PanelDue, Motors all skipping/stuttering, and endstop LED question.



  • I recently ordered a DuetWifi with PanelDue from Filastruder here in the states, and have had tons of issues since the order, with no help from them so I thought I'd post here. I have two main issues I am trying to solve first. First off, I am testing it on my Geeetech Me Creator 1 before dropping it into my big CoreXY 300x300x300 build.

    Question 1: I had ordered the PanelDue 7" kit, but only received the LCD, not the little breakout board and wire to connect it to the DuetWifi. Is it possible to connect the LCD without the little board, as they aren't answering any of my emails?

    Question 2: I have a ton of motor skipping/stuttering issue that I cannot track down, and I am wondering if I have a dud board. I disconnected all things from my Geeetech machine, and reset the config files to stock and tried it on my Giant Kossel (Running an older Duet 0.6 btw, love the duet products!), and the issue followed. So I redid the code for my CoreXY, and this time left the bed, hotend, fans, and all other motors disconnected. Only connected one motor to the X spot, following the main wire diagram, ensuring that I had it in the order of 2B on the SD card side, then 2A, 1A, and ending with 1B on the power input side. Set it to move by 10, and it clicks and shakes, but doesn't spin. I tried all 5 motor outputs and got the same. I have connected 12 different Nema 17's to this thing now, so I am sure it's not motors. Has to be coding or board. Here is my movement portion of my config.g:

    ; Machine configuration
    M569 P0 S1 ; Drive 0 goes forwards (change to S0 to reverse it)
    M569 P1 S1 ; Drive 1 goes forwards
    M569 P2 S1 ; Drive 2 goes forwards
    M569 P3 S1 ; Drive 3 goes forwards
    M569 P4 S1 ; Drive 4 goes forwards
    ; If you use an endstop switch for Z homing, change Z0 to Z1 in the following line, and see also M558 command later in this file
    M574 X1 Y1 Z1 S2 ; set endstop configuration (X and Y and endstops only, at low end, active high)
    M667 S1 ; set CoreXY mode
    M350 X16 Y16 Z16 E16 I1 ; set 16x microstepping with interpolation
    M92 X80 Y80 Z2560 ; Set axis steps/mm
    M92 E100:100 ; Set extruder steps/mm
    M906 X800 Y800 Z800 E800 ; Set motor currents (mA)
    M201 X800 Y800 Z15 E1000 ; Accelerations (mm/s^2)
    M203 X5000 Y5000 Z50 E5000 ; Maximum speeds (mm/min)
    M566 X600 Y600 Z30 E20 ; Maximum jerk speeds mm/minute
    M208 X150 Y150 Z150 ; set axis maxima (adjust to suit your machine)
    M208 X-8 Y0 Z-0.5 S1 ; set axis minima (adjust to make X=0 and Y=0 the edges of the bed)
    G21 ; Work in millimetres
    G90 ; Send absolute coordinates…
    M83 ; ...but relative extruder moves

    Question 3:
    This isn't all that important, but an aesthetic thing. My endstops are all functioning correctly, however, the LED is on when they are not depressed, and off when they are depressed. Is there a way to inverse this?

    Thanks for any help/info!



  • I may be wrong, but your endstop logic might be messed up. Here's a way to find out. If you have your endstops plugged in and not triggered, look at the board at the leds next to the motor connector. If they're lit up, that means they are programmed to be triggered when they're not. I did that my first time and got the same shaking, because it thinks it's hitting an endstop, goes back to hit it again, but hits an endstop, and repeats. If this is the case, you need to change endstop configuration to active low.
    I'm sure you'll get more help with this, I'm pretty new, but I know how that light works.

    As far as contacting tim goes, he's been real good to me, and I see him on reddit a lot, so maybe try reaching out to him there, he's really got great customer service, by my experience as well as others I've heard from.



  • How do I change that setting? I've tried:

    M574 X1 Y1 Z1 S0
    M574 X1 Y1 Z1 S1
    M574 X1 Y1 Z1 S2

    All 3 in that spot give me the same motor shake/stutter. Even the extruder axis does it though, and no endstops hooked up there.



  • As a side note, I looked through the GitHub to answer my first question. Luckily I have an ARM Microcontroller, the ATSAM3S2BA-AU, so I should be able to do it. Just gotta dig up all the capacitors and such.

    But yeah, on the motor issue, didn't think of it in my last post, but I even tried a motor on every driver with no endstops hooked up (Yes I did turn off the power before swapping it out).



