Sevo motores - External controllers
-
Hi all.
I want to "upgrade" a BLV MGN Cube and add a third motor to the Z axis for self-leveling. I will also take advantage and change the spindles that it mounts for ball screws.
Since I have to change motors, I would like to make a change for the better (more precision, if possible and more speed).
I have searched for all the information I could about using external controllers, particularly for connecting servo motors.
I have found only one case in the forum where the use of external drivers with stepper motors seems to have worked for it.
The question is, do you know of use cases of external controllers with servo motors that work well?
One of the problems when connecting external drivers (particularly with servo motors) is that the differential signals on the expansion board are 3.3 V.
dc42, says many times:
"Most external drivers are happy with the differential signals provided by the breakout board. However if your drivers really do need 5V then you can achieve this by connecting PUL +, DIR + and ENA + from the external driver to + 5V on one of the heater / servo connectors on the breakout board, instead of to the + pins on the 6-pin output connector ".
Ok, it seems, those are most of the time, but it seems that there are cases (especially in servo motors) in which not.
That has been clear to me, but let's go to the assumption in which they are necessary:
f I use 5 external motors that need the 5V differential signal, do I connect the PUl +, DIR + and ENA + of the 5 motors to the 5V of one of the heaters?
That is, the 3 positive pins of each motor x 5 motors, do I connect the 15 to the 5V of one of the heaters?
And how will the board know which motor it corresponds to?
One of the posts where there is more information is this:
https://forum.duet3d.com/topic/17970/configuring-for-servomotors?_=1597447796699
... but in the end, the bottom line is that for whatever reason, the servos you use are not suitable for use in 3D printing.
For all that said, if someone has servo motors running on a 3D printer and could contribute their expertise, that would be great.
A greeting. -
@Pritt said in Sevo motores - External controllers:
f I use 5 external motors that need the 5V differential signal, do I connect the PUl +, DIR + and ENA + of the 5 motors to the 5V of one of the heaters?
There are no servo or step-servo drives that I know of that require 5v differential signals. 5V differential signal driver chips do not exist - we have searched for them.
Common external stepper drivers e.g. Leadshine need 5V single-ended signals. A few high-end ones need RS422-style differential signals (typical signal level 3V). High-end drivers for large motors sometimes need 24V single-ended signals because they are designed for use with PLCs.
-
I've used these drives on the step/direct interface no problem off the differential signal, despite the manual quoting 5-24VDC signals.
Here's a couple additional notes re: servos for anyone who cares:
-
The motors I use have a relatively high encoder resolution (20,000 counts/rev) which allowed me to use a high step rate. These types of servos tend to jump/stutter unless you feed them lots of steps very quickly. Because good servos will generally have configurable steps/rev settings, your best bet is to set this to the highest value that the duet can output (keeping in mind pulse timings and max speed requirements). In my example, I set my servos to 10,000 steps/rev (1/2 of 'full' encoder resolution) and then fed them steps at 800 steps/mm.
-
Expect to do quite a bit of tuning on the servo parameters for proper performance. Servos are not plug-and-play, they require tuning PID parameters, filters, etc. for your specific machine and loads. Make sure the servos you buy have good, well documented software and examples or you may spend a very long time in 'trial and error' mode here.
I think a lot of the reasons people have difficulty here is either non-familiarity with the equipment used, or using the lowest cost options which have limited tuning capabilities and poor manufacturer support.
-
-
I am not an expert.
I'm just an amateur.
I don't mind fighting and trying ... but up to a point.
I have never used servos: I have had a CNC Rotuer and currently I have a mini CNC mill, and both with stepper motors.
Now I would like to make the leap to servos ... but I see that you can even have more difficulty in 3D printing than in a CNC?I had planned to use these Servos with embedded controller:
http://leadshineusa.com/UploadFile/Down/IES-1706d10.pdf
Do you think they will suit me?
Thank you.
-
@Pritt said in Sevo motores - External controllers:
I am not an expert.
I'm just an amateur.
I don't mind fighting and trying ... but up to a point.
I have never used servos: I have had a CNC Rotuer and currently I have a mini CNC mill, and both with stepper motors.
Now I would like to make the leap to servos ... but I see that you can even have more difficulty in 3D printing than in a CNC?I had planned to use these Servos with embedded controller:
http://leadshineusa.com/UploadFile/Down/IES-1706d10.pdf
Do you think they will suit me?
