Hy Torgeir, do you have a picture of your board wiring ?
To which ground point did you go ?
my problem is not solved , the two drivers are dead, so its not able to print anymore.
Chri
Hi Chri,
I've added pictures and details about this previous in this tread, have a look at topics on 27 of May.
I did not go externally to ground, but just made a jumper across the actual resistor on all the driver IC's.
Thanks.
Torgeir.
1
Thank you @Torgeir for the clarification. 
Silly question for @torgeir since I know how to solder but I have zero electronic knowhow. It's the same to short it than take it away and solder both points or the resistor actually needs to stay there?
Hi neotko,
No question is silly...
To remove a surface mounted component, -even a resistor, can be risky. So the best way to do this is to short the resistor, use two cord of 0.1 mm twisted copper wire.
You'll never know a proper ground on such a PCB, as there may be several points that seems to be ground, but are not!
Remember the size of the soldering tip, if you want to see what tip I'm using I can make a picture.
You'll solder one leg first, then make a short loop to the other side of the resistor, cut the twisted wire here and make the (solder) the other side. Then the resistor is 0 ohm.
Do not use to much solder, as it may float around and short something nearby.
Remove/clean the flux(solder paste) by using reed spirit and inspect to see there is no bridging to other components near by.
It's always smart to study the PCB and the place you want to solder before any work here.
Also use a good magnifier glass, 5 to 10 X.
OK. Good Luck.
Thanks.
Torgeir
2
Ah thanks for that info torgeir, didn`t checked the pages before page 6 
As i wanted to have it easier removeable i used the common connection point on the left (R81/R82/) and wired them to a ground point on the left side.
This is how it looked on my board
It may look as the left cable end ond R81 may shorten to the lower pin, but it doesn`t there is a air-gap in between, checked that three times before powering up the board.
Chri
1
Hi DR1pper,
I'll think this is caused by the same problem, the missing step.
Thanks
Torgeir.
Hi Torgeir,
Please correct me if i am wrong but if the problem in the photo's i provided (on the um2+ vs m200) were caused by the missing steps issue that is causing zebra stripes, wouldn't one expect there to be a repeating pattern in the layer misalignment in my um2+ print? It's just that i've tried printing different objects multiple times in a row with the same g-code and cannot see a repeating pattern. I do not see the layer line misalignment's occurring in the same places. For all intensive purposes, they look rather random.
Edited by GuestAre you saying the zebra stripes are located in a different part of benchy each time with the same gcode file? If so that is probably because it homes before the start of each print and the home location can move the tiniest of amounts so bench is printed the tiniest bit in a different spot. But the "missing" steps occur in the same spot it's just that benchy moved. I dont' think of the step as missing by the way - they just are smaller then the next step which is twice as big.
Note also - I mentioned this many pages ago - that you only get these step related issues on benchy because it has a plane (wall) that is tilted *both* slightly off vertical *and* slightly off the X or Y axis. If it was tilted much more the zebra stripes wouldn't be such an obvious pattern. If only one axis was tilted it would be horizontal or vertical stripes which are easily confused with other issues (z banding, ringing, cad model lines, z scar, etc). If it was flat in XZ or YZ plane you get no stripes at all.
Most models line up with X and Y and Z axes. So I'm not really surprised that this is rare and only obvious on a few unusual models like benchy.
Sorry for any confusion.
I was talking about the visibility of print layers on the m200 vs um2+ prints. Here's a more recent test i've performed showing what i'm talking about:
Both were printed in abs and even with pla versions of this test object i made, the results seem exactly the same on the um2+. Now by no means am i saying that the m200 print surface is perfect but i do consider it much closer to perfection than the um2+ print. For some reason, the layers tend to misalign to a much greater degree and with greater frequency than it does on the m200 which to me is why the m200 prints pull off an injection molded look to them more than the um2+ prints.
But if you look at the first image in this post and look at the orignal file size and zoom in you can see on the um2+ print (left one) that areas of layer alignment looks as good as the best areas from the m200 so it doesn't seem like it's not possible to achieve the same quality but for some reason it just does not do so everywhere to too the same level as the m200 everywhere.
update: This is what i mean
To the right side of the left print (from the um2+) there are areas with layer alignment as perfect as the m200. If you open the image in full you'll see what i mean. But then to the left side there are quite a lot of layer misalignments that just ruin the surface quality of the print and stopping it from looking injection molded like the m200 print(s). It's really really frustrating because it's so close and yet so far. But the fact that areas of the um2+ print are as good as the m200 print gives me hope. Just need to figure out why it's not always as good everywhere.
But i've spoken with other um2 owners and they've told me they have the exact same problem with their um2 prints. So i don't know.
Thats interesting. I only get those lines when my temp is too high. Otherwise my prints are 'pretty much' perfect. I have seen the occational line here and there but you just dont tend to get them with thinner layers i guess? I could be wrong, but i print a lot and my prints are very consistent.
