[Krenovian]

I've read a lot about the Ethernet Smooth Stepper and in general it gets high marks. For those of you who have been using the ESS are there any persistent bugs or negative aspects that you've encountered? Are there any features that you would like to see implemented that aren't present? I'm looking at the ESS vs. one of the CSLab CSMIO ethernet controllers, probably the CSMIO/IP-M units. Thanks for any input you might have.

Curtis

[Spk64]

I have been using ESS for several weeks and have not had any issues. Works everytime.
Only downside for me and it is minor is supplying the 5 volts. This required a separate PSU or a multi-tap PSU. Would be nice to use 12or24 volts to power.

[Analias]

There is an outstanding issue of homing slaved axis with homing switches on both axis. The developer has acknowledged the problem but hasn't provided a fix yet. A workaround is possible. I had to disable the homing switch on one of the axis to allow homing to work.

Sent from my SGH-I337 using Tapatalk

[mactec54]

There is an outstanding issue of homing slaved axis with homing switches on both axis. The developer has acknowledged the problem but hasn't provided a fix yet. A workaround is possible. I had to disable the homing switch on one of the axis to allow homing to work.

This is not a fault of the Smooth Stepper

That most likely would never get fixed or changed, because you should not be trying to Home the slaved axis using (2 ) separate switches any way, You could damage your machine trying to do this, if you had a precision laser to set up the ( 2 ) switches/ Gantry Etc, then you may be able to do it, But even then if one motor looses some steps, you will damage you machine if you have ( 2 ) switches

I think that anybody that has the Smooth Stepper, once they have everything set up, they don't have any problems with them, I never have had anything go wrong with it

It also depends what you want to do with it, I use the U100 also, which works very well to with no problems, this only has 100Khz output the Smooth Stepper is up to 4Mhz output

[Analias]

This is not a fault of the Smooth Stepper

That most likely would never get fixed or changed, because you should not be trying to Home the slaved axis using (2 ) separate switches any way, You could damage your machine trying to do this, if you had a precision laser to set up the ( 2 ) switches/ Gantry Etc, then you may be able to do it, But even then if one motor looses some steps, you will damage you machine if you have ( 2 ) switches

I think that anybody that has the Smooth Stepper, once they have everything set up, they don't have any problems with them, I never have had anything go wrong with it

It also depends what you want to do with it, I use the U100 also, which works very well to with no problems, this only has 100Khz output the Smooth Stepper is up to 4Mhz output

In my case I have the A-axis slaved to the X-axis which is a gantry driven on both sides. Without the ability to home each side, you cannot assume that the gantry is perpendicular to the Y-axis or that it is repeatable. My homing switches are adjustable and repeatable with-in less than .0010". If both slaved axis could be homed independently for the last few thousands of an inch then you can trust that the gantry is square and perpendicular to the Y-axis.

Now granted, I've only had my CRP-4848 for the last 18 months and I wouldn't call myself an expert by any measure, but I don't see where I've strayed on this approach. If I'm wrong, please let me know how - I am teachable.

If the ESS cannot do this, it's only because of the firmware and IMHO it should support the ability to independently home slaved axis.

[mactec54]

Analias

The A-axes being slaved to the X-axes, become one, so only one switch can be used, if each axes is independent of the other then 2 switches can work, put one in the center of your machine for a better result

[ger21]

This is not a fault of the Smooth Stepper

That most likely would never get fixed or changed, because you should not be trying to Home the slaved axis using (2 ) separate switches any way, You could damage your machine trying to do this, if you had a precision laser to set up the ( 2 ) switches/ Gantry Etc, then you may be able to do it, But even then if one motor looses some steps, you will damage you machine if you have ( 2 ) switches

This is a native function in Mach3 that thousands of users use and rely on. The bottom line is the Smoothstepper does not support it properly. Call it what you want, but it makes the ESS a less attractive option for anyone that needs this function.

If you're going to build a motion controller that works exclusively with Mach3, than you should support all of Mach3's functions.

[mactec54]

ger21

Yes I agree that it should do all the functions that the control can do, But this feature should not be use to twist the Gantry to square it up, which most seem to think that is what you should do

If everything on the machine is built/setup correct, assembled square Etc, it also should never be needed

[Spk64]

What procedure should be used to "home" both screws on a double motor setup?

So far I have not setup home/limit switches on my machine and just do a manual squaring of the gantry and hope I do not lose any steps with a crash or moving the screw with power removed.

Homing each drive screw with individual home switches seems like a logical method to insure both sides of the gantry are in the setup / square position.