  • Tim here, from Filastruder. I've looked through emails and the only one I'm seeing from this week is the one sent at 1AM EST today, about 7 hours ago. I've replied to that one, concerning the Duet's stepper outputs. The Duet is tested before shipment and it seems very unusual that all stepper drivers are faulty, so there may be some troubleshooting needed. The only other email I saw was last week about adding items to your order, and I replied to that one within 2 hours.

    Regarding the LCD - I looked at your order and you only ordered the LCD, you did not order the PanelDue.

    What other issues were you having?



  • @elmoret:

    Tim here, from Filastruder. I've looked through emails and the only one I'm seeing is the one sent at 1AM EST today, about 7 hours ago. I've replied to that one, concerning the Duet's stepper outputs. The Duet is tested before shipment and it seems very unusual that all stepper drivers are faulty, so there may be some troubleshooting needed. Rocking back and forth like you describe often means a miswired stepper.

    Regarding the LCD - I looked at your order and you only ordered the LCD, you did not order the PanelDue.

    What other issues ("tons of issues") where you having?

    I am following this diagram:

    https://duet3d.com/wiki/Duet_WiFi_wiring_diagrams

    I have worked with steppers professionally for many years. I am taking it in to work today though to have a colleague of mine see if it is just me and my equipment here. Always a possibility.

    On the LCD, I was sure I'd ordered the one with the Ormerod Board. But then again, I was sure I'd ordered 3 E3D V6 Heatsinks only, but after the order was placed, I saw they weren't on the order. I emailed 2 times on that, then again about that LCD when I first received the package using the email from my order, but hadn't received word back. After the previous poster had mentioned how good you are on replies, I thought it may have been a bad email address or even that laptop I was using to complete the contact form with so did it on another pc with another email. I've even tried to sign up to the website so I can actually log in, but when I click log in, it tells me "Invalid login credentials." I signed up with the original order, then again when I was trying to find order info after I saw the E3D heatsinks weren't on the receipt. Just checked the receipt again and it shows LCD only. Gah. Well, I'll be ordering that Ormerod later today.

    I sent about 4 or 5 more times through that contact page on the website to find troubleshooting stuff on the Wifi board since I got it in the mail 24 hours ago. I was trying to find wiring diagrams, etc. Previous Duet boards I ordered came with a little pamphlet in them showing the wiring schematic, pin layout, etc, but this one came in with nothing. So after a lot of searching, I found this site through github where I always get my Duet firmware.

    The original email used to sign up order was xxabyss23xx@yahoo.com. Maybe it's mostly been issues on my internet end, and I am sorry for that. I had 3 other orders that came in bad with it, one of which I ordered 3 E3D Titans from Amazon for $178, and got 3 cheap MK8 brass kits in its stead, and all of my extrusions for this build from SMW3D came in at the wrong lengths, and 2 rods from another company came in bent, so I've been quite frustrated this week. Really sorry if that's came out in my post. Of every build I've attempted from scratch, this has been the worst!

    On the board itself, here are the problems still persisting:

    1: When I run an M119, it reports that only Zprobe and extruder endstops hit, but the lights are on for x, y, and z. I can manually push in an endstop switch, and the light goes off then. I've changed out the code for them being active high/low to no avail. It has stopped reporting as triggered on the M119 command finally though since I have left it set to M574 X1 Y1 Z1 S1.

    2: If I tell it to move say X +10 or -10, X and Y motors will jitter back and forth for a bit then stop, but no smooth motion. Same for Y. Z will move, but it is very jerky. When only hooking a single motor up: The X jitters back and forth no move, moving it to Y, jitters but no move. On Z it will move about 3 turns of the shaft, stop, move 2 or 3 more and stop, etc until it completes the movement amount. Scoping it revealed current to be passing into Y and X concurrently when sending a move command to only one, not sure why there. It's reading at 400mA at each driver when the config is set to 800mA per channel. Might be firmware, I did update to the newest on everything as soon as I got the board. For the Z, it is showing intermittent spikes of current, which appeared normal to me, but then I decided to check the output for both coils, and one coil was getting 1/2 the power of the other coil. E1 and E2 send out nothing when sending extrude commands. The heater will heat up and reports roughly the right temperature without even tuning it yet.