Thank you.
I can't comment on the mechanical suitability, but the electrical spec indicates that they will work with 5V single-ended drive.
-
@Pritt That's a closed loop stepper, not a brushless AC servomotor (like the iHSV type I am trying to use). Since it is essentially a stepper, I would expect it to be easier to get it working in a printer than the AC servomotors I am trying to use, but I guess we'll have to wait and see how you get on with it.
-
@mrehorstdmd: No idea. Leadshine has the same type of servo and stepper motor.
The step by step is this:
https://leadshineindia.com/products/leadshine-integrated-stepper-ist-1706-nema-17
I don't know exactly the differences.
Anyway I am a bit dizzy with servos - hybrids - closed loops.
In the end I see that there is not a single case with servos that has not had problems. The only one I've seen that says it works correctly has been increasing the resolution by pulleys.
... I do not see it clearly.
Although I see that dc42 is a guy who knows a lot, he always says that "Most external drivers are happy with the differential signals provided by the breakout board.", I have also seen that even a technician of a brand has said that this it is like that and that really, at least in its brand, it needs them.
I have also seen this post:
https://forum.duet3d.com/topic/17780/clearpath-servos-with-1xd-expansion/37?_=1597066436587
It gives the impression that the Duets are not very fine with external drivers and more specifically with servos due to a problem with the electronics itself.
I say this because I have seen that the same servos that have been mentioned here are working on CNC machines, also at a lower speed, with good precision and without any problem.
So, I don't want trouble: I have looked for a few Oriental Motors step by step and will make the switch to them.
A greeting.
-
@Pritt said in Sevo motores - External controllers:
https://forum.duet3d.com/topic/17780/clearpath-servos-with-1xd-expansion/37?_=1597066436587
It gives the impression that the Duets are not very fine with external drivers and more specifically with servos due to a problem with the electronics itself.The investigation I did on this led me to conclude that the Duet 1XD expansion board I tested adequately met the published specifications provided by ClearPath.
OTOH, we know that there are some servo controllers that are designed for use with PLCCs and need drive signals much greater than 5V.
We have several OEM customers using step-servo drives with Duets.
-
@dc42 said in Sevo motores - External controllers:
The investigation I did on this led me to conclude that the Duet 1XD expansion board I tested adequately met the published specifications provided by ClearPath.
OTOH, we know that there are some servo controllers that are designed for use with PLCCs and need drive signals much greater than 5V.
We have several OEM customers using step-servo drives with Duets.So what do you mean, that the problem is with the servo?
And with the rest of the servos that have been talked about in the forum?
If you have customers that use servos, maybe they should be encouraged to post ... these servos work well and are configured like this.
That would encourage us fans to install them.
Thus, if in addition to the large outlay, what one sees coming are possible problems, it is difficult that there are many fans who decide to do so.
At least, in my case, I assure you that my intention was to install servos ... but if in addition to the great expense and headaches to choose them, you have to unite a sea of doubts as to whether they will work ... what I said ... I'll take them step by step.
A greeting.
-
@Pritt What do you actually hope to achieve with closed loop motors? If steppers don't skip steps, print quality will not change in switching to closed loop motors. If steppers skip steps, maybe there's a problem that should be fixed instead of throwing more expensive motors at it.
My experiments with servomotors are going on because I have the motors that I bought to use in the sand table where they resulted huge improvement in its operation over steppers. I do not expect to see the same benefits in much slower 3D printing, other than quieter operation.
As far as driving the external drivers using step/dir/en signals, that should be an easily solved problem- maybe you can connect a scope to verify the voltage swings at the driver inputs. Maybe there are some parameters in the driver configuration that need to be tweaked. My motors have dip switches to set the steps/rev, but I also see some programmable parameters in the drivers that appear to do the same, so I'm not sure if the dip switches or the driver parameters or some combo of the two set the steps/rev. It's behaving oddly in the printer, so I'm starting to think it has something to do with the driver parameters.
-
More than a specific objective, it was "investing" for the future.
I just have to change the motors ... and that's it, well, I thought, well, we're going to the servos and in the future maybe I'll change the belt system to ball screws where its speed would be great and I could gain precision.
What one does not expect when, in principle, "improvement" is to encounter problems ... and when I refer to problems it is not that it takes X time to configure them, but that it may not be successful.
A greeting.