[mactec54]

If 1 output is running 2 motors, can each of the motor drive lines be properly homed? Is it possible due to tolerances between the drive screws or if one motor misses steps for some reason, tiny errors would accumulate over time and cause binding if both screws aren’t homed independently?
Jayne

I don't think peteeng would be using many screws for his Robot build, they do use a lot of Belt drives though.

Both motors are getting the same signal, so both should be out, by the same amount, unless one side was over loaded, they could still each be Homed using a separate switch, I don't think this would be a good idea though, some do this thing called Gantry squaring, ( 2 ) switches, if you can square your Gantry by Homing you have a structural problem with your machine, and this will always be a problem if you have to use Gantry squaring

I don't use Steppers and never will, with a Ac servo you can use the encoder Z pulse / Index pulse, which will give you the same perfect Home position every time, once you establish the Encoder Z pulse, on both motors you can then mount or clamp the Ballscrew mount in place.

The same should apply to setup your Steppers, but lost steps are lost steps, which could happen if one side of you Gantry was over loaded, this does not happen with Ac servos they don't loose steps, ( 1 ) output may not work for you, if you need independents of each screw, another way would be to use ( 1 ) motor and a belt drive to the 2 screws, if your Gantry is not very wide then this would be better than using ( 2 ) motors.

[davida1234]

I never liked auto gantry squaring because you home first and then you force the gantry to go back on only one side for the auto square function. Distortion of the gantry, however small; uneven load on the screws; forces on the gantry to carriage bolts will all add up to a sloppy gantry over time.

I do two precisely squared hard stops on each side and check the gantry on those if I think the steppers might have lost a few steps.

[peteeng]

Hi Jayne - all axes can be homed independently. Same issue with a dual drive axis. Missing steps is a major issue, means some mechanical jamming or design fault. All other motion parts are made to tolerance classes and you need to pick parts that achieve your required machine tolerance. Some "errors" like microstepping average out, some accumulate. Need to be aware and design for this. Your gantry can be driven by one BOB output and use two drives in parallel or you can use the A axis output as a slave to your other axis.

I can see the issue with running two drives from a single BOB axis and not being able to home them accurately, have to think about that....I agree with comments on structural issues if you are forcing the gantry square. The structure should be correct as best can be in the first place & then stiff enough to maintain that geometry. The robot will use servos so slightly different approach there... Now YaG/Scoot is running correctly I will start paying attention to its metrology. I pick up a large aluminium extrusion today to use as a reference beam for squaring. Still looking for a scanner or laser for this job as well.... Peter

[JayneV]

Hi everyone,
Thank you for the explanations to help me understand. I agree that using some kind of gantry squaring routine would only be masking a problem with the frame and end up causing longer term issues if the frame is continually being forced into squareness by the transmission system. I'm still having trouble visualising how an open loop system like one using stepper motors could continually operate without accumulating errors over time. Maybe I'm overthinking the situation, I'll keep following along to watch and learn more, I'm sure the penny will drop eventually.
Jayne

[machinedude]

Hi everyone,
Thank you for the explanations to help me understand. I agree that using some kind of gantry squaring routine would only be masking a problem with the frame and end up causing longer term issues if the frame is continually being forced into squareness by the transmission system. I'm still having trouble visualising how an open loop system like one using stepper motors could continually operate without accumulating errors over time. Maybe I'm overthinking the situation, I'll keep following along to watch and learn more, I'm sure the penny will drop eventually.
Jayne

Stepper motors will repeat unless you run into lost steps then you get into real problems. just because they repeat does not make them accurate in a machine. if you have lead error or backlash the stepper motor has no way to correct any of this. This is a down side to a open loop system. When you you move up into a servo system it's still only a semi closed loop. so to get around backlash and what ever else that can mess up position you get into high precision parts so the mechanical end is as tight as possible. the other approach is a duel loop feed back system for a fully closed loop with the encoder on a servo providing information and a signal from a linear scale on the load side providing information going into a PID loop making corrections when they get to far apart from one another.

so knowing how close you need to hold your sizes for what ever you plan on making lets you know how complex you need to go on the control side of machines. the price can jump pretty fast the more involved the control becomes.

