Print stops randomly

I'm a bit concerned about electrical noise from stepper wires too. As I need to extend the stepper wires anyway, I was looking for a suitable shielded twisted pair cable. My NEMA17 stepper wires seem to be 24 AWG (not printed on them, but they are a bit smaller than 22AWG hook up wire I have), which is the same as Cat5/6 ethernet cable, except Cat5/6 is single core wire and it is possible to break them when flexed, which could happen a lot on a reprap. I was thinking something along the lines of this:
[uk.rs-online.com]
22AWG stranded wire, twisted pair, two pairs, pairs individually shielded.
Though it says "These cables are not to be regarded as power cables or cables for the direct conection of the equipment to mains power supplies." but does have a rating of 300V. Any good?

Those wires will work... the "not rated for mains" means don't run 115/240VAC household power through them. That's probably more to do with liability than electrical engineering.

FWIW, another data point: I double checked the power settings on my laptop... the printer paused when the power settings dialog box opened. Only for a second at most... and then when I switched the "Turn off monitor" from 5 hours to Never, the printer paused again. I have a strong feeling this is exactly what killed my prints: the monitor turning off. I do not know how USB is related to this power saving setting... but could explain why some people are affected by it and not others. I can imagine Dell (or pick any manufacturer) would tie different subsystems in together.

Hoping this current print doesn't go to waste, too...
Chris

It got all the way through the print, but the extruder couldn't pull the PLA off the spool so I get to do more R&D on my setup...

Mounted my spool on bearings and got the print restarted around 8pm last night. Came in to work and it had frozen again. Same fault. This time 6 hours had passed. This is getting bad. Now going to try ubuntu to see if that makes a difference.

Check the USB driver settings, possibly in the device manager. There may be a power saving mode in the driver configuration for the USB controller. The ethernet controller on my laptop doesn't even work unless I turn this feature off. I think more than one person has mentioned Dell. Am I right?

As for the PLA spool binding, I had a similar problem Friday night. The filament was wrapped tight on the spool, and part of the feeding was trapped under several other strands. Crash!! When I woke up the next morning, there was a huge nest of extruded plastic all over the place, no model, and the X carriage had been pulled off the rods. Thankfully nothing broke.

I took the spool down and loosened it up.

jcabrer wrote:
> Check the USB driver settings, possibly in the
> device manager. There may be a power saving mode
> in the driver configuration for the USB
> controller. The ethernet controller on my laptop
> doesn't even work unless I turn this feature off.
> I think more than one person has mentioned Dell.
> Am I right?

Wow, checkout the attached screenshot. ONLY the "Root" usb devices had "power off" tick boxes checked. Wow, I hope this is/was my problem. I'm like 40 hours into troubleshooting this. I gave ubuntu a go, but the java host software was unable to establish communications on the port. Threw an obscure error that google said was only in the source code. So, back to windows again...

Is there anything besides the java host software for ubuntu? Sadly, it seems far easier to use windows for reprap work.

Well, back at it. If I can't get it sorted by Tuesday, I think I'll order a rapman. It's fun and all to tinker, but I'm trying to print a prototype for my business. I can't spend much more time on tinkering around with the configuration I'm using...

Yaay,
Chris

Chris,

Get yourself an SDRAMPS board for running from microSD card. Then all you need is to copy the Gcode to the microSD card, and run the machine using any simple serial terminal.

Commands to control the card can be found in the wiki here.

The board itself is designed to plug straight in to the RAMPS shield, so it is a very neat solution.

I have been running my machines like this for weeks now, and the problem only reappears when I have my laptop connected.

For information, I operate my repraps from Ubuntu, and my laptops are all Dells.
I found this page for settings USB device power settings, but I haven't made much sense of it yet.

Even with activity, the USB communications can be interrupted when the power saving feature is turned on. The reason is that the operating system polls the USB ports individually for this, and other functions. Try a little experiment: Start a print and as soon as it gets going, plug in a USB device. Im going to guess the machine stops.

I just encountered the same issue on a new reprap dedicated computer. Disabling the power off seems to have fixed my problems.

This is interesting I have only run from desktop PC's so have not seen this problem.
I am about to run from my New laptop also trying to running from one of my daughters old Netbooks as well.
Reading the whole of the thread is sounds like it is the host power saving that is causing the problem.
Looks like I will need to also solve this problem on both of these portable machines.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

I have had this happen a few times. I have now turned off the power saving option in the usb setting for my little net-book I run my printer off. So I will see how that goes.

It has not done it when I have plugged usb device in while printing though. And oddly enough it has never done it while I have been playing music in it too. (Its my entertainment media center in my shed.)

I will see how it goes. This is all with repsnapper BTW.

I don't think it is anything to do with the hardware. (Though I'm nor ruling that out.) I'm running gen6 and a lot of other people seem to have exactly the same trouble running other board. So I'm wondering if it is an error when repsnapper loads the gcode.

