Printer needs to communicate more with MMU
I've noticed that the PINDA probe on the printer is sort of useless when using the MMU V2. I just had a jam with one of my filaments, 21 hours into a 36 hour print. Woke up to the printer waiting for me to fix the MMU before continuing, all I could think was "please be on a retraction and not a feed sequence!".....Nope it was on a feed sequence! So now I have a skipped layer. Just a little bit frustrated, the print was going to be for a good friend of mine at our Halloween party tomorrow, but now I don't have time to reprint it. I think the problem was due to filament inconsistency, out of the 21 hours of printing it was just the one mishap. So the filaments print with great tips, but occasionally one is just a little to big.
But I digress, what I really wanted to talk about is how the printer and MMU communicate. Hopefully the Prusa research team is working on this. With the two probes, the one on the MMU and the one on the extruder. The system should know if the jam occurred on a feed or retract cycle. On a feed cycle, the filament goes into the probe on the MMU, the printer should not move until it senses the filament pass through the PINDA probe, and detects movement. If it doesn't detect the filament, then it should time out (that's another thing that could be tweaked, why does it wait so long to time out and request an MMU fix?), then park the extruder head and ask for a MMU fix. Once fix is done, it should continue the layer that it was on when the filament got stuck during the feed. Same thing for a retraction sequence, printer knows when filament has left the extruder, so when it jams up and then waits for the MMU to be fixed, continue on to the next layer.
POOF! No more skipped layers.
I don't know anything about programming firmware, but I hope this can be done, and hopefully soon. 😀
Re: Printer needs to communicate more with MMU
I've noticed that the PINDA probe on the printer is sort of useless when using the MMU V2.
I think you mean the optical filament sensor, and it's more than just useless, it's actually disabled.
This has been the topic of considerable discussion here. I think the consensus is that:
1) The MMU would be a lot more reliable if the optical sensor was used to check for proper loading, and
2) The MK3's optical sensor was disabled with the MMU because it isn't reliable enough to be useful.
So the first step really needs to be to get the optical sensor reliable enough to actually be useful.
In My Humble Opinion, the optical sensor in the MK3 was a good concept but flawed in practice. They really need to use something more like a scroll wheel that can work well with any kind of filament.
Re: Printer needs to communicate more with MMU
In My Humble Opinion, the optical sensor in the MK3 was a good concept but flawed in practice. They really need to use something more like a scroll wheel that can work well with any kind of filament.
Better to use a microswitch connected to the Bondtech idler door. That way you would know when filament is actually in the main drive gears.
Peter
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…
Re: Printer needs to communicate more with MMU
In My Humble Opinion, the optical sensor in the MK3 was a good concept but flawed in practice. They really need to use something more like a scroll wheel that can work well with any kind of filament.
Better to use a microswitch connected to the Bondtech idler door. That way you would know when filament is actually in the main drive gears.
Peter
That would also work. I was thinking scroll wheel because it can sense motion and not just the presence of filament, giving you the ability to trap a wider range of failure modes.
Re: Printer needs to communicate more with MMU
Better to use a microswitch connected to the Bondtech idler door. That way you would know when filament is actually in the main drive gears.
Peter, you sir, are a genius. That is a nice way to do it. And I heard there's a free wire on the optical sensor - you could use it for this and be quite happy about it I think.
I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list.
Re: Printer needs to communicate more with MMU
Peter, you sir, are a genius. That is a nice way to do it. And I heard there's a free wire on the optical sensor - you could use it for this and be quite happy about it I think.
First off, I had this idea and only after I suggested it, I was informed that it had already been tried in an early prototype:
However, things have moved on and MMU2 developed further and I believe this to be a viable option.
With regard to the optical sensor, there are certainly positive, zero and return lines present. You would use two of those with a microswitch - return + positive OR zero, depending on whether the return line needs to be pulled high or low. It may be necessary to add (say) a 10k resistor to the third line (zero or positive) to hold the return at the correct (opposite) level to that required on the return.
Peter
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…
Re: Printer needs to communicate more with MMU
Since there's already power, ground available, you should only need one dedicated wire (signal) for a switch like that.
I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list.
Re: Printer needs to communicate more with MMU
Since there's already power, ground available, you should only need one dedicated wire (signal) for a switch like that.
Which is already present. Just need to work out which one...
Peter
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…
Re: Printer needs to communicate more with MMU
Since there's already power, ground available, you should only need one dedicated wire (signal) for a switch like that.
Which is already present. Just need to work out which one...
Peter
EDIT, no it's not, sorry 🙁 Only SCL and SDA, plus the 5V and 0V.
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…
Re: Printer needs to communicate more with MMU
Which is already present. Just need to work out which one...
EDIT, no it's not, sorry 🙁 Only SCL and SDA, plus the 5V and 0V.
I think there's a pin on the board's header? Someone mentioned making a switch and using one of the sensor's free pins and I believed them.
I guess there's already a firmware which watches a switch in that area - tracking it down would help. It sure would be awesome if Prusa would take a little bit of the reins and be communicative so the community of people who CAN test/build/code/solder could work out some of this with a little assurance it might get mainlined.
I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list.
Re: Printer needs to communicate more with MMU
It sure would be awesome if Prusa would take a little bit of the reins and be communicative so the community of people who CAN test/build/code/solder could work out some of this with a little assurance it might get mainlined.
It's totally fine if individuals go out, make changes and re-program the firmware. Unfortunately, that's fine for those individuals and the few brave users both knowledgeable and willing enough to take a chance. But for most users, it really does need PR to step in and fix the problem. And quickly. Not just for some but for all users.
So, keeping my fingers crossed that someone at PR will read these forums and understand just what they need to do.
Peter
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…
Re: Printer needs to communicate more with MMU
I'm with you there. I've put in hours getting videos and taking notes, in the hopes that the data provided would be utilized to move the development forward.
It's not been my experience that the team is interesting in fixing or even identifying issues. I wish they would, so we could interactively better format our assistance in ways which would be beneficial, but I mean a month of asking twice a day is a pretty clear message: Your help isn't wanted and your problems aren't interesting.
I still welcome the opportunity to move the development forward. At least the non-official stuff is enabling - I've been able to do honest print tuning and limit pushing using the communities' efforts.
I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list.
Re: Printer needs to communicate more with MMU
So, keeping my fingers crossed that someone at PR will read these forums and understand just what they need to do.
I hope so too Peter. I am so disappointed in the MMU2 so far. I have turned my fully functioning Mk3 into a machine that barely completes one out of five prints and then only with intervention. MMU2 is not the device I believed I was buying from watching the announcement videos. To say I am unimpressed with it is a massive understatement.
Re: Printer needs to communicate more with MMU
So, keeping my fingers crossed that someone at PR will read these forums and understand just what they need to do.
Peter
That was one of the reasons I started this thread, in the hopes that someone at PR would see it. I myself have no experience with tweeking firmware, so wouldn't know where to even begin to mod this myself. It just seems like the most logical thing to do, I mean, there are two filament sensors, you would think that the printer and MMU would use them in such a way to prevent skipped layers. Granted the optical sensor has it's problems with regards to detecting filament movement, but at least the printer would know that filament has arrived from the MMU. If the extruder sensor never gets triggered, then the printer can pause the print awaiting for user to fix the jammed filament. Once it's fixed, then the printer could continue that layer, instead of just printing empty air.