Imagine something interesting here...
FYI for those like myself who don’t have every G and M code memorized:
https://docs.duet3d.com/User_manual/Reference/Gcodes/M670
M670: Set IO port bit mapping
Parameters
Pnn:nn:nn... - List of logical port numbers that bits 0, 1, 2... control (supported in RRF2.x and earlier)
Cnnn - Used to specify the pin name(s) to be controlled (supported in RRF3 and later, see Pin Names)
Tnnn - Port switching time advance in milliseconds
So I've been toying around with the idea for a bit of creating a Virtual Reality / Augmented Reality interface for a printer. This would essentially be a more interactive, 3D version of the web interface already available. There may also be the ability to backport it to 2D devices as well similar to Microsoft's Mixed Reality viewer. Here's some possible use cases:
Immersive Printer Status:
Augmented Reality Printer Status:
Factory View
So thoughts?
Useful? Not Useful? Never Use it? I'd like some feedback and opinions if it's something worthwhile to develop. With the VR devices coming down in price ($299 for Oculus Quest 2), this might be an interesting interface in the future.
Wanted to put in a request for a few updates to the next rev of the tool boards:
I realize some of these changes might increase the price a bit... I for one would gladly pay a bit more to not have to make a custom connection (which ends up being a couple of hours of my time after I get it wrong a few times...).
Does anyone know if there is a way to limit the ramp rate of heaters? I’ve done a bit of digging, but haven’t been able to find anything.
Thermal stress in parts (hot end, bed, chambers, etc...) is proportional to the temperature gradient from heating. With higher temperatures, more power is needed to combat radiative losses (proportional to temperature^4) and convection losses (proportional to temperature^3/2 with free convection). Higher powered heaters over-perform at lower temperatures where radiative losses are minimal. The resulting ramp rates can cause high thermal stresses and potential damage to printer components.
From a safety standpoint, implementing a ramp rate parameter would also be a good idea. Starting out with a lower PWM (with a lower max-temperature capability) and verify heater performance before reaching high-powered states. In the event of a problem (admittedly likely only a software problem), the heater would never exceed safe ranges.
I think the easiest method of implementation is to have an array of calibrations at different PWMs / temperatures, and then use interpolation to find the appropriate PWM to achieve a desired ramp rate. I could go into more of the algorithm details if this put into the firmware at some point in time. This might be achievable more manually with conditional g-code, but I haven’t looked too much into it.
Anyone have any thoughts or ideas?
Just another item on my wishlist - it’d be great to incorporate a duet camera (or just software) into the ecosystem (CAN-bus ir POE?). Easier camera integration into the UI - and the long shot feature for me, but AI print monitoring. Should be feasible to detect common issues such as print pulling off the bed. Maybe provide a cloud service that would analyze the images every minute or so to look for errors?
Anyway, just a few ideas.
So after battling an issue for almost a year, I finally resolved it, - I really wish I’d know this years ago, so posting it here.
Ground your 24V / 48V power supply “0V” line to your ground. I had not done this until tonight, always thinking that I was avoiding a possible ground loop. Turns out, some sensor readings are impacted by this, particularly if you shield cables. I’ve never had the best of luck with Thermocouples, they’ve always been on the fritz. Turn out, this was the likely cause.
I recently observed some very odd behavior on my Thermocouple lines as documented here: https://forum.duet3d.com/topic/31601/mystery-help-4-tc-lines-all-error-at-the-same-time/9?_=1705372886695
At the time, I had a custom PCB with 240V passing through with SSRs - I attributed the noise to that, and ultimately redesigned my electronics. I have no regrets spending about a year doing that (well, mostly), but I was shocked to encounter the same issue with a new board, wiring, and most components.
After adding filters and ferrite clamps, the noise got cleaner but was still quite visible. I saw a good 0.8V peak to peak at 60 hz, not only on my thermocouple lines, but also the SPI lines and 24V input. Somewhere either the noise or the ground shield over the lines was interacting with the signal and making everything go haywire.
It’s worth to note that almost all other aspects of the Duet 3 boards worked as intended with no visible problems.
So…. In the event you have sensor issues, try grounding your 0V line on your power supply. There’s a good number of Reddit posts you can find that explain the merits and downsides of doing so. Personally with the Duet 3, I wish I’d done this ages ago.
In case anyone else encounters this issue in the future, I thought I'd add a recent finding:
I've been having this issue frequently near a BMG extruder outlet. I'd cut the tube a couple of times to give it fresh material to clamp onto, which did help a bit, but the issue was slowly getting worse.
I had a couple of the E3D Bowden fittings spares laying around and thought I'd try replacing the inner fitting. I cut 1/4 of it off to get it out, but was actually fairly surprised when I got it out.
The original tube (also from Anycubic) had left some residue on the Bowden clip - you can see it in white. This was essentially coating the clip in PTFE as well, which substantially reduces the friction holding the tube in place (PTFE on PTFE has the nearly the lowest friction coefficient possible). I switched the tube to some leftover Capricorn tubing and changed out the fitting - I've noticed the difference immediately despite the smaller outer diameter of the Capricorn.
TLDR?
Public Service Announcement: Throw away your Anycubic PTFE tubing and replace the Bowden insert if you're having this problem.
I was looking at the maximum power on page two for current estimates.
