Shifting Forward and Backwards.

Thoughts about Raise3D, 3D printing and making in general.
Huntpor03
Posts: 3
Joined: Tue Mar 22, 2016 1:22 am

Shifting Forward and Backwards.

Postby Huntpor03 » Thu Apr 07, 2016 6:56 am

Hello Everyone,

I have a quick question on why my Raise N2 + started to shift Forwards and Backwards.

So... How it started, I have no Idea, loaded a file to the printer like I always do, then walked away for about two hours so that I could grab some sonic and watch new episode of Arrow. When I returned to my printer, I got this.
IMG_1105.JPG

took it off the bed and looked at it from the side view
IMG_1108.JPG


The only thing that I changed from the last print would have been that I resumed a print and never really finished the Print.

Thanks,

Hunter

User avatar
John@Raise3D
Posts: 483
Joined: Fri Feb 26, 2016 8:42 am

Re: Shifting Forward and Backwards.

Postby John@Raise3D » Thu Apr 07, 2016 10:24 am

How long have you been printing with N2+?
I would suspect the stepper driver over heat first. Open the electronic box cover when reprint and see how that helps.

User avatar
walshlg
Posts: 1350
Joined: Wed Mar 23, 2016 12:01 pm

Re: Shifting Forward and Backwards.

Postby walshlg » Thu Apr 07, 2016 1:10 pm

also make sure belts aren't loose

Andy Cohen
Posts: 170
Joined: Mon Mar 21, 2016 4:43 pm

Re: Shifting Forward and Backwards.

Postby Andy Cohen » Thu Apr 07, 2016 7:01 pm

YA know, we've seen this before. Others with the same issue. One possibility is the steppers are missing steps. I hope I am wrong. :oops:
So... Print speeds??? Going too fast can do this...
I'd move the carriage (disengage the steppers) by hand slowly and try to feel for constant movement Vs it catching or snagging. If it does snag look at the rods. With this system if the gantry rods are not nice and straight it could inhibit movement since the carriages on the sides ride these rotating rods. Grease won't help all that much.

It could also be nozzle crashes, but if it were it would not be so much and in such small increments and back and forth.

gyro
Posts: 18
Joined: Mon Mar 21, 2016 10:04 am
Location: Auckland NZ
Contact:

Re: Shifting Forward and Backwards.

Postby gyro » Sat Apr 16, 2016 7:07 am

Hi All,
I have a similar result when printing a long fluted column.
The first 360mm was perfect yet from there on the print looking quite shaky, I was initial wondering if it was the cable chain rolling as the blips are almost the same distance the the chain pivot points but I don't think that is the case.
My second thought was some racking on the lead screws or resistance in the 4 corner guides as the problem self corrects, a stepper mis-step would result in no correction, am I right?
To test the latter I started printing atop a 360mm high column of timber to see if it would repeat but that print was perfect.

The attached image shows the wobbles on the left and the attempt the recreate it on the right.
Can anyone share their thoughts on the above, as I'm nervous of a problem like this kicking in on a large (time consuming) print.
Attachments
Wobbles.png
Last edited by gyro on Sat Apr 16, 2016 9:24 am, edited 1 time in total.
Thanks in advance
Andrew@Gyro

www.gyro.co.nz

User avatar
John@Raise3D
Posts: 483
Joined: Fri Feb 26, 2016 8:42 am

Re: Shifting Forward and Backwards.

Postby John@Raise3D » Sat Apr 16, 2016 7:43 am

If it wiggles back and forth, especially when it is fine for a certain period, I would suspect stepper driver over heat the most.
@gyro, you forgot about the pictures.

Hescockrl
Posts: 29
Joined: Thu Mar 31, 2016 4:43 am

Re: Shifting Forward and Backwards.

Postby Hescockrl » Sat Apr 16, 2016 8:24 am

I haven't got my N2+ printer yet but from experience with my other printers that is a bad adjustment on the stepper drivers. I'm not sure how this board is set up yet and if the drivers have trimmer pots on them but when I get this I readjust mine till they give a VREF reading of .41v. To high and they overheat and allow skipping, too low and there is not enough juice going to the motors and the momentum allows the stepper to slip because there is not enough force to hold its position.