Mind you my printer is next to my garden door on the kitchen floor so there is a lot of opening and closing doors and drafts? Those m200 prints do look near perfect though ill have to admit. But i need everything in pla.
Hey cloakfiend. These were printed at 0.14mm and 0.15mm respectively. UM2+ print was Prima ABS at 220*C. I'll show you the pla print tomorrow/later-today and it was printed in primavalue pla at 195 or 200*C if i recall correctly. The visibility of these "lines" on the pla print are equal to (if not more than on) the abs printed version.
Also, are you saying that the visibility of these "lines" on your um2 prints are as good (i.e. as visibly minimal and subtle) as on your m200 prints?
Edited by Guest
i print at 0.06 all the time now, used to do 0.04 but overhangs were just too annoying. i have a um2 and ill post a closeup of my surface. something looks wrong with those ramps. ive used many um2s and never seen that uneven layering. even with abs back in the day.
but i could be remebering things wrong. lol. ill get my camera now.
here goes... but im sorry i don't own a macro lens!
please note that my head is wobbly and im pretty sure i need to replace my bearings on the rods. but ill push it untill they die!
@DrR1pper and @Torgeir, I came across an interesting topic on reddit.
Obviously, the solution you suggested works because there is an immediate improvement in surface quality. But here on Reddit there is this guy who is pretty confident he achieved the same goal by changing something in the firmware, claiming the fault lies with Marlin.
What do you think about this?
Could it be that the stripes can be fixed in both ways?
Edited by Guest
A firmware fix would be great!
i print at 0.06 all the time now, used to do 0.04 but overhangs were just too annoying. i have a um2 and ill post a closeup of my surface. something looks wrong with those ramps. ive used many um2s and never seen that uneven layering. even with abs back in the day.
but i could be remebering things wrong. lol. ill get my camera now.
here goes... but im sorry i don't own a macro lens!
please note that my head is wobbly and im pretty sure i need to replace my bearings on the rods. but ill push it untill they die!
hmm, that does look practically perfect, without a hint of the issue that i am having.
Would you mind printing out my test object for me please to double check my own please? Firstly with my original gcode and secondly with your own pla slicing settings but with the same 0.15mm layer height to keep that variable consistent please. It would really help me.
g-code: http://www.filedropper.com/drr1pperprintsurfacetestpla'>http://www.filedropper.com/drr1pperprintsurfacetestpla'>http://www.filedropper.com/drr1pperprintsurfacetestpla'>http://www.filedropper.com/drr1pperprintsurfacetestpla
stl: http://www.filedropper.com/drr1pperprintsurfacetest
If you can't spare the time to do so though, no problem, i understand.
Edited by Guest@DrR1pper and @Torgeir, I came across an interesting topic on reddit.
Obviously, the solution you suggested works because there is an immediate improvement in surface quality. But here on Reddit there is this guy who is pretty confident he achieved the same goal by changing something in the firmware, claiming the fault lies with Marlin.
What do you think about this?
Could it be that the stripes can be fixed in both ways?
Interesting...
Well you are showing some bad layer issues all right! Most likely caused by bad Z movment (or possibly fluctuating temperature which is trivial to check for - just watch the temp and see if it oscillates the entire print).
What happens is the Z axis gets stuck and only moves half a layer and then later it moves a layer and a half. Or other amounts but the result is that some layers stick out a bit and some stick into the part. You get stripes. Irregular.
The fix is to take apart and clean the 6 key Z parts:
z screw
z nut
the 2 rods
the 2 bearings
My z nut has no play but on some printers it has quite a bit of play. The quality on these varies from printer to printer. But you might just have a bit of friction or dirt. You want zero oil on the 2 rods - you want them clean. Also the bearings can get dust in them especially if you oiled the vertical rods. If so take them off the rods and clean them out with WD40. People have been able to make the Z stage move perfectly with a little bit of messing with it.
Hi DR1pper,
I'll think this is caused by the same problem, the missing step.
Thanks
Torgeir.
Hi Torgeir,
Please correct me if i am wrong but if the problem in the photo's i provided (on the um2+ vs m200) were caused by the missing steps issue that is causing zebra stripes, wouldn't one expect there to be a repeating pattern in the layer misalignment in my um2+ print? It's just that i've tried printing different objects multiple times in a row with the same g-code and cannot see a repeating pattern. I do not see the layer line misalignment's occurring in the same places. For all intensive purposes, they look rather random.
Hi DrR1pper,
This is interesting I may say.
Try to turn your print object 90 Deg and make another print. To me it looks like one of your axis (X or Y) have an offset problem. The missed step may also be in here, but may be hard to visualize.
If you give us the file.gcode of the picture below, I'll give it a try with my printer.