[mactec54]

Stepper motors will repeat unless you run into lost steps then you get into real problems. just because they repeat does not make them accurate in a machine. if you have lead error or backlash the stepper motor has no way to correct any of this. This is a down side to a open loop system. When you you move up into a servo system it's still only a semi closed loop. so to get around backlash and what ever else that can mess up position you get into high precision parts so the mechanical end is as tight as possible. the other approach is a duel loop feed back system for a fully closed loop with the encoder on a servo providing information and a signal from a linear scale on the load side providing information going into a PID loop making corrections when they get to far apart from one another.

so knowing how close you need to hold your sizes for what ever you plan on making lets you know how complex you need to go on the control side of machines. the price can jump pretty fast the more involved the control becomes.

There is no such thing as a semi closed loop. when using servos, it is closed or it would not work.

The control can be any normal control, the duel loop is closed in the servo drive, there is nothing complex about doing this. and the extra cost is minimal, servo drives with duel feed back are common, and most manufacturer's have had them for many years.

Here is a link to one manufacture that has duel feed back servo drives. Yaskawa / Delta and most name brand servo drives, all have this duel feedback feature.
https://www.a-m-c.com/what-is-dual-l...-is-it-needed/

[peteeng]

Hi Jayne - Where you are headed you will achieve +/-0.1mm Ok. To get to +/-0.01mm is a big step in mechanical part accuracy. Then to get to 0.001mm you need to address mechanics and electronics. Then to get to 0.0001 your in a very different universe then to get to microns look up KERN mills. The C7 ballscrews you are about to buy is the limiting factor (all else being considered in a first build). Machine errors is a huge subject... Also accuracy, tolerance, repeatability, resolution all have specific meanings in engineering... Peter

[machinedude]

There is no such thing as a semi closed loop. when using servos, it is closed or it would not work.

The control can be any normal control, the duel loop is closed in the servo drive, there is nothing complex about doing this. and the extra cost is minimal, servo drives with duel feed back are common, and most manufacturer's have had them for many years.

Here is a link to one manufacture that has duel feed back servo drives. Yaskawa / Delta and most name brand servo drives, all have this duel feedback feature.
https://www.a-m-c.com/what-is-dual-l...-is-it-needed/

single loop semi closed loop is 6 of one and a half dozen of the other but your article goes into detail of the point i was trying to make. i never said duel loop was something new i just said it is an option based on needs. i have seen some drives with it built in like you mentioned but they all don't have this feature so if your drive supports it then what you say has some merit if the servo dive does not support it then not so much. but thanks for the link and bringing up the fact that some drives have this built in as i am sure the extra information will be helpful in their decision making process. The fancy duel loop drives i am sure cost more and then you still need linear scales to provide the second loop so this adds cost to the control and it is more complex as a result so this was the only point i was trying to make as being more complex. these extra cost might be minimal for some people but if the budget is tight they could add up to be more than planned. always a pleasure Mactech54 you have a great day

[peteeng]

Evening All - Todays jobs where to rewire BrevisNo1 and to do something on Scoot. After wiring Brevis I decided to put some aluminium onto Scoot. The endmills have not arrived yet but hey there was an old bit in the machine and I had a chunk of al/ply/al laminate in my hand. So I did a 30mm diameter interpolated hole via UCCNC CAM tab and ramped it 0.3mm and away we went. It hummed through it quite well. Minimal burr, easy to deburr... I did spray a small amount of the ubiquitous WD-40 on it. This bodes well as I want to do this sort of material in Frankenrouter. Its quite circular but needs a little metrology to get the size right. That's next, more calibration of the axes for Scoot... Peter

[machinedude]

i thought you sold Scoot? WD-40 is good stuff i use a lot of it in the home shop. it works pretty good on Aluminum. looks like something different was done to one of the outer holes? but if you get a proper end mill for Aluminum that might help with the burrs? a plain straight router bit might be most of the problem with your holes. it looks like it was pushing pretty good on those outer holes.

