Manual - How to install TL-Smooth - Zebra Strippes Fix UMO - UM2 - UM3?

[mutley3d 3]

1 hour ago, tinkergnome said:

 

This should be already done at startup (if the correct pin numbers are defined):

https://github.com/Ultimaker/Ultimaker2Marlin/blob/master/Marlin/stepper.cpp#L698

 

pin numbers in your case should be 54 to 56 for x-axis and 59 to 61 for y-axis, right?

 

...but there are perhaps some more things to consider:

• https://github.com/Ultimaker/Ultimaker2Marlin/blob/master/Marlin/Configuration.h#L310
• https://github.com/Ultimaker/Ultimaker2Marlin/blob/master/Marlin/Configuration_adv.h#L165

 

Hi tinkergnome and good shout on inverting the enable and step signals. I am (perhaps dangerously) assuming that because the drivers i'm using are "usually" drop in replacements for 4988's, that I shouldn't have to make changes to those. But I will certainly keep this in mind now that you mention them.

 

Using the analogue pins seems to be working, although I am having a bit of trouble. In diagnosing, if I fit just one external driver, and set the pins accordingly, I have both X and Y running (one from the offboard driver, one from the original onboard 4988). But if I wire in both off-board drivers and set pins accordingly, I get one axis running.works. Just working my way through a few tests. Do you think the inverting Enable and/or Step is going to be necessary? I wouldn't have thought so given the "drop-in" replacement nature of these.

 

I am so close to my goal.

 

 

[neotko 1,417]

2 hours ago, halfie said:

 

I'm actually wondering something... Would it be possible to use a different controller (ie the duet3d) and just use the ulticontroller plugged in via ext1 and 2 like they are for the ultiboard2?

 

Actually the plug looks the same, but no, isn't possible. Well, better put, noone made the code. It's 'doable' but only if the controller is very basic since some of that pins are using also for internal stuff on the duet3d (there's some chats about that on their forum) ultimately, I though that I would miss the panel, but not at all, I use all my printers 3 with the phone, send the jobs from the computer and do the file selection, baby stepping for z addjustments, preheat, everything from the www phone, works like a charm.

[halfie 8]

Just now, neotko said:

 

Actually the plug looks the same, but no, isn't possible. Well, better put, noone made the code. It's 'doable' but only if the controller is very basic since some of that pins are using also for internal stuff on the duet3d (there's some chats about that on their forum) ultimately, I though that I would miss the panel, but not at all, I use all my printers 3 with the phone, send the jobs from the computer and do the file selection, baby stepping for z addjustments, preheat, everything from the www phone, works like a charm.

 

That's what I figured. I'm planning to splice in a Raspberry pi 3 soonish, and run it from the phone as well, it seems like a great way to keep an eye on things while I'm at work.

[neotko 1,417]

2 hours ago, mutley3d said:

 

@neotko yea but does it have the bed levelling routine? @halfie seems to be having probs with it.  Ill check it out once I have these drivers sorted, but taking one step at a time, too many changes in one go is dangerous. For now just using the Ultimaker/Marlin from their github repo

 

 

There's no need IMO for that... It just works different. You set the Z height, then move near the possition and ajust the Tool offset to mimic a calibration. OFC it doesn't have fancy menus. I did get use to that, is very fast to change a configuration, since it's just a text file, you hit reboot and you change everything (not like marlin that need's a total firmware update for some stuff).

 

Anyhow, indeed, nobody made a assisted calibration, since the duet3d supports practically all the bed calibration addons on the market. Ofc I also don't use any XD 

[mutley3d 3]

Well i'm almost there. Got each axis running when mapped and connected individually one each to on-board, and off-board driver. When I map both axis to off board drivers, i get odd behaviour from one axis. Calling it a night for now, as one of the off board driver carriers blew the voltage regulator, through plugging in an old blown driver (4988), it happens when tired. The odd axis behaviour might be down to how I was grounding both carrier boards (as per manuf' instruction), and sods law just when I'm about to eliminate that concern with separate ground wires, the magic smoke escapes.

 

It didn't help that I had a whole pack of faulty 8825 drivers throwing me off track, maybe the plane from China took a lightning hit

 

The intention is for a relatively straight forward documented method of circumventing the 4988 related print quality issues once and for all, whilst retaining the original board. From where this "idea" was a few days ago, it has come along pretty far. Of course the other option is that 2.1.5 board fitted out with TMC drivers. I have one on order. Certainly the off board driver carrier approach is being a lot more hassle but once finalized, some documentation will make it pretty straight forward. at least thats the plan.

