After watching videos of the clear path servos on NYC CNC I have dreamed of putting a set on my Tormach. It's just a matter of time before people will be.
If the comments you added above are true then is something I will save up for.
In another thread, TMarks said
Has anyone done that- that is, just replaced the Tormach steppers -series II or III- with the Clearpath servo? The implication is that these are a very simply path to servo drive; no change in machine controller, no change to the drivers, just swap motors and done.
If it is that simple, it would appear to be a viable alternative to upgrading a series I or II instead of the polyphase steppers/drivers/control board that Tormach offers. Or even upgrading a series III. Having the closed loop motion would be an attractive alternative if there is nothing else to do.
A summary from anyone who has done the swap would be very interesting.
Similar Threads:
After watching videos of the clear path servos on NYC CNC I have dreamed of putting a set on my Tormach. It's just a matter of time before people will be.
If the comments you added above are true then is something I will save up for.
I'm very interesting in this as well because the Clear Path servos are what I'm looking to use on a Emco PC Turn 105 CNC lathe. I'd love to do a retrofit to PathPilot.
Awall
Awall - The Body Armor Dude
CoolCNCStuff_ on Instagram - CoolCNCStuff.com
What do you get out of the upgrade?
Sent from my SM-G900V using Tapatalk
Arguably, one gets a closed loop control- servo=no missed steps. Some other Clearpath reviews would suggest faster motion, too. Since the Tormach open loop seems only to lose steps under extremes (too much load, actual drive failure, like that) I am personally not sure there IS value in what looks like a $1200-1500 change to avoid a problem that doesn't happen to me. Maybe higher speeds would change that assessment.
I'm not advocating the change. Regardless of WHY it was done, has anyone decided this IS worth the cost? And if so, how was it done? And how did it work out? WAS it just a simple motor swap, nothing else required? Is it as easy as the Clearpath site suggests?
A total lack of reported attempts will be a data point as well....
I push my series 3 ad hard as it will go and I never have problems.. That being said I'd like it to move faster, I'd pay for a simple replacement option that would give be more speed.
Sent from my SM-G900V using Tapatalk
I just purchased some clear path servos for an inspection tool i designed at work.
They are very configurable with built in drivers. I've only had a chance to play with them briefly, but so far seem like quality units that are very easy to configure and integrate using their software.
I have different servos on my Torus Pro mill and love them. Quiet and fast. Servos maintain torque through the total speed range, where stepper torque drops off as the speed increases. Not that that is an issue with these mills, they seem to work fine.. But I had a stepper based mill (not a tormach). The biggest problem I had was skipping steps without knowing. I would come back at the end of the run to discover the part was all jacked up mid way through the run when it skipped and lost position . With the servos, if there is a loading problem, it generates an estop immediately, saving the part and a lot of time.
And I think we have an answer.
No one has tried it (or, anyway, haven't reported it).
I've looked at the Clearpath SDSK manual. I may be wrong, but I see no obvious way to implement the closed loop/no skipped steps OR higher speed move servo advantage without changing power supply, drivers, and probably machine controller.
Clearpath servos are a powerful design option. They are not a simple drop-in for the Tormach.
What do you see as a problem? they look like a drop in replacement to me
The closed loop part is all local to the drive. They are, in fact, servo motors
with a built in step/dir interfaced driver mounted on the back of the motor,
So they would replace both the Step motors and the Leadshine drives
Last edited by PCW_MESA; 04-19-2017 at 07:36 PM.
There was a Tormach 1100 with clear path servos for sale on the facebook tormach group some time ago, but no statement/review of the resulting combination.
Also a thread on linuxcnc and here about a 4th axis configured this way, no review of the results (other than that the OP loves them) once the wiring issue was sorted out.
https://forum.linuxcnc.org/pathpilot...t?limitstart=0
http://www.cnczone.com/forums/tormac...pathpilot.html
Last edited by tmarks11; 04-19-2017 at 08:11 PM.
Tim
Tormach 1100-3 mill, Grizzly G0709 lathe, PM935 mill, SolidWorks, HSMWorks.
Servos are ALWAYS better, because steppers ALWAYS "lose steps"! NOT!!! A properly designed stepper drive system, operated within its design limits, will NEVER lost steps. That is a fact! A servo system will NOT "save the part" any more than a stepper will. Servos throw a fault only AFTER sensing a significant "following error". They do NOT throw faults just because you are 0.001" from where you're supposed to be. In fact, the following error required to throw a fault is typically on the order of tens, or hundreds, of steps. When they throw a fault, the part is almost certainly already scrap.
The Tormach stepper systems are VERY conservatively designed, and should NEVER experience "lost steps" unless you are expecting them to do something they are physically incapable of doing. A comparable servo drive will do exactly the same, under the same conditions.
