Welcome to the SeeMeCNC Community

Hi guys,

I recently changed out the steppers in my Max V2 to .9 degree motors, and while I was at it, I changed the M2 belts to steel-reinforced belts. So far, so good.

When I brought it back to life, it has picked up a habit of "dancing" around as it heads down to the bed to begin its print. It'll go up and down, sometimes sideways, and even back and forth on the surface of the bed. It will then eventually begin to print as if nothing has happened.

The prints come out fine, and it hasn't crashed, but it's a bit nerve-wracking to say the least.

It's been doing this on every print to varying degrees of amusement/panic (on my part), so I don't think that it's G-code related.


Any thoughts?


Thanks!

It might be speed related - .9 steppers require more steps, so a rapid rate that works fine with the 1.8s may not work well with the .9s. (just guessing here).

g.

Yeah, that's my thought too. Maybe it's time to upgrade the controller...

But not until I get a video of it first.

I had another thought about this around 4am... the firmware may be smart enough to know that it can't move all three axes at the same time due to the higher steps per mm, so it may move a single axis at a time in order to reach the desired travel point.

g.

More likely the firmware isn't smart enough to keep the motors in sync when it can't generate the steps fast enough.

Reducing the maximum travel speed may be a temporary solution until you get a faster controller.

Hmmm, both good points.

In any case, I took the plunge and bought a Smoothie, and I just realized that I had forgot to order the display. Fixed that just now by ordering the Viki 2.

And while I've got things apart, I'm going to install ceramic fiber paper underneath the heated bed. So once I'm done with all of this, I expect that there will be an extended period of tuning required... :/

You should've gotten a Duet. Unless it's been added recently, the firmware the Smoothieboard runs doesn't use the excellent calibration system that RepRapFirmware does.

g.

I also would pick the duet, since the calibration firmware 618Pilot had for the smoothie, IIRC, had trouble fitting into the ram/flash constraints of the Smoothie, and more advanced calibration, or better web interface features would only make shortcomings of the tight resources more apparent. However, if you're already using Smoothieboards on other printers, then I would pick a common base instead of going for the absolute best option.

Admittedly, the V2 smoothieboard would alleviate a whole host of shortcomings relative to the Duet, but it's been nearly 15 months since the last mention I can find of the hardware progress. It seems they bit off a bit more than they could chew, as some of the features they were hoping would be major game changers for them, like the Edison socket are already end of life (Thanks Intel), and their decision to use a CAN-styled network system over another physical interface (gadgeteer style), while building three rather different boards, is enough of a project that with the volunteer effort and time they had, they haven't released anything in long enough that a lot of builders are going with other alternatives, meaning that financial support for it is waning, while with no test platform to work with, the software folks have a much harder time helping, and with less incentive, to say nothing of the amount of testing and bug reporting that an active, technical userbase can provide.

Hopefully though, they manage to get it solved. Competition and progress are only ever good for development of new technology. And hopefully you end up satisfied with your Smoothieboard as well.

Yeah, I debated the pros and cons of each of them, and the unlikelihood of SBv2 ever seeing the light of day. But either one of them is going to do fine, and either way, this is going to be a major upgrade from what I've currently got.

Another big thing the DuetWifi has going for it are the stepper drivers. I've got one of them in Orange Menace - it's a Rostock MAX before there was a "v1" and it's a big echo box. It's nearly silent. I've got an 0.8.5 in another v1 and it's really loud - same drivers as the RAMBo.

g.

So I've been holed up trying to get my Smoothieboard working for about three weeks now, and I've thrown in the towel. It's going back (it stopped working), and my Duet Ethernet is on its way. I've also sprung for the 7i display.

You're going to love that board.

g.

I hope so, it's been awhile with no printing. In other news, my buddy is giving me his old Ultimaker.

Well, my Duet Ethernet arrived today, and it really looks nice. I didn't know of the Duet Ethernet version when I bought the Smoothie, or I would have gone with that from the get-go. The WiFi version of the Duet was a non-starter for me since I don't want to run anything even remotely controlling my printer over WiFi.

We've got teenagers in the house you see...

Can I still use Matter Control with the Duet? What would be a good choice otherwise?

Thanks!

You can use any slicer you like, if you export the GCode to file and then upload it to the Duet using the web interface. Some slicers (e.g. slic3r and Cura) have the facility to upload to a Duet themselves after slicing.

*chants* Cura! Cura! Cura!



g.

I was thinking of manual jogging and such for the tune-in process. I suppose I can use the display...

I haven't had great luck using Cura, because I end up getting layer heights different from what I ask for. Looks like I will have to figure it out...

Well, it's homing properly, the bed and hot end heat up (but the hot end tends to throw that > 15C fault. And that's after it being auto-tuned). The X and Y axises are swapped from the Rostock norm, kinda wondering how bad that will work out...

The X & Y axes are the same - (if my collandar-like memory is correct), X is on the left, Y is on the right, Z is in the center (as you face the machine)

g.

As long as they're perfectly swapped the only impact should be that you'll need to mirror any asymmetrical parts for them to turn out right.

geneb wrote:The X & Y axes are the same - (if my collandar-like memory is correct), X is on the left, Y is on the right, Z is in the center (as you face the machine)

g.

I call it my two-step memory. Two steps and I've forgotten what I was doing.

Yeah, that's what it should be. I swapped X and Y at the connectors (both the steppers and the endstops), that should do it. I hope...

IMBoring25 wrote:As long as they're perfectly swapped the only impact should be that you'll need to mirror any asymmetrical parts for them to turn out right.

Swapping the motor and endstop leads between X and Y fixed it.

Well, it's printing! Not pretty (yet), but it prints.

Temperatures are within 1/2 degree, that's a nice change.

I did the twelve-point calibration, but it did as it has done in the past, which is not too much from what I can tell. It's high on the XY (front) side, while very low on the Z (rear) side. This is the same pattern that I've gotten using Matter Control's calibration tool for the last three times.

I've a tool in the past where I had to add an offset to the results to get the values positive and it did extremely well. I can't remember the name of tool, if anyone knows it, or has a recommendation, I'd love to hear it.

Thanks for your help guys! And Gene, you're right, I can't even hear the motors any longer!

What type of Z probe are you using: FSRs under the bed, or something else?

How many factors did you ask the auto calibration to adjust?

Z probe? I wish... Probably the next thing to be added though.

I used the 13-point calibration routine that comes with Matter Control.