    3: The wifi has lost connection many times and is sitting next to my router. My phone and laptop cannot connect to it. I click connect on it's server page, it goes yellow and says "Connecting…" but just stays there. It will still connect fine on my server machine though. This all worked fine before I updated the firmware. My current firmware looks like this:

    irmware Name: RepRapFirmware for Duet WiFi
    Firmware Electronics: Duet WiFi 1.0
    Firmware Version: 1.17e (2017-02-10)
    WiFi Server Version: 1.03 (ch fork)
    Web Interface Version: 1.14a
    Web Interface by Christian Hammacher
    Licensed under the terms of the GPL v2

    I just decided to try something else this morning. I connected a cheap external pololu shield and all 3 motor drivers on it worked flawlessly when setup. I now have it moving around fine on x, y, and z running from that board using the expansion header and using M584 X5 Y6 Z7 and the M569 to set the active and pulse width. While at work today, I will try to also reinstall firmware on it using their servers in case it was a bad download. In the meantime, any suggestions would be appreciated. I have a full slew of test equipment available here and at work (I teach Space Systems Engineering at a local University),



  • @BReeves23:

    I am following this diagram:

    https://duet3d.com/wiki/Duet_WiFi_wiring_diagrams

    I have worked with steppers professionally for many years. I am taking it in to work today though to have a colleague of mine see if it is just me and my equipment here. Always a possibility.

    I'll leave the Duet diagnosis to the experts/designers (dc42, Tony, Roland). But if a replacement is necessary we're absolutely happy to take care of that.

    @BReeves23:

    On the LCD, I was sure I'd ordered the one with the Ormerod Board. But then again, I was sure I'd ordered 3 E3D V6 Heatsinks only, but after the order was placed, I saw they weren't on the order. I emailed 2 times on that,

    I did receive those emails regarding the heatbreaks, and replied about 2 hours later. Looking more closely, it seems like my reply didn't get to you because of the email address entered in the contact form (which differs from subsequent emails):

    I'm happy to waive the shipping on a second order if you'd like, since it was just a communications problem.

    @BReeves23:

    I've even tried to sign up to the website so I can actually log in, but when I click log in, it tells me "Invalid login credentials."

    It looks like you did not set up an account when you placed your order, thus there are no valid login login credentials. I'm sending an account invite.

    @BReeves23:

    then again about that LCD when I first received the package using the email from my order, but hadn't received word back. After the previous poster had mentioned how good you are on replies, I thought it may have been a bad email address or even that laptop I was using to complete the contact form with so did it on another pc with another email. I sent about 4 or 5 more times through that contact page on the website to find troubleshooting stuff on the Wifi board since I got it in the mail 24 hours ago. I was trying to find wiring diagrams, etc.

    I don't see any messages from anyone with your last name concerning an LCD, and the only message I have regarding the Duet Wifi is the one from this morning. I haven't heard of any other problems with the contact form, but I'll look into it. You can also use the email address listed on the contact page to get in touch more directly. (support@filastruder.com).

    @BReeves23:

    Previous Duet boards I ordered came with a little pamphlet in them showing the wiring schematic, pin layout, etc, but this one came in with nothing. So after a lot of searching,

    I actually suggested this to the T3P3 guys a few weeks ago, they're working on something now.

    @BReeves23:

    I've been quite frustrated this week. Really sorry if that's came out in my post. Of every build I've attempted from scratch, this has been the worst!

    That does sound frustrating, and I'm sorry for what we contributed to that! I expect David or Roland will be along shortly to assist with the Duet Wifi. I'm refunding $10 for the hassle, hopefully that helps.


  • administrators

    Hi,

    Regarding the motors not moving consistently:

    1. Are you certain that you have identified the phases of the motors connections correctly? See https://duet3d.com/wiki/Connecting_stepper_motors#Using_the_internal_drivers.

    2. Make sure that you are not trying to move them too fast or with acceleration that is too high, especially if you have increased microstepping, or you are running the motors well below their rated current.

    3. Make sure the power supply voltage is not too low or is dropping when you drive the motors. If the voltage drops below approx. 9.5V then the drivers will be shut down until the voltage reaches 10V again. You can see whether this has happened by sending M122 and looking to see if any under-voltage events have been recorded. Check that the VIN terminal block screws are tight.

    Regarding the endstops, the endstop LEDs on the Duet light when the switches are closed. Therefore when you use normally-closed switches (which is recommended), the lights are on when the switch is connected but not triggered, and of when it is triggered. This is described at https://duet3d.com/wiki/Connecting_endstop_switches.

    HTH David



  • @elmoret:

    I did receive those emails regarding the heatbreaks, and replied about 2 hours later. Looking more closely, it seems like my reply didn't get to you because of the email address entered in the contact form (which differs from subsequent emails):

    http://i.imgur.com/uf3Xr6c.png

    I'm happy to waive the shipping on a second order if you'd like, since it was just a communications problem.