[peteeng]

Hi MD - Yes I sold Scoot but the name was popular so this model is Scoot Series 2 or Scoot-S2. I shall be changing the saddle from welded stainless steel to billet aluminium and this will be Scoot-S3.... I used WD-40 with teflon. Don't know if that helped but I have ordered some single flute upcut bits to play with. I see by tracking they are half way here... Image Alum 2 was deburred Alum 1 was off the machine... was happy with the result for an old 1/4" bit.... but the cut was light. Will play more with the new bits... Peter

Re names - Scoot was my first development router, used 16mm belts. Sold to a boatbuilder after I had taken it as far as I could go. Brevis-HD was the inception name for this thread but it became bigger and beefer then Brevis (meaning brief or small) ever was going to be so then called the project YaG after the lasers that make the parts. YaG was confusing for enquiries so reverted to Scoot but gave it a series name. Which means it can go onward forever, maybe Scoot-S124789 in year 2060, Like windows-12567 Hmmmm Peter

Heres Scoot-Delta as it was called at time of sale...

[machinedude]

Which means it can go onward forever, maybe Scoot-S124789 in year 2060, Like windows-12567 Hmmmm Peter

Heres Scoot-Delta as it was called at time of sale...

i hope we all last at least that long

[peteeng]

Hi All - I have finished threshold 2 and made some cleats for the shelving. Scoot is now feeling like a normal machine vs a schizophrenic ryobi. So onward to more stuff. Peter

[davida1234]

Man, Brevis, Brevis-HD, YaG, Scoot, Brevis#1, Scoot S2, Scoot/Yag, Scoot S1........... quite some lyrics here

[peteeng]

Hi David - I have explained the name development a couple of times in the thread. Here's the short story. Brevis is the name of a small CNC platform I built to cart around trade shows. Brevis means "small". I then decided to make a heavy duty version called Brevis-HD and that developed a bit then market info declared it should be bigger and taller, so I changed it to YaG which is the name of the cutting lasers the contractors use. YaG was fine for an in-house name but clumsy in the marketplace so I recycled the Scoot name (Scoot was the first development router I built and people liked that name) , adding a series number so it could last as long as needed like cars do. Maybe that's why it behaved badly didn't know its heritage... Peter

Machine nicknames change over time. This thread was started in jan20 so been awhile, things change. In my commercial work I've had machines called Psycho, Trex, Betsy, M30 and one that I can't print...

[davida1234]

Thanks, Peter .......... was meant to be a tongue in cheek comment.

[peteeng]

Thanks for the levity David - I've been having sleepless nights over backlash and need some mental health relief. Maybe its time to think about Christmas. I wonder how Santa did with Covid... back to sealing the washroom. Peter

[JayneV]

Did somebody say backlash???

I like the threshold Peter, looks good. Is it safe to say that Scoot-2/3 has rid itself of those pesky gremilins or is it too soon to celebrate?

Jayne

[peteeng]

Hi Jayne - I think the light has been shed on the Gremlin and it has dissolved into mush. Its totally predictable if you pull the USB power out of the BOB it goes skitzoid, put it back in and its solid. So until the next one appears; all extremities crossed & Scoot is now commissioned. I have started checking all the rails for square and level. The gantry pair need a little attention. Todays job was to make a finial for one of the shower battens. Will show that when installed but here's its profile. Need to move onto some more challenging parts... so get Frankenrouter moving... Peter

[JayneV]

Hi Peter,

I'm so glad to see Scoot commissioned and clear of gremlins. Can't wait to see how Frankie and Milli evolve. BTW, I am loving your Milli thread. It's huge so will take a long time to catch up, but it's so interesting, and yes, I have read the last few pages to try and stay up to date. Keep up the wonderful work.

Jayne