---

Make your Mendel twice as accurate.
[www.thingiverse.com]

Are you sure RepSnapper has finished loading the gcode before starting the print?

In my environment, repsnapper does not automatically refresh the 3d render pane. It is only refreshed when I click in the render pane. It sounds like maybe you do not let the file load complete ("Done" will show in the status bar) before clicking print.

It finished the print!!! I am so happy! I was believing this wasn't going to be possible.

Unticking the "Turn off" option in the device manager did it! (see my attached pic in above post)

The power saving features I had to disable:
- from screen saver, click on power settings and be sure that your profile never turns off the monitor or hard disks.
- (I also told it to never go to screen saver, though I may start a short print and try it)
- within device manager, go into the usb section and for each and every entry click on every tab to ensure any type of "Turn off to save power" setting is not checked.

@johnnyr : Good point. I believe the repsnapper beta version I'm using never says done, but I do wait until the green progress bar completely goes across. Also, I add my own GCode at the end of skeinforge to tell the extruder to turn off heat then go to y=0 then x=0 so that it doesn't affect the finished part. So I check for that at the end of the "result" gcode area.

Chris

Looks like your Y axis bars are out of a printer.. they look similar to bars I have removed from a Lexmark printer.

What is the object you have printed?

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

@BodgeIt : I didn't source the parts ... I got a jump-start with my printer from another forum member.

The part is a section of a small fuel tank I am prototyping. There are 4 large pieces like that one, 1 mid-size piece, and 3 small parts. Once glued together, will be something we can try to fit as proof of the drawing's concept. I used a CMM arm to scan the original part as well as the surroundings. Then I used CAD software to conceptualize and design my replacement part that's 50% bigger. I'm using the reprap to prototype a full size model before paying $$$$ to have a mold made.

This small tank+reprap is my trial run ... as some of my projects approach 3 feet long, that is a LOT of parts printed out of the reprap to be glued together... But I think this is worth it given the alternatives. As long as I have a work flow, it's ok if it takes a week to "print" out a full size model.

Thanks,
Chris

Mine kept stopping and so I have done the usb power saving settings thing with my little netbook.

Curiously it never stopped when I was playing music from it at the same time while printing!.

I just finished a large part I was printing with no problem.

So fingers crossed and thanks for the tips.

---

Make your Mendel twice as accurate.
[www.thingiverse.com]

@unexpectedly: What kind of CMM arm and software do you use? I use a Romer with Spatial Analyzer. Did you just grab a huge mesh of scanned points? What method did you use to expand the tank size?

I ordered a kit about a week ago from Makergear and have been waiting eagerly to receive it and begin building!

---

Germanicus

I'll take my payment in picture of succesful prints

johnnyr Wrote:
> Are you sure RepSnapper has finished loading the
> gcode before starting the print?
>
> In my environment, repsnapper does not
> automatically refresh the 3d render pane. It is
> only refreshed when I click in the render pane. It
> sounds like maybe you do not let the file load
> complete ("Done" will show in the status bar)
> before clicking print.

JohnnyR, you are absolutely correct. I specifically tried not doing this last night. And the print died. Oddly, it did not die at the point of importing the gcode when I hit print, but much before. I was clicking in the viewer and watching progress of the part. Instead of dying at the half way point, it died between 1/4 and 1/3. So I restarted repsnapper and reloaded the gcode, waited for it to say done, and now it's 5 hours into the print, doing well, and more than halfway done.

germanicus Wrote:
> @unexpectedly: What kind of CMM arm and software
> do you use? I use a Romer with Spatial Analyzer.
> Did you just grab a huge mesh of scanned points?
> What method did you use to expand the tank size?
>
> I ordered a kit about a week ago from Makergear
> and have been waiting eagerly to receive it and
> begin building!

@germanicus I have a romer stinger II infinite arm and I'm using Rhinoceros 3D CAD software. I scan using the point probe and use tools the same as if I were designing: lines, curves, lofts, circles, layers. I use the ball tips to help locate bolt holes, etc. To help with the design (expansion, etc) I'm using T-Splines. I cannot recommend their software highly enough. It allows for the direct manipulation of organically shaped objects by vertices, edges, and faces. My problem with T-splines is that I subdivide parts too quickly and end up with too many surfaces. I need to work on that...

I attached a screen capture of my work scanning an off-road motorcycle. I take many pictures with the camera to reference what I'm looking at in the CAD drawing. In this image, I am showing the results of scanning the bike with the rear suspension extended as well as fully compressed.

Thanks,
Chris

Hey guys,

It seems this has happened to people using various operating systems and PCs. For me it has always happened on electronics using the Arduino MEGA 1280.

Has anyone else had it happen with other electronics?

---

[emaker.io]

[emaker.limited]

@jmgiacalone :

I think we have found that the culprit is not the electronics. This is probably resulting from loading g-code in RepSnapper, but not waiting for the 100% conversion of the file before beginning the print.

