Posts made by brandonh

@dc42

Yep - did a re-calibration from actual room temp and the values are back to fully stable.

Also, I removed the housing from the effector thermistor, looking for any issues, and the crimp was clean. The wire is pretty small so maybe a good crimp wasn't after 2 years of shaking. I went ahead and soldered it just in case, and I haven't had any temperature errors in about 5 hrs now. It's not proof that I've fixed the issue, but it's a good sign! About the only thing left as a possible source would be the thermistor wiring atop the effector that connects to the Duet.

@veti I have the new-style E3D with the sock, which uses a cartridge. There's a screw to hold the cartridge in, hence my confusion.

3 hrs of printing in, and no hint of the issue. That's a good sign.

Update:

I tried jiggling the board-to-effector cable and the effector crimp cable for the thermistor, and I could not see any evidence of the problem seen last night. That makes me think that nothing from the effector up is relevant. I unplugged and replugged the thermistor plug (connecting the effector to the thermistor) too. Now the problem seems to have gone.

I'm back to +/- 1C variation now. This doesn't prove out that the crimp connections are the source of the issue, but the fact that I never saw the issue for ~2 yrs and ~300-500 hrs of printing, then it became intermittent (once every ~50 hrs), then last night it was every few minutes, makes me think that this crimp experienced some kind of quality degradation over time that progressively worsened. I'll only know if after another 50 hrs I don't see the problem, but I'll consider it fixed for now, and have ordered another thermistor just in case.

In the graph below, the +/- 1C variation here seems due to not having any filament present; the increased thermal mass and increased thermal drain from filament present, plus increased cooling from having a part below, probably accounts for the sub - +/- 0.2C variation I'm used to.

@dc42 - Definitely interested in your thoughts on this.

@veti What would loosening the thermistor screw do?

@Veti (edited because I totally misread this) i'm using the screw-held thermistor. Got about two years on this one with no changes.

Also - I think I've got a good next step - re-run the same gcode file with no fan and no filament to rule out any effects there (like changing cooling as the print builds)

@veti It's a genuine E3Dv6 on a Smart Effector.

Some other issues I'm having, both related:

(1) Can't resume heating after a heater fault, unless restarting the whole board
After a heater fault, shouldn’t this clear the error: M562 P1 ? Regardless of this command, after a temp fault, any attempts to set a new temp value don't yield any change. Same thing with M562 by itself.

(2) Blob after resuming
This one I've also seen for awhile; after restarting then resuming after the print failure (M916), I typically see a large blob on the print. The extruder moves into near the starting position, seems to dump filament, and then restarts. In some cases, the blob is so large that the extruder catches on the next level and causes the print to fail.

I'm running the 2.03 Duet WiFi firmware now and have seen both today after the upgrade and since at least 2.02. Has anyone also come across these?

@dc42 - I've set M929 but this doesn't show any more detail - just that it's out of range, not that it's above or below the setpoint.

After ~10 hrs of printing I hadn't seen the issue, and then I saw it a few times in a row, within minutes. I'm used to a rock-solid temperature, and now I'm seeing something very noisy:

!

When the extruder is not moving, the value is nicely consistent:

!

I re-calibrated the extruder heater but the values were pretty similar, and now, watching the PanelDue, I see the temp drop ~20C from the setpoint nearly instantaneously, then jump back up to setpoint. When heating from cool (as in the inline pic above), I don't see this (it would take many seconds to change 15 degress!), so I really have no idea how such fast changes are even possible.

The part I'm printing now is just a cylinder and it's seeing huge noise:

I'm wondering if I have a failing thermistor or heater connection somewhere... it just doesn't seem like the temp measurement is trustworthy. The fairly weak extruder fan I have now seems unlikely to cause such a temp drop.

As always, any help would be appreciated.

Typical temperature variation is well below 1C. With a silicone sock and recent PID tuning, the values tend to range within 0.1-0.2C of target.

Thanks @dc42 - I'll add the logging for this. Is the temperature at the time of fault printed out to the log? If it's way off of the target, then I know it's a wiring issue, but if it's right below the target, then a heater or over-cooling issue would be more likely.

4 hrs into a print yesterday, I saw "heater fault" on the PanelDue. I was able to resume the print (with a blob) and successfully complete it, but for months I've been seeing these occasionally and I'm not sure what to do.

This seems random, and I'm fairly certain it's not the usual over-cooling on the nozzle or anything. I had thought the problem was related to a failing Smart Effector power connector, as I was getting effector lighting cut-outs that seemed related to either failing connectors or wires. But after replacing the crimps and wire block there, the lights don't cut out (good! confirms a fix), but I still saw a heater fault.

Besides replacing the wires and crimps for the thermistor 8-wire connector, and double checking the heater power connectors, I'm not sure what I could do. Maybe check the thermistor wire connection on the Smart Effector itself?

When this happens, what can I do to get some debug on this? Nothing specific prints to the G-code console on the PanelDue. Any other debug/fix suggestions would be greatly appreciated, as everything is great on the printer, except for this one thing. Thanks!

This looks great. Lots of questions!