Do the stepper drivers on these boards have adjustment photometers on them? You may need to install cooling fans to blow across the board if they don't and overheating is a problem.

gyro
Posts: 18
Joined: Mon Mar 21, 2016 10:04 am
Location: Auckland NZ
Contact:

Re: Shifting Forward and Backwards.

Postby gyro » Sat Apr 16, 2016 9:31 am

John@Raise3D wrote:If it wiggles back and forth, especially when it is fine for a certain period, I would suspect stepper driver over heat the most.
@gyro, you forgot about the pictures.


Just thinking back, I had the bed at 70deg and suspect as it got lower in the cabinet the warm air was being pulled through the electronic box, on the second test I had the bed at zero degrees.

If that is the case do you suggest making the Electronic Box Cover Mod?

Thanks
Andrew
Last edited by gyro on Sat Apr 16, 2016 10:30 am, edited 1 time in total.
Thanks in advance
Andrew@Gyro

www.gyro.co.nz

User avatar
John@Raise3D
Posts: 483
Joined: Fri Feb 26, 2016 8:42 am

Re: Shifting Forward and Backwards.

Postby John@Raise3D » Sat Apr 16, 2016 10:28 am

I suggest to take out the current cover and apply additional cooling to the board. If the problem goes away, the make the mod. If not, then the problem is something else and we need to keep looking.

User avatar
walshlg
Posts: 1350
Joined: Wed Mar 23, 2016 12:01 pm

Re: Shifting Forward and Backwards.

Postby walshlg » Wed Apr 27, 2016 4:22 pm

Hescockrl wrote:I haven't got my N2+ printer yet but from experience with my other printers that is a bad adjustment on the stepper drivers. I'm not sure how this board is set up yet and if the drivers have trimmer pots on them but when I get this I readjust mine till they give a VREF reading of .41v. To high and they overheat and allow skipping, too low and there is not enough juice going to the motors and the momentum allows the stepper to slip because there is not enough force to hold its position.

Do the stepper drivers on these boards have adjustment photometers on them? You may need to install cooling fans to blow across the board if they don't and overheating is a problem.


I've heard this same advice several times now. When you get your printer, please post a photo to show where to test and adjust.

Jetguy
Posts: 3085
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Shifting Forward and Backwards.

Postby Jetguy » Thu Apr 28, 2016 12:11 pm

That voltage was for the extruder stepper drivers and has NOTHING to do with an XY shift.

Jetguy
Posts: 3085
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Shifting Forward and Backwards.

Postby Jetguy » Thu Apr 28, 2016 12:20 pm

Here's the facts and I brought this up very early on in the start of all this. The TMC2100 stepper driver is a silent stepper driver, but the reason why it's so quiet is it's not driving the motor as hard all the time during motion and thus not making the noise. As such, if your motion system is free moving, low mass, and reasonable speeds, you probably won't exceed motor safety torque margins and thus do not skip steps. If you skip steps, it's because you are exceeding the motor torque capacity. You can put other drivers in and decrease the chance of skipped steps, but it comes at the price of noise during printing.

This driver overheating is way overblown. If a driver goes into thermal shutdown, it's for several seconds until the driver cools meaning the axis stands still for several seconds and huge positioning error is lost. We are not seeing that. The only way to see that is if you seal up the printer, lids and doors closed, printing ABS with the bed at 110C, and on a long print where the entire inside of the printer becomes quite hot, and ONLY on the N2 and N2+ since the electronics are inside the same "chamber" as the print space. In that specific situation, then and only then is additional cooling "the fix" and that's because the electronics are being cooled by warm/hot air rather than cool air from room temp. All of those mods simply feed direct room temp air to the cooling blower. The blower is a huge 12v unit blowing like a hurricane. If that can't cool your electronics, then adding some fan isn't going to do squat.

Just trying to bring some sanity to the discussion.

Jetguy
Posts: 3085
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Shifting Forward and Backwards.

Postby Jetguy » Thu Apr 28, 2016 2:15 pm

Again, the basic problem is that silent stepper drivers were chosen to run in "stealthchopping" mode. It's literally in the documentation of the driver chip that this mode may not be ideal for all situations where skipped steps are not acceptable. I personally enjoy the silence of the TMC2100s in this mode but also know that the risk of skipping steps was higher than other drivers that may be noisier. As such, I went to great lengths to ensure and test that my XY axis was free moving and reduced the mass to single extruder. I also know that increasing speeds and acceleration values could lead to skipped steps because that's higher torque load.