In some cases, it might be related to the use of power saving features on laptops, but in this case, I think that was a red herring.

jcabrer,

I disagree. Not waiting for the g-code to load in repsnapper has never been the cause for me. Proof of this came when running at Maker Faire earlier this year. I had a file loaded in repsnapper which I was running continuously throughout the day. This problem occured randomly during those prints.

More recently, the effect of using my bench grinder leads me to believe it may be related to electrical noise.

---

[emaker.io]

[emaker.limited]

My Gen3 motherboard was really bad at picking up conducted noise down the USB or via the ATX psu, I never 100% identified the source or eliminated it completely but the machine usually stopped when I was using a big drill or switching big strip lamps connected to the same electrical circuit.

RAMPS seems a better setup - as the Mega has the USB converter on-board so you don't at least have a long TTL serial cable as per Gen3.

I have only had one freeze using RAMPS+MEGA1280 in the last 3months and I know the firmware was still running fine but the communications with the (desktop) PC had messed up and again I was using a Pillar drill and Lathe at the time.

Use a short as possible USB cable (1 or 2M) - (and also try changing it for a different type, I had a brand new dodgy 3M long one once)

---

[richrap.blogspot.com]

Is there anyone with electronics knowledge that can recommend a better way to filter the power spikes?

It all depends on if your Firmware is crashing due to electrical noise being picked up in the Reprap electronics or the PC comms is failing due to noise on the USB or something else.

I expect you are all seeing different things.

PC ATX power supplies and motherboards are usually quite well filtered to stop any conducted mains interference getting out onto USB or VGA outputs etc.
So it's actually more likely that the MEGA and RAMPS are picking up radiated electrical noise, or you have poor Ground connections or Ground loops in your system.

Good practice -
Screen your Reprap electronics (put it all in a metal box and connect the metal box to GND at your star-point).
Keep cable runs away from the MEGA and RAMPS shield.
Use separate psu for powering the MEGA (don't power from RAMPS)
Have a single point for all your ground connections, (make sure all grounds come back to the main 'star-point').
Short USB cable (and good quality).

Don't run it all from a long extension cable out to your shed or garage

And if you get desperate - Try using a UPS for the PC and Machine.

---

[richrap.blogspot.com]

Greg Frost Wrote:
-------------------------------------------------------
> Is there anyone with electronics knowledge that
> can recommend a better way to filter the power
> spikes?

+1 to the UPS
+1 to putting the arduino / embedded controller in a metal box
+1 to ground being your friend

At the end of the previous page, I put up a picture of how I routed my usb cable through EMT conduit and strapped it to ground.

... since then, I have experienced the "losing its mind" fault from the mains. I use glue to attach sections of my parts together and then a soldering gun to melt the junction for more permanent effect. One time, I plugged the soldering gun into the same source as the arduino's atx power supply. The instant I pulled the trigger on the gun, the printer stopped. (recall my arduino is powered by my notebook)

Now I've got a problem with the hot end... thermistor is reading incorrectly and I'm melting the PTFE tube. So now I'm doing like everyone else and building my own hot end. Because I think my idea is better.

I've had this issue also, but only when using Kliment's Sprinter firmware. I switched back to johnnyr's modified tonokips and the issue went away. I assumed it was a firmware issue, but this thread makes me wonder if something else was causing it.

I'm having printer stopping problems too, but if I just pause/unpause it continues merrily. When it stopped for you, was it permanent? It may be a completely separate issue, but I'm wondering if my problem could be because of the usb power saving issue as well. I'm using Ubuntu though, and I don't know how to disable usb power saving, anyone know?

When it stopped for me, it was permament in the sense that repsnapper thought it had finished the print. But the machine was still functional; the temperatures were maintained, as well as axis positions. So the firmware had not been reset.

I've tried to find out about usb power saving on ubuntu in vain. I also would like to know how to do that if anybody knows.

---

[emaker.io]

[emaker.limited]

I am having the same issue(s), here is what I have confirmed:

Running my bench grinder does indeed reset ramps - 99% of the time, but running my lathe there is no problems at all - I suspect this is due the quality of the electronics used in each appliance.

When my laptop goes into screen off mode - there is a worrying pause but it always seems to continue (I did switch off power saving on the usb port but need to double check those settings again)

Whenever it stops for no known reason, I look in the comms log and find an error repeating itself over and over - there is no way out of this, But if i make a small change to skienforge settings then most of the time the print will finish. The print almost always stops at the same point in the code!

This is all with tesla standard firmware on ramps and repsnapper (old version). I am sure the repeating error is along the lines of 'Line Number is not Last Line Number' But it is odd that it didn't happen so often on my older laptop. I am not sure what is going down with the comments about gcode not being loaded fully, I can't start a print until the gcode is fully loaded (the preview rendering is displayed)?