The one and only real advantage of a servo drive is a PROPERLY DESIGNED servo drive WILL give you faster rapids, and faster cutting feedrates, due to their flat torque response. But, choose the wrong motor, and you'll just as likely make things worse.
I'd be very surprised if the Clearpaths could be simply dropped in, without, at a minimum, replacing the DC power supply. Perhaps you'd get lucky and the Tormach supply would, by dumb luck, be properly sized (both voltage and current) to match the needs of a suitable Clearpath motor, but that would be nothing but dumb luck. There would for certain be cabling differences.
Overall, sounds like a lot of money to spend for faster rapids. And if you're going to spend that kind of money, why not go for the current "gold standard" in CNC drives - AC servos. Prices have come down dramatically in the last few years. I have AC servos on all my machines, and they will do 500 IPM without batting an eye, and have enough power to snap a 1/2" endmill without slowing down or even losing position.
Regards,
Ray L.
Good points Ray. Couple things to add, unless your machine runs all day I doubt you old see an increase in productivity with faster rapids. I do want faster rapids but I like to soup things up. Faster cut speeds need horsepower and rpm to be really productive. So add a higher hp motor, servos and change the pulleys to get 7000rpm...........
RAD. Yes those are my initials. Idea, design, build, use. It never ends.
PCNC1100 Series II, w/S3 upgrade, PDB, ATC & 4th's, PCNC1100 Series II, ATC, 4th
There are places where rapid speed can make a BIG difference. Two that come to mind are peck-drilling, and HSM machining, where doubling rapid speeds can cut the machining time in half. But, other than that, it won't make a huge difference in most other operations. If I could have my 'druthers, higher spindle RPM would top the list, followed by more spindle power. The former would greatly reduce machining time with small tools, and the latter would speed up roughing, rigidity permitting. I already do most of my roughing in 6061 at well over 100 IPM using inexpensive HSS tools, and with a faster, more powerful spindle I could get benefit from carbide tools, and increase MRR considerably.
Regards,
Ray L.
I agree. The closed loop part is local to the servo. So, when a repeated missed step happens -say due to an overload- how does the controller know there's a problem? The servo attempts to take the step again; the controller, knowing no better, asks for another step. The servo is still busy trying to execute a previous command. And so on. Perhaps I'm imagining a problem that can't exist, but it seems to me that a lack of an external fault communication to the controller could be a big deal.
The servos alone won't replace the steppers and drivers, as I understand the manual. the SDSK needs a direction and drive signal which looks (again, to me, and remember I'm only reading the manual- don't have one in my hand) like it comes from a driver. But, if the servo alone does replace both stepper and driver, is that compatible with the Tormach controller? I don't know. There was also a lot of stuff in the manual about having the right kind of power supply (mostly, it seemed to amount to "don't use an electronic one").
It's possible the SDSK will do as a drop-in replacement for the stepper. My question was, and remains, whether it provides the advantages of a servo without further modifications to the system.
Tim, thanks. I hadn't seen the 1100 facebook post, but that's directly on point to the question. Perhaps the owner will comment here.
The 4th axis thread is also helpful. It appears that one can swap servo for stepper on a 4th axis (and, if there, why not elsewhere?), but not as a straight drop-in; some control code changes are needed.
Per page 53 of the Clearpath User Manual, the minimum "tracking error limit" is 1/4 rotation of the motor. This is the mininum error that will cause the servo to fault, and stop moving. 1/4 turn on a 5mm ballscrew means the machine will, at the point the fault is triggered, be 1.25mm from where it was commanded to be. So much for myth that "the servo will save my part". Even with that, unless you add wiring to trigger an E-Stop when ANY of the drives faults, you'll never know it happened until you look over and see that's it's not cutting what it's supposed to be cutting. Just like with steppers. With that added wiring, the machine will stop, but only after your part is already scrapped.
Again, a properly designed stepper system (which the Tormach IS), operated within its capabilities (which is ALWAYS up to the operator, not the hardware), will NEVER "lose steps". So, what problem is being solved here? As I said, I have AC servos on all of my machines, and they ARE wired to E-Stop. Not ONCE has having servos "saved the part". If I program a move the machine cannot physically make, the part is ALWAYS scrapped, long before it throws the E-Stop.
Regards,
Ray L.
As i know step and dir are no coming from any driver , they come straight from BOB wich is connected to pc .
The Clearpath drives (which are 75V AC servos) have an output that can be programmed as a fault output that is asserted when a following error occurs
The SDSK drives take step/dir signals the same as the Leadshine drives that Tormach uses
Because they have a step/dir interface and there is no encoder feedback to the controller you lose some of the advantages
of a full closed loop servo system that LinuxCNC/Pathpilot can support (real time smart following error monitoring, homing to index
no need to rehome on estop/faults etc)