SandervG
DrR1pper and Torgeir, I came across an interesting topic on reddit.
Obviously, the solution you suggested works because there is an immediate improvement in surface quality. But here on Reddit there is this guy who is pretty confident he achieved the same goal by changing something in the firmware, claiming the fault lies with Marlin.
What do you think about this?
Could it be that the stripes can be fixed in both ways?
SandervG,
As firmware is a software to control the hardware, many thing can be done with firmware.
Even the “configuration change” I did can be done by the firmware IF the designer made it possible.
By using the firmware to increase the current to the actual stepper motor (that's have a missing step) may cure this problem, also by increasing the voltage to the stepper motor “can” do the same.
There is not much talk about another important parameter when we discuss stepper motors. I'm thinking about the impedance (resistance) of the actual stepper we're using. The impedance and ampere tell us a lot more about how “strong” this stepper “can” be.
What we do here is to increase the energy to the stepper motor (make it stronger – a little more torque), we remember that the reduced current for micro stepping is about (10-12)% of max current.
I've also read the link you refereed to here:
This guy changed the firmware to: “repetier firmware “. How do they handle this problem I do not know, but if there was a rounding error, as he asumed, would not the error still be present on my printer then?
Thanks.
Torgeir.
Well, that doesn't sound good in the slightest bit if the case.
I did think maybe it was some slack in the x/y belts causing the position to over/undershoot slightly but when looking at opposite printed sides of a cube/cuboid along the x or y axis, you can see the same layer shift/misalignment pattern which suggests to me that it's not this. If it were a random over/undershoot caused by some slack in the x/y, i would not expect the pattern to be identical on both sides. I originally thought this might be the issue because when feeling for the tension in the belts connecting the x/y steppers to the main rods, these belts were super tight on the zortrax as compared to the um2+. I did toy with the idea of loosening the stepper bolts to make that belt have more tension but decided not to on the basis that surely the amount of tension, though less on the um2+ than the zortrax, is designed to be this way by ultimaker.
Which then lead to me think it might be either an extrusion consistency issue or z issue. My current bets are on it being a z axis related issue.
I just tried to see what the "play" was like on the z-axis for both printers and found that whilst there was some on both, the amount on the zortrax was completely undetectable by the eye, i could only feel it ever so slightly whereas on the um2+, you can visually see the play. You can move the bed on the um2+ up/down from a locked z-stepper position a good millimeter at least. Another observation is that with both printers powered off, the bed plate on the zortrax does not fall down on it's own under gravity unlike on the um2+. It takes a good amount of force/effort to push it down manually on the zortrax whilst on the um2+ if it hasn't already fallen to the bottom, you need only give it the tiniest bit of momentum and it's overcome it's static friction and very easily and freely drops down to it's lowest position. I'm not really sure how or even if at all this is related to the possibility of the z axis being at fault.
Thanks for the suggestion but i already tried printing it's longest length along the printers x and y axis and it made no difference in the outcome which only strengthens the theory that it could be a z-axis issue somehow/someway.
Thank you for offering to print it for me too. Here is the g-code i used and the stl.
g-code: http://www.filedropper.com/drr1pperprintsurfacetestpla'>http://www.filedropper.com/drr1pperprintsurfacetestpla'>http://www.filedropper.com/drr1pperprintsurfacetestpla'>http://www.filedropper.com/drr1pperprintsurfacetestpla
stl: http://www.filedropper.com/drr1pperprintsurfacetest
I will confess that i'm pretty burnt out from trying to troubleshoot this problem and getting it looking like m200 prints with respect to the visibility of layer lines but i'm starting to think it's not possible.
I've not seen or heard of anyone suffer this issue with an m200 before, though that may simply be a cause of me not looking hard enough ofc. But those i've spoken with multiple m200 workhorses have said they've not experienced anything like this on any of theirs and they all print consistently to the same quality.
Edited by Guest
i already know i would be able to print them out just fine, its definitely an issue with the printer/zscrew/bearings/moving parts, i just dont have those issues, i already did a bunch of overhang tests which i posted after i installed my OB that came out perfect. however its worth giving the zscrew a clean as gr5 suggested. but if loads of people are experiencing this then that may be a concern. i vaguely remember having something similar in my early carefree days of using the printer, and noticed a vast improvement in surface quality after many months of printing once i got a damp cloth and toothbrush and completely degreased and regreased the zscrew. tiny bits of plastic dust and fragments sometimes build up over time.
however i also remember someone printing bracelets that had that issue but then they solved it replacing something, cant remember.