[halfie 8]

6 hours ago, neotko said:

 

There's no need IMO for that... It just works different. You set the Z height, then move near the possition and ajust the Tool offset to mimic a calibration. OFC it doesn't have fancy menus. I did get use to that, is very fast to change a configuration, since it's just a text file, you hit reboot and you change everything (not like marlin that need's a total firmware update for some stuff).

 

This works pretty ok, but I noticed on playing with manual bed leveling that I was getting odd movement: if I jogged the wheel, sometimes it would move 0.3 or 0.5mm before the stepper kicked in and did it's move, and then if I wanted to back up, I had to jog it waaaaay back and then go back in slowly, it was odd.

 

I'm wondering if it's to do with the steps per mm... Using the 4988 my z steps is 400, but using the 2100s it was still 400. The umo thread mentioned having to double your steps per mm from 200 to 400...

 

I also wonder what would happen if I left the microsteps for z at 16 instead of dialling it back to 1, so then it's 16 (board)x16(driver). The z doesn't need to move all that much so it doesn't need speed like the X and Y

Edited March 20, 2018 by halfie

[neotko 1,417]

2 hours ago, halfie said:

sometimes it would move 0.3 or 0.5mm before the stepper kicked in and did it's move, and then if I wanted to back up, I had to jog it waaaaay back and then go back in slowly, it was odd.

 

I had tons of troubles with the endstops on the bottom. They do lack of the necessary precision (near 0.1 error) for me. I did move the endstops back to where umo uses it and now (after buying better endstops also) I get perfect z repetition along 3 printers. But that’s just my experience ofc...

[mutley3d 3]

A quick before and after photo. I think something might have been seriously wrong with the drivers on board, or maybe i'm just used to the finer prints from my other machines, but I swear my UM2's used to print better than the result on the left side. But I have same issue on both the machines. Piece on the right is after TMC mod.

 

I think the stepper carrier approach is a right royal PITA, it took a while to get the second TMCdriven axis to work, testing different remapped pins. However I think the result is pretty conclusive, at least in my case. This machine is now back to producing production parts once boxed back up. When the 2.1.5 board arrives I will fit this to the other machine, if all goes smoothly, I will order another for this machine to keep things as tidy as possible and recover the carrier boards back into my spares bin.

 

For reference the printed item is just a thin walled hollow item similar to the cylinder, but with some straight edges, rounded and sharp corners. Printed in black (ABS) so as to highlight surface striations, 0.8mm wall thickness, 0.2mm layer height. I think that is a wrap !

 

 

Edited March 21, 2018 by mutley3d
Added closeup image

[halfie 8]

This sounds dumb but could there be slop introduced into the driver over time with heat generation or something?

[mutley3d 3]

2 hours ago, halfie said:

This sounds dumb but could there be slop introduced into the driver over time with heat generation or something?

Ive hacked about with these machines quite a bit over time. They both run Flex3Drives with a suitably modified firmware but none of the changes caused the issue. I may have had an accident at some point tinkering where the board switched off but then reset, so I don't know if maybe there was a damaged component on the board, or partial failure of a driver. Having said that I have reverted to "from scratch" firmware to make sure no old late night forgotten changes playing tricks, mapped to the spare driver chip, and plenty other tests and checks, and replacement of donor parts to no avail. Going to off board driver was instant fix for me so maybe board component, but i cant recall ever getting such print quality out of the machine. Defo the best I have had, and i'm very satisfied with this level of output now.

[halfie 8]

@mutley3d are you running in spreadcycle or stealthchop? 

[mutley3d 3]

@halfie Spreadcycle...for now.

[halfie 8]

Tried flicking my 2100's over to spreadcycle on the weekend, and I think I burned out one of the chips trying to bridge that tiny tiny little gap. No matter, the A4988 Black Edition I got are fantastic, and I have just thrown it into a back room to avoid the noise.

 

I also noticed that 90% of the noise is from the extruder motor, so I'm going to chuck the astrosyn on just that, and take them off the x and y axis, I think.

[mutley3d 3]

1 hour ago, halfie said:

Tried flicking my 2100's over to spreadcycle on the weekend, and I think I burned out one of the chips trying to bridge that tiny tiny little gap. No matter, the A4988 Black Edition I got are fantastic, and I have just thrown it into a back room to avoid the noise.

 

I also noticed that 90% of the noise is from the extruder motor, so I'm going to chuck the astrosyn on just that, and take them off the x and y axis, I think.

Ouchies. I dont use the solder pads, my eyes cant cope with that small. I put a bridging wire across pins config1 and gnd, and i'm assuming from the results i see that i am getting spreadcycle mode. Certainly feels like more torque