What we have here is a collision of goals and problems.
The goal was silence but that comes at a tradeoff in reliability of step loss.
The goal was to ship heavy duty printers- however that same heavy duty mass being shipped can lead to minor alignment and resulting friction issues by the time it reaches the end customer.
The goal was a plug and play printer- however, current technology and goals such as silent stepper drivers running right on the margin of skipping steps and only work well if the mechanism is absolutely friction free and perfect alignment.

So yes, this is exactly what one should, could, and would expect to see, a small percentage of users might experience skipped steps on marginally operational machines with friction issues. Then add to mix novice users and lack of 3D printing experience and the failure rate rises a few percentage points.

Reference material http://www.trinamic.com/products/integr ... er/tmc2100
As configured, resistors/jumpers for CFG 1 and CFG2 are not installed, thus the driver is confirmed to be in 1/16th microstepping and stealthchopping mode.

I'm literally seeing everything said in this thread https://groups.google.com/d/msg/deltabo ... boa0em5_EJ
"What I haven't figured out yet is why these drivers seems to heat up the stepper motors more than the 4988. "

Again, stealth mode seems to heat the motors a little more than any other driver, but at the same time can skip steps under higher loads and conditions. It's not so much of a "problem" of the motor heat, just an observation. We are not skipping steps because of motor heat.

And, that same statement covers what REAL overheating acts like (AKA symptoms). Overheat means a total shutdown of the driver for longer periods of time. It's not a subtle shift in the print, it's a massive and unrecoverable loss of position, often ending up with the print head against the side of the printer. These subtle shifts are either mechanical or just loss of steps due to the way stealthchop is trying to be silent. Silent operation is a lack of force causing noise. It's a tradeoff.

If I were to recommend alternate drivers, it would be the latest ones here http://www.panucatt.com/product_p/sd6128.htm
Just note, I did test them and while not exceptionally loud, if you are used to the TMC2100s, these seem louder and there is no mistaking the difference. The point is, you (the average user) are unlikely to be able to change the mode of TMC2100s because of how they are configured.

I've included a close up of the tiny, tiny, tiny jumper resistors soldered or missing on the stock XY drivers as supplied by Raise 3D.
In addition, I replaced my Z and extruder motor drivers on my N2 with TMC2100s from http://www.filastruder.com/collections/ ... tor-driver
However, note, the Filastruder (selling OEM Watterott http://www.watterott.com/) drivers, have different default options set and thus are not by default in stealthchop mode. Second picture showing those defaults.

All this to say, great, if silence is your key goal, then you must ensure the mechanics are absolutely free and mechanically sound. If you are willing to trade off sound for reliability under less than perfect mechanical conditions and give you a little more torque safety margin, then changing to the Panucatt SD6128s is a safe bet. You could also use the same drivers that are used for the extruders (A4988s) but the SD6128s are a better driver (both thermally and internal tech).

Note, beware of other THB6128 drivers on the market. By default, the THB6128 driver IC has an inverted enable signal when compared to other typical drivers (A4988s, TMC2100s). Panucatt is the only vendor with a custom board that is inverting the signal on the board and does not require any firmware changes (basic drop in compatible).
Attachments
TMC2100 Filastruder.jpg
OEM TMC2100.jpg

Jetguy
Posts: 3085
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Shifting Forward and Backwards.

Postby Jetguy » Thu Apr 28, 2016 4:39 pm

Probably should post a real test more than anything.
#1 identify the nature of the shift. Is it minor less than 1mm or greater than 5mm?
Does it happen multiple times in a print?
Does it seem to happen later in a print on a tall object?
These all help determine the scope of what it might be a drive further tests rather than throwing things at the wall to see what sticks.
#2 If it short shifts or periodic and repeated minor back and forth shifts, are there clicking or other noises? Is this an N1 or N2. N2 has couplers that could break, N1 does not. As such, if this is possibly a backlash issue, the backlash test is simple. Use the control panel to enable the steppers. While locked, attempt to move the head back and forth and side to side with moderate pressure. Any movement other than minor flexing of the belts and really, the rods is not ideal. This should be able to identify when a loose pulley or broken coupler exists. Then unlock the motors (disable), move the head by hand in all directions and listen for friction, abnormal noise, tight spots, or other problems. It should take very little pressure to move the head with the motors disabled. This is different than when the printer is completely off, the motors are generating voltage when moved in the off state. With the PSU on, there is no additional drag on the motors.