Do the least intrusive fixes first before you start taking things apart! I absolutely dread taking things apart and always worry about making matters worse. just installing my OB was a nightmare but turned out ok as i thought i broke the original heater but it turned out i just cracked the casing revealing the inside, but it still worked! i went into panic mode and bought uneccesary spares of everything, but i actually prefer the old heater as i can manage it well and the bouncing temps are not an issue when you can control them.
the biggest mistake most people do in my books is to set the temps and let it go. i ALWAYS slowly drop the temps after the first layer. it is a good indicator if your nozzle is blocked. You should always know the lowest temp you can print with (without clicking/skiping) as that becomes your benchmark for hunting down errors.
if the filament is coming out exactly as you expect, then look to the screw and rods.
obviously ive not seen your printer, but ive seen some printers that look like they came out of a bombsite, dirty nozzles, chamber full of plastic dust and fragments, plastic melted around the head, all sorts. i always have a pristine nozzle. when it first heats up i always wipe it clean with a cloth. i also never change nozzles lol, even with the OB, lol. its in for life. (but thats just my choice until a better one becomes available.)
hope you sort it out and sorry for the essay, especially if none of it is relevant to you!
cleaning the z screw from top to bottom doesn't take too long, but do it perfect, otherwise theres just no point!
....and my bed does not move easily when powered off, it takes a lot of force to move it!
...another note, but check that the filament is not touching anything as its being fed into the feeder as i had to get a bit of plastic and screw the mesh netting from getting in the way (worst place ever for the mesh to come out of!), it could affect the flow hence surface quality if the filament rubs against it as mine did, (pic below) obviously i have the old um2 feeder and have no idea if this is an issue on the new ones! i also don't print on rolls....
Thanks @cloakfiend for the suggestions. I appreciate the effort you put into your response and no need to apologise for it looking like an essay at all. One paragraph, 10 paragraphs...either way, much appreciated that you take the time and effort to help.
I have lost count of how many cold pulls i've done (even using transparent nylon to ensure it's clean as a whistle against the black filament residue inside the nozzle). Fyi, this um2+ is pretty new, so any possible fault with the z i think would most likely have been from the get-go.
Yeah, i'm really not liking the idea of stripping the z down to check if this resolves it though.
i also don't print on rolls....
May i ask how come?
I also fact checked my memory of how easily the bed falls down when not powered on and i was wrong. It's not as easy as i stated (i.e. it doesn't just require a bit of momentum for it to then fall all the way to the bottom under gravity) but still has a lot less friction than on the m200. When you push it down, it sometimes has sufficient momentum to continue falling down a little bit more on its own but it then meets more friction and comes to a halt. The amount of friction is not consistent is what i mean. I don't think any of these observations play a role on the issue i'm having though. Could be wrong but just my current thoughts on the matter.
Ideally, i would really like a few independent confirmations that the um2 can print as well as the m200 in layer visibility (at the same resolution) before i even seriously consider attempting to take it apart in hopes that it'll resolve the issue. If anything, i'm quite skeptical that it'll work based on the fact that the printer is so new and that i've heard from at least one other person with both printers as well tell me that they have the same print surface quality difference in the layer visibility between their own m200 and um2. But ofc it is possible they are just experiencing the same issue as me and its just never been fixed.
Edited by GuestHi folks,
Just made some prints after all this adjustments and thoughts...
I'll kind of like this one:
They said it could be done, but none of my boats come close to this - before...
This little other object is ment as a calibration tools for those "chemical printers", I could not resist.
All is printed with Verbatim (transparent) PLA. Height: 0.1, extruder temp: 192 deg. Celsius
No heat bed (just glue stick), infill: 20%.
The "chemical printer" calibration tool; height 0.06 and other as above.
Edit: And only using the "old" standard nozzle 0.4 mm.
Thanks.
Torgeir.
1
MariMakes posted a topic in UltiMaker Cura,
ArunC posted a topic in UltiMaker 3D printers,
Recommended Posts
Torgeir 267
Thanks for the reply Torgeir. I have a limited understanding/knowledge of electronics in general let alone of stepper drivers. When you say "having installed this (A498X) series of stepper drivers" it sounds like you are saying that you installed a new physical stepper driver onto the printer. Is that correct or am i misunderstanding something.
And just to confirm, you are saying this mod should work with any other 3d printer using different electronics than the um2?
Thank you.
Hi.
The stepper driver I've been talking about is the one currently installed into the UM2 and this part number P/N is; A4988. I never changed this driver on my printer, just changed a resistor from "10 000 ohm" to "0 ohm". I have described this in detail previous in this tread.
When I talked about other 3d printers with this problem, they might have the A4988 installed or can also have a driver belonging to the "same family of drivers", meaning starting with the P/N; A498X, where X can be other number than 9.
As an example, UMO are using A4983 (if I'm right here) and later model/upgrade may have A4988 as the latest.
Thanks.
Torgeir.
Link to post
Share on other sites
neotko 1,417
Silly question for @torgeir since I know how to solder but I have zero electronic knowhow. It's the same to short it than take it away and solder both points or the resistor actually needs to stay there?
Link to post
Share on other sites