• Do you have any pics of your switching extruder setup? Really curious about that one.
• What is the material for the core of the extruder?
• What do you use for a pump?

The lights are great but create moving shadows and keep cats up, which keeps me up at night. And now that I have a nearly silent printer, where all the noise is from the part cooling and hotend fans, I prefer to do long prints at night.

I'm thinking of cutting traces near the LEDs and routing small wires to an on-off microswitch, to be mounted atop the effector with a bit of glue. This seems like the easiest solution, given that I don't see extra wires. If there were some way to attach a header, that would work well too, and probably at less weight.

I've got 2x effectors - one for a Volcano and one for a regular E3D - so I wanted to check here for any ideas before making this mod twice. Thanks for any suggestions!

@mrehorstdmd said in Fans vs Berd-Air:

Printed centrifugal blower/compressor.

Same question here.. any long-term experience updates?

@chrishamm By speed graphing, I mean the ability to see the speed of the printer over time, on the UI, just like the temperature graph. This is really useful for debugging because the instantaneous readout can be hard to keep up, and just knowing how fast the printer is going can be useful for debugging slicing and machine settings. Plus if you change these settings but want to run an older g-code file it can be handy too.

I did take a look at the code and it seems pretty reasonable to add, but I doubt this is something I'd work on anytime soon.

The new DWC looks good to me. Installed super-quick and looks more modern. I'm very happy to see X axis time labels for the heater diagram - it sounds mundane but this is really useful. Good work!

Minor issues I saw:

• The developer console shows an error upon page load:
  GET http://bigkossel.local/rr_download?name=0%3A%2Fsys%2Fdwc2defaults.json 404 (file not found)
• G-code Jobs shows "No jobs" until loaded; would be good to show "loading"
• Clicking the back button for Gcode Jobs after going to a folder takes me to the Settings tab and system folder, oddly.

Suggestions:

• When I'm in a printing Status view, the thing I care about are the estimated print times. These could have the font vastly larger. In this view, speed graphing would be AWESOME for debugging.

@danal

The STL is here (not mine):

https://www.thingiverse.com/thing:1159886

I got these results with 1 or 2 perimeters and a few mm of brim to keep the boxes together, using Slic3r PE 1.41.2. I assure you there's nothing interesting in my g-code.

The only change between those two was the addition of M593 F51.6 to the config.g file and running this beforehand.

@dc42 Sure, go ahead

David, thanks for all the work you've done to make Dynamic Acceleration Adjustment (M593) available. Seriously, if you're ever in the Bay Area, I'm buying the beer (or offering my homebrews).

I've spent hours and dollars making changes to my large Kossel to chase out ringing issues, as documented here:
https://groups.google.com/forum/#!topic/deltabot/d_JHfa4qFck
... changing out the frame (1515 -> 2020), carriages, effector (printed -> Smart Effector), rods, extruder style (bowden -> flying), cooling, 12v->24v, hotend, you name it.

In 10 minutes, with a single command (M593 F51.6), I have possibly the single most obvious print-quality bump I've seen in 5 years on this printer. The pic below is a 10mm double-wall multi-test cube print; the face shown is one where the head makes a long travel move to the shown left face, then moves right, for a 40 mm/s external perimeter.

The motion is not 100% gone, but it's obviously vastly reduced. I don't know if this is due to frame flex, rod flex, the weight of the extruder connected via a bowden time above, or something else. But whatever, it's really good now. Going to try out pressure advance next.

Also, I'm very excited that the WiFi continuously tries reconnecting in post-2.0.0 builds! That was the single biggest annoyance from the past. The upgrade to 2.0.2RC7 is great.

(The original thread covering this is locked - including this link for anyone who comes across this post):
https://forum.duet3d.com/topic/5951/periodicity-of-ringing

@dc42

Thanks DC.

I see the tag on chrishamm's repo, but no downloadable release there, or on https://github.com/dc42/RepRapFirmware/releases - is this available for direct download somewhere?

I'm using a Duet to upgrade an old Kossel Air delta that has no autoleveling, so I'm trying to manually calibrate it. But the movement buttons in DWC don't seem to behave like I remember.

In the machine control tab on DWC, there are buttons marked for 'Head Movement', but the motion they incur is in the corresponding tower. That is, if I click the Z+5 button, the Z carriage moves 5mm, resulting in a new head position that modifies the Y and Z positions (say, to [0, 6.16, 6.81] from [0, 0, 5]).

This makes calibration painful when I actually want the buttons to adjust the head position, and move all the axes to move together in a coordinated way. I have to manually type Z0.5, Z0.4, Z0.3, ... then Z0.2, Z0.22, Z0.24, etc. I know that autocalibration removes this work, but I also had no need to recalibrate this particular machine for 2 yrs, once I got it calibrated, and the small printable area doesn't leave much space for off-the-shelf autocalibration options.

Is this a recent regression / intentional change / something else? Is there some way to have a button cause the motion I want (outside of writing a macro for the motions)?

Thanks. Running firmware 2.01beta1(RTOS) (2018-06-23b1) and DWC 1.21.2-b1.