Andy Cohen
Posts: 170
Joined: Mon Mar 21, 2016 4:43 pm

Re: Shifting Forward and Backwards.

Postby Andy Cohen » Thu Apr 28, 2016 9:10 pm

Nuthin like a detailed Jetguy post. The dual N2/N2+ have less then optimal downward cooling yet increased carriage mass.Conflicting requirements to me is like chewing aluminum foil. REAL bad.
My fear has always been that the quiet mode for the drivers selected by R3D would cause this. As such I run my N2 slower then my other machines. I have left it as a dual, though. Since I added the downward fan and removed the right side nozzle until those times I do dual extrusion I have not seen a problem. Slight bumps that are ignored by all my other machines are a skip on my N2 and a ruined print. The gantry has to run smooth, the nozzle can't bump into anything and hands must be kept away.
A driver upgrade may be in my N2's future. Reliability in a big machine like this is a bit more important IMO then how loud it is. It's out in my garage anyways.

firesped
Posts: 854
Joined: Mon Mar 21, 2016 9:23 pm
Location: Tucson, AZ

Re: Shifting Forward and Backwards.

Postby firesped » Thu Apr 28, 2016 10:25 pm

I've been working on a mod to monitor the stepper motor heat levels. I put temp probe on the 4 stepper motors I have access too. I found that the ones on extruder were at 48 degrees C. the other two were around 41-46 degrees C when running. I think with the machine sealed up, after a few hours they might be overheating. I got some heatsinks from Digikey and put them on the stepper motors, but I'm having an issue with keeping it attached to one of them. I was able to attach a heatsink to the stepper motor above the filament and I have a fan blowing towards it. That stepper is running 10 degrees C cooler then the other stepper motor in the back. I am new at 3D printing myself though. so I would look to Jeyguy for what he recommends.

I wonder if I should put the last thermal probe in the electrical box.

edited to correctly state what my mod is doing.
RL name: Michael Nolen
printers:
raise3D N2 kickstarter Early Bird
Trinus Deluxe (running smoothieware on Azteeg X5 GT board)
Monoprice Maker Select v2

firesped
Posts: 854
Joined: Mon Mar 21, 2016 9:23 pm
Location: Tucson, AZ

Re: Shifting Forward and Backwards.

Postby firesped » Thu Apr 28, 2016 11:44 pm

is the SD6128 basicly plug and play? just switch it out and don't have to re-solder anything?
RL name: Michael Nolen
printers:
raise3D N2 kickstarter Early Bird
Trinus Deluxe (running smoothieware on Azteeg X5 GT board)
Monoprice Maker Select v2

Jetguy
Posts: 3085
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Shifting Forward and Backwards.

Postby Jetguy » Thu Apr 28, 2016 11:45 pm

Remember, X and Y are the only ones stock with TMC2100s. All others, Z axis, and the 2 extruders are A4988s.
This is why the original voltage mod was for the extruders- for the very temp rise you are seeing.

I'm just concerned this thread is mixing and matching.
There is no current "official" approved mod dealing with XY. If it's skipping steps, I seriously doubt the cooling is the issue, yet every time in every thread in pops the cooling duct mod. It's a giant red herring IMO.
It's not cooling, it is a placebo and people "think" it fixed or changed the problem, only to announce days layer the issue still exists.
That's because it simply mechanics meeting the limitations of the stepper subsystem. The steppers stock have a limit and if your mechanical friction ratio is slightly higher than normal, or your print jobs and calibration is off and you are getting nozzle strikes into the print- it's going to skip steps.

Jetguy
Posts: 3085
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Shifting Forward and Backwards.

Postby Jetguy » Thu Apr 28, 2016 11:47 pm

Plug and play but also set the microstepping switches under the driver socket. You want 1/16th per stock.
I cannot recommend running higher than that because this is still an 8 bit AVR running Marlin. You will see degredation in performance when the axis steps per mm is increased as microstepping increases. Don't fall into that trap.

Jetguy
Posts: 3085
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Shifting Forward and Backwards.

Postby Jetguy » Thu Apr 28, 2016 11:49 pm

Axis direction might change too, I cannot remember if it did. Simple and easy fix, no soldering, but a point of what could happen.


Return to “General”

Who is online

Users browsing this forum: No registered users and 2 guests