In any case, I just made a duplicate board and connected it to my test printer with a 24 AWG, 3ft ethernet cable and it worked just fine. Literally just plugged it in and it worked. I have a 7ft cable for my main printer coming in on Friday - I expect it to solve the problem.
All ethernet cables are NOT created equally...
FYI, I've done a lot with CNC machines and rigidity is your biggest problem with the setup above. If the head mounts on that frame, I'd bet you could barely do wood reliably. There's a lot of CAD/CAM software out there that can take a 4th axis to do 4-axis milling, although it's generally more expensive as 3-axis linear milling is the bread and butter unless you're doing something really complicated. If you're looking for something to do with the rotational axis, I'd highly suggest you put that on top of a 2-3 axis linear system instead of substituting it. If wood is OK, you can find small Chinese routers at about the price of a single Duet card.
If you are interested in CNC milling, I'd recommend you strip the electronics and put it on something like a Grizzly G0704 (~$1k) or a used Bridgeport (~$1-5k). Grizzly should be able to do aluminum or mild steel. Make sure you have a Bridgeport or similar if you ever plan to do any materials that are harder to machine (stainless steel / titanium). Downside to a Bridgeport is that they typically weigh a ton (literally, >5k lbs) and operate on 3-phase power. They can't be beat on rigidity and setup tolerances though.
Also, CNC milling is a profession - if you're interested in it, I highly suggest you take a college course on machine shop basics before diving in. It can be highly dangerous if you don't know what you're doing and there's a reason many professional machinists are missing fingers. Steppers of that size can crush fingers easily not to mention other dangers while actually cutting.
Hello,
Can you please update the Duex documentation to include v0.9?
I recently bought the new version and rather embarrassingly couldn't find the 12_EN jumper (I had the card mounted in such a way that it was hidden by the 12V header). It was only days later that I actually realized where it was...
Also, I assume the 12V line is a 12V bus essentially? That's quite useful in my setup - it would be nice if you took credit for that in the features and documentation.
@moth4017
So I tried on the bed heater OUT0 line - and I got the same response with buzzing!
As I was poking around thinking the buzzing was a weird wiring short, I was able to change the pitch when I touched the power inlet ferrite inductor beads. I removed them and the problem was solved.
But WHY????
Ferrite inductors should smooth the voltage at the power inlet, although also probably changing the impedance. Anyone have an idea why that would cause buzzing from the board (mosfets?)?
@droftarts Exactly why I had it connected to one of the heater output lines. 
@moth4017 i’ll give that output a shot next. I’ll
Be down an output for another fan but it shouldn’t matter in the short term.
@moth4017 it’s a somewhat larger (200mm?) fan with high airflow. At 20% or so it does a burst sequence (on 0.5 seconds, off 5 seconds) about three times before arriving at a constant speed.
In the fan side everything seems to work perfectly fine. The issue is the buzzing that appears to be coming from the motherboard.
@moth4017
Yep that was my first thought too. Changing it changes the pitch of the noise but not the intensity / loudness.
I've encountered an odd issue - my newest Duet board has a loud buzzing when connected to a fan. The buzzing is about 2X the loudness of the fan itself at ~20% PWM.
Anyone know if this is an indicator of a faulty board? Other than the buzzing, it seems to be working just fine.
So after battling an issue for almost a year, I finally resolved it, - I really wish I’d know this years ago, so posting it here.
Ground your 24V / 48V power supply “0V” line to your ground. I had not done this until tonight, always thinking that I was avoiding a possible ground loop. Turns out, some sensor readings are impacted by this, particularly if you shield cables. I’ve never had the best of luck with Thermocouples, they’ve always been on the fritz. Turn out, this was the likely cause.
I recently observed some very odd behavior on my Thermocouple lines as documented here: https://forum.duet3d.com/topic/31601/mystery-help-4-tc-lines-all-error-at-the-same-time/9?_=1705372886695
At the time, I had a custom PCB with 240V passing through with SSRs - I attributed the noise to that, and ultimately redesigned my electronics. I have no regrets spending about a year doing that (well, mostly), but I was shocked to encounter the same issue with a new board, wiring, and most components.
After adding filters and ferrite clamps, the noise got cleaner but was still quite visible. I saw a good 0.8V peak to peak at 60 hz, not only on my thermocouple lines, but also the SPI lines and 24V input. Somewhere either the noise or the ground shield over the lines was interacting with the signal and making everything go haywire.
It’s worth to note that almost all other aspects of the Duet 3 boards worked as intended with no visible problems.
So…. In the event you have sensor issues, try grounding your 0V line on your power supply. There’s a good number of Reddit posts you can find that explain the merits and downsides of doing so. Personally with the Duet 3, I wish I’d done this ages ago.
FYI for those like myself who don’t have every G and M code memorized:
https://docs.duet3d.com/User_manual/Reference/Gcodes/M670
M670: Set IO port bit mapping
Parameters
Pnn:nn:nn... - List of logical port numbers that bits 0, 1, 2... control (supported in RRF2.x and earlier)
Cnnn - Used to specify the pin name(s) to be controlled (supported in RRF3 and later, see Pin Names)
Tnnn - Port switching time advance in milliseconds
@T3P3Tony
Any update on the availability of the motors? I haven’t been able to find them at distributors yet.