Sent from my iPad using Tapatalk HD

[mactec54]

Spk64

If your Gantry is not square to start with, then you can not use 2 limit switches, if the Gantry is out, & you square it some how, with having 2 switches, then you will have one side of the Gantry under tension with the other side, so one motor will be pushing the Gantry & the other side will be trying to hold the Gantry square, not a good thing to be doing, It won't run smooth when it is doing this, if your gantry is 4' or more wide, & it is out say .010" you would not even be able to measure it over the 48" if you were cutting wood/MDF Etc

[ger21]

That most likely would never get fixed or changed, because you should not be trying to Home the slaved axis using (2 ) separate switches any way, You could damage your machine trying to do this, if you had a precision laser to set up the ( 2 ) switches/ Gantry Etc, then you may be able to do it, But even then if one motor looses some steps, you will damage you machine if you have ( 2 ) switches

This function was specifically implemented in Mach3 to do what you're saying nobody should be doing.
The problem is that you're making two rather large assumptions.
One, that everyone has an extremely rigid gantry that never gets out of square, and two, that everyone is using motors strong enough to tear this very rigid gantry apart. For tha majority of users, neither is true.

The fact is that most home built gantry routers have a gantry that can flex enough so that one side can easily move a small amount independently of the other side. In most cases, if one motor loses enough steps during homing to cause a problem, binding will usually occur in the linear bearings whichwould cause the other motor to stall before it can do any damage.
And, if a motor is losing steps during homing, then it would also most certainly be losing steps during operation, which would then require some method of re-squaring the gantry.
When I first got my machine running, I had inadvertently set the homing speeds different for each side of the gantry, so that the two sides were homing at different rates. One side would get about 3-4 inches ahead of the other side, which then caused it to bind, and allow the other side to catch up. This process repeated itself a few times until I figured out what was going on and corrected it Nothing was damaged or destroyed. It just shows that my gantry beam connection to the sides is not very rigid, but in normal use the dual screw drive masks the weakness of the assembly.
This function has been in Mach3 for probably close to ten years now, with thousands of users using it. I've yet to read a single instance of it causing damage. Mainly, because the function works. Secondly, most stepper powered machines just don't have enough power to tear themselves apart.

Also consider that whenever you turn off the power to your steppers, there's no guarantee they will both be in the same position when you power them up the next time. Homing to two switches gets both motos back to a consistently "Synced" position.

Using two switches is a simple and very effective method of keeping a gantry relatively square. In a perfect world, everyone would have a oerfectly square machine that was rigid enough to stay that way regardless of how it's been used or abused. But that's not the reality that a lot of users face.

Sure, as you say, aligning two switches to get the gantry perfectly square is nearly impossible. But as you also said, they don't need to be absolutely perfect.

[mactec54]

ger21

I have seen this on many machines & used it,(Hands on) as you have, There is nothing wrong with having (2) switches,(if the machine is floppy enough) If your machine has a rigid Gantry, It is not a good idea to use (2) limit switches, most machines now are getting build better, more rigid, making squaring the Gantry even more important, so there is no need for the (2) switches,if you have a square rigid Gantry to start with

Any good machines now that are using servo's with encoder feed back, the encoder is in control, no lost steps, & if you are using or able to use the Z-pulse for homing then you have the same perfect position all the time,with only (1) limit switch being used

Getting (2) switches in sync, is harder to do, than squaring the gantry to start with

As for a machine with (2) ballscrews, you would be using (2) switches, but they would need to be in sync, or your Ballscrews will not like it (1) being loaded (tension) & the other holding tension in the opposite direction, they will bind for sure, a rack & pinion machine is more forgiving if each side is a little different, there is enough clearance every were, not to be problem, the only thing that will bind is the linear rail bearings if it is to far out of square when running

The bottom line is square the Gantry to start with, (Build the Gantry Rigid ) then you won't have any of these problems

Squaring up the Gantry by forcing the motors to do it, makes for an inaccurate machine

[xjdubber]

I used to work for a multicam distributor as a field tech. All of multicam's machines have slaved xa and xb axis, all have seperate homing switches. I know several other manufactures that also do this. In Multicam's controller a known square value is used the machine then goes up to the limit switches until triggered and continues to do this adjusting position on either side until it is within 0.002 of that number. Their machines never fell apart mechanically.

This would be a major downfall of the ESS if it cannot do that.

The other downfall I have found working on machines with the ESS is not being able to hookup to another network while programs are running.

as far as quality goes, i find its always in the details and CS-labs sure did a nice job finishing of their products.

[Spk64]

Thanks Ger,
That is pretty much how I seen it also. My gantry will move about .050 on one side before starting to get tight.
With my gantry squared up all works smoothly. It is that gradual loss of steps or potential crash that I would like to recover from quickly versus going through the squaring up process again.

Sure I could have extended my cars a little further and made it even more rigid but it is more than adequate for everything I have worked on so far.


Sent from my iPad using Tapatalk HD

[Analias]

The other downfall I have found working on machines with the ESS is not being able to hookup to another network while programs are running.

I haven't seen this problem.

I run the CRP-4848 with the ESS on the same network as the rest of the shop. It's plugged into the same network switch as everything else. I have no problems using the PC (Win7 I5 3.4GHz 16GB RAM) that controls the router while accessing other resources on the network or the Internet.


Sent from my Xoom using Tapatalk 4

[Spk64]

Yep, network is not an issue. I stuck a wireless card in to access network. A wired card would work just the same. All of my files are on Dropbox, so I have access from anywhere to the files.


Sent from my iPad using Tapatalk HD

[Krenovian]

Thanks for the responses to my queries. I appreciate the information.

Curtis