    I checked my server this afternoon and it appears I was losing lots of packets last week in the upstream. When I placed that second order today, it took 5 times for me to get the credit card to accept. No need to refund anything, this was entirely on my side. Besides, I added that PT100 board, so shipping was free. 🙂 No need to rush it though. The other company is re-cutting my extrusions at no cost as this was their error. Fabrication of the extrusions is going to be at a week time frame though due to the dimensions, so I'm in no hurry. 🙂 Thank you again for your wonderful service.

    @dc42:

    1. Are you certain that you have identified the phases of the motors connections correctly? See https://duet3d.com/wiki/Connecting_stepper_motors#Using_the_internal_drivers.

    They are 6 wire variety, but they came with the phases already split out on a connector at 2B 2A 1A 1B. I tested the resistances last night to ensure the phases were correct, and even went so far as to test the motor with an LED. I know, ghetto, but was afraid maybe my scope and meter were failing me. When connecting them to a sanguinololu, Duet 0.6, or a UltiController v2.1 board, the motors all performed flawlessly.

    @dc42:

    2. Make sure that you are not trying to move them too fast or with acceleration that is too high, especially if you have increased microstepping, or you are running the motors well below their rated current.

    I tried lowering the accelerations and movement rates to 1/4th of what they are. I've had microstepping at 1/16 the entire time until I get everything ironed out. I bumped the rated current all the way up to 95% max, still the same problem. If I only have 1 motor hooked up on the X output and I send a +10 on X, it jitters, but will move about 10 - 15 down the rail, stop, then go back around the same amount, but not exactly the same each time. If I send a +10 on Y, it jitters and will move slightly. Z and both E's it remains motionless as it should.

    @dc42:

    3. Make sure the power supply voltage is not too low or is dropping when you drive the motors. If the voltage drops below approx. 9.5V then the drivers will be shut down until the voltage reaches 10V again. You can see whether this has happened by sending M122 and looking to see if any under-voltage events have been recorded. Check that the VIN terminal block screws are tight.

    Checked this purposefully at work today just to make sure. I am using an Corsair AX760 that has a single 12v rail at 756w. I know overkill, but this power supply was having to power a 12v 300x300 heater in another build. My kossel machine that I tested it on is a Mean Well NES-350-12 12V 350W DC power supply. At work, a few of my colleagues helped me test it using several other 4/6/8 wire nemas we had laying around. I used a Keysight Technologies 6652A variable power supply to ensure smooth power during this. I scoped it again with an Agilent MSO9254A, a bit better than than one I used at home. There is on average 43% of the amperage that is supposed to go out the X output going out the Y output.

    I did think to check the traces while I was there. I didn't use the x-ray, but just a standard scope and ensured there were not solder or trace breaks across the X and Y outputs, and made sure that they weren't connected in anyway so I'd say it is on a software level. I will take time to dig through all the firmware and flash it all with the most up to date from github on Monday.

    I've actually plugged in and unplugged motors so many times one of the plastic connectors on the board is now broken, but the pins are fine. It was just the tall support piece that the connector locks into. I am going to take a break from troubleshooting it over the weekend. I'll check back though if there are any other suggestions, and report back after I have re-flashed it Monday.

    @dc42:

    Regarding the endstops, the endstop LEDs on the Duet light when the switches are closed. Therefore when you use normally-closed switches (which is recommended), the lights are on when the switch is connected but not triggered, and of when it is triggered. This is described at https://duet3d.com/wiki/Connecting_endstop_switches.

    Thanks! I didn't see this in one spot on doing the endstops, but found it in another on the wiki after you mentioned this. I swapped them out for normally-closed switches and the LEDs all went out.



  • I took the advice of a local friend that has this board and tried to SSH into the SD card to look at the firmware. I was getting some errors when trying to open the files. So I decided to just pull it out and look at it on my pc using an SD card reader. It immediately popped up a file structure error and told me to format the SD card. It failed in the attempt. So, I decided to go to the github (dc42, your repository of files here is awesome!). I grabbed the Sys-CoreXY build, popped it on a fresh SD card, and ran through the configurator for detailed options. Works like a charm now.

    I don't know if my original firmware update failed due to a power/net issue, or it was just a dud SD card destined to fail, but all is awesome now. I want to thank you guys for your help and advice. The M122 code tipped me off a bit on it, thanks for pointing me there for the voltages yesterday dc42. Here was the readout I was concerned with:

    SD card 0 detected, interface speed: 11.7MBytes/sec
    SD card longest block write time: 582.4ms

    My friend told me that is should be interfacing at around 20 MB/s and writing under 50ms for most of the up-time.


  • administrators

    I'm glad you found the problem. Yes the SD card speed is normally 20Mbytes/sec on a Duet WiFi, or 21 on a Duet 0.8.5. Were you using the SD card that was supplied with your Duet?


Log in to reply