertical axis dropping when the system is powered off).
You could potentially use one on a 4th axis, but you would have to generate code that continually unlocks, moves, then locks the motor in between machining. Otherwise you still need sufficient torque from the motor to hold things steady against the cutting forces.


Now Steppers and servos.
What you have to bear in mind, is a stepper is essentially a form of brushless servo motor. Using a suitable encoder and servo driver, you can run a stepper motor as a servo motor.
The big draw back though, is due to the internal design of a stepper motor, you get magnetic detents as the slotted rotor aligns with the permanent magnets, which affects performance compared with a properly designed brushless servo motor, which will have hardly noticeable detents.

Steppers don't lock solid. There is an air gap between the rotor and coil, so you're relying on a magnetic field to hold the rotor, which means there is a bit 'spring' to even the full step position.
As Hanermo has mentioned, microstepping reduces holding torque. The worst point is at the halfstep point, as you have two coils 50% energised, which theoretically puts the rotor exactly between the motors natural detent point, meaning the motor itself is trying to push/pull the rotor to the nearest detent point.

Where servos have the advantage, is the lack of magnetic detent improves performance, and you have an encoder. Servos are rarely perfectly held on position. They will normally always be dithering at least an encoder count or two, especially if they're subjected to any kind of varying load. Under normal use, even with perfect tuning, they will always be out a few counts, however compared with a stepper motor, servos should produce near continuous torque at any point in their rotation, and produce near constant torque over their entire rated speed range.

However, you don't really need to know any of this. If you design the system around rated torques (in the case of steppers, look at the torque/speed graph, to get the torque at the maximum speed you think you'll be machining), then you shouldn't have any problems.

Don't rely on servo peak torques, as they're more to allow for rapid acceleration. If you exceed the rated torque, depending on the motor/driver, the driver will shut down after a set time (my drivers calculate how much energy has been put in the motor, and use an algorithm to calculate if the motor has overheated), the motor may have a thermal switch to shut things down, or worst case scenario with no overheat detection system, you end up with a very hot paperweight.

hanermo2
09-03-2017, 12:55 PM
1. A slight correction.
Old-tech type servos, dc brushed servos like geckodrive 320, of which I have 7, used to dither.
The geckos are not in use anymore.

New ac brushless servos find the position commanded, and lock.
Zero dither.
They are "live" about 0.5 secs, and then lock solid.

I have 2 brands, about 15 total in use and in stock, all are the same.
From 400W / 60V, 750W/220, 2.5 kW/220V.


The bigger servos mostly have a led display, you can configure, as std it shows the error count.
So you can see the error on the led, which always goes to zero, and then the servo locks.

The moral:
A 1.3 Nm (400W) servo is the same size as a Nema 23 stepper.

A full set costs == 290€ EU 22% VAT.
A Nema 23 new stepper set == 40 + 50 + cables ==100 €.
A servo is about 200€ more / cheap small nema23 stepper, per axis.

But a Nema 34 stepper system, high voltage, with a 150€ driver, is about 250€ all-in.
The modern ac servo system is vastly better.

So the steppers are very cheap, pretty accurate, but have low dynamic range.
This means stepper systems are either accurate, fast, powerful, but not all 3.

Vast numbers of excellent routers have been made with steppers.
Including mechmate-sized 10k$ systems for workshop use.

I am not a servo-zealot by any means.
For one, you must have hw limit switches, imo.
Unlike with steppers.

Likewise, almost all lathe conversions I see are with steppers direct coupled.
This is a terrible idea. Imo. Ime.
And I tried 2000 hours.



1.
Servos are rarely perfectly held on position.

They will normally always be dithering at least an encoder count or two, especially if they're subjected to any kind of varying load. Under normal use, even with perfect tuning, they will always be out a few counts, however compared with a stepper motor, servos should produce near continuous torque at any point in their rotation, and produce near constant torque over their entire rated speed range.

However, you don't really need to know any of this. If you design the system around rated torques (in the case of steppers, look at the torque/speed graph, to get the torque at the maximum speed you think you'll be machining), then you shouldn't have any problems.

Don't rely on servo peak torques, as they're more to allow for rapid acceleration.

m_c
09-03-2017, 10:39 PM
1. A slight correction.
Old-tech type servos, dc brushed servos like geckodrive 320, of which I have 7, used to dither.
The geckos are not in use anymore.

New ac brushless servos find the position commanded, and lock.
Zero dither.
They are "live" about 0.5 secs, and then lock solid.

Unless you happen to have a constant or no load, a servo has to dither. The drive only knows how much power the motor needs to maintain position, by the motor moving off position.
Modern drives are more accurate and will dither less, but they still need to dither to obtain position. The drive display may tell you it's exactly on position, but an oscilloscope on the encoder will likely tell you otherwise.


I have 2 brands, about 15 total in use and in stock, all are the same.
From 400W / 60V, 750W/220, 2.5 kW/220V.

Who does a 2.5kW 220V servo and drive?
And is that 2.5Kw continuous or peak?

hanermo2
10-03-2017, 09:25 AM
Em...
I did mean when the servo is in-position, ie stopped.
At that point there is no dither.

Of course, if you try to move it off-position, it then activates and tries to get back into position.

I think all manufacturers make servo drives in to multiple kW or more.

Mine is 2.5 kW continuous.
Much bigger ones are available for not much more money.

30 Nm peak, 10 Nm cont.
Belt drive at 1:3, with HTD8 profile belts, 30 mm wide, taperlock pulleys.
So 90 Nm at the spindle.

The motor mount is a frame, from 30x40mm tool steel bars, and the motor is hard mounted to a steel plate 2 cm thick.
Motor is on top of HS, so heat does not distort spindle (grow it).


I use step/dir with a csmio-ip-s controller
+, enc, mpg, extra io addons



Unless you happen to have a constant or no load, a servo has to dither. The drive only knows how much power the motor needs to maintain position, by the motor moving off position.
Modern drives are more accurate and will dither less, but they still need to dither to obtain position. The drive display may tell you it's exactly on position, but an oscilloscope on the encoder will likely tell you otherwise.

Who does a 2.5kW 220V servo and drive?
And is that 2.5Kw continuous or peak?