MK2.5S:MMU2S Configuration

[habitforming11]

Please consider a MK2.5S compatible variant of the lastest TZB MMU2s firmware. Unfortunately my technical ability in coding is nearly zero, but I will be happy to support through detailed testing.

[TheZeroBeast]

Ok, I have a plan of attack, I'll be hooking up my bench supply @ 12v this weekend to my MMU2S to tune Stallguard which is why it keeps trying to home.

I'll get a new build to you shortly after, should be one of the only things the voltage differences will affect.

[AbeFM]

On my mk3S/mmu2s if I say "home all" it consistently puts the selector on pos A, the drum on pos B, feeds out the wrong filament next to selector.

If you stop it, it'll move on to the next one (I believe the printer doesn't do any checks, so it just goes on), each time the drum is off by one so you get all your filaments sticking out of the machine and the selector moving over to snap them off.

I never liked how 'load all' worked, but it works worse now. :-P

Otherwise I've had few issues.

On Mon, Sep 30, 2019 at 11:52 PM Robert McKenzie notifications@github.com wrote:

Ok, I have a plan of attack, I'll be hooking up my bench supply @ 12v this weekend to my MMU2S to tune Stallguard which is why it keeps trying to home.

I'll get a new build to you shortly after, should be one of the only things the voltage differences will affect.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/TheZeroBeast/TZB-MMU2S-Firmware/issues/135?email_source=notifications&email_token=AAAKLEPCI27VRYMCMJKGL53QMLXTJA5CNFSM4IWLH4QKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEAAGAQQ#issuecomment-536895554, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAKLELTTZRVLG7IRRFOQP3QMLXTJANCNFSM4IWLH4QA .

[TheZeroBeast]

Can you open a bug report for that @AbeFM with the template so I can get it fixed?

Edit: Hey @AbeFM, I just loaded all five and double checked half a dozen time from fresh boot to after ejects and loads and always did all five from left to right as it should. Is it still happening on the latest release?

[TheZeroBeast]

@habitforming11,

I've had a win, I pipped 12v into my MMU2S and setup voltage reading to identify when it's on a 12v system and adjust itself.

I'm just running a test print then I'll post the compiled hex files.

[TheZeroBeast]

Hi @habitforming11,

Can you please try this set, from what I can tell it should be fully functional on your MK2.5S setup. If so I'll add it to the releases page 👌👍 TZB-300-MK25S-Bundle.zip

[habitforming11]

I should have some time to give it a whirl this evening.

[habitforming11]

Hi @TheZeroBeast

Happy to report the firmware has been successfully loaded, and the MMU and printer both appear happy so far. I haven't had a chance for a test print, but the MMU homes properly, the display on the printer is working, and the MMU properly shows as connected.

[habitforming11]

I just did a test print, have a strange issue. The MMU idler isn't moving off of the active filament during a print. I can hear the filament clicking against the bondtech in the mmu, and see the shaft turning from time to time as filament is being pulled through. This happened on both lane 1 & 2 during a test print just now. The load process selects lanes correctly, so I don't think it's a mechanical issue.

[TheZeroBeast]

Hi @habitforming11,

That's awesome progress!! 👏

I have a few questions.

1. How many full rotations past flush are the two idler tension bolts on the MMU2S?
2. Can you try grab the serial logs from the printer for a load sequence?
3. Does the load to nozzle in the menu work and allow you to unload afterwards?

[habitforming11]

for the serial logs will that work if i just do load from the printer screen, or does it need to be during a print?

once i get to the printer in about 30 min I'll answer all of the above.

[TheZeroBeast]

If the load_to_nozzle works from the menu it'll need to be from a print load thanks mate. When you get time 👍

[habitforming11]

1. Half a rotation past flush
2. I will post in a separate reply
3. I now cannot load as the printer no longer acknowledges the MMU. I was doing "load all" and must have hit the incorrect button to skip each empty lane. Now lane 5 red & green flash every time the MMU boot finishes, even after a reset. I press the middle button and the MMU returns no normal state, but as soon as I power off or reset MMU it comes on with lane 5 red/green flash again.

[TheZeroBeast]

Cool, might need to go further, half a rotation doesn't apply much force to the filaments to engage. Your MMU2S is disabled in the Service Position (allowing use of printer without MMU2S) This is a TZB FW feature. Use Left button to direct back onto a channel and reset printer to reconnect the two devices.

[habitforming11]

Wow, that's a very useful feature!

1. 1.5 turns past flush
2. to follow
3. Load to nozzle is select-able. The load is successful, however the MMU extruder motor gets dragged forward and backward during load/unload. Unload is also completely successful.

[habitforming11]

Here is the log for item 2

tzb_bn2462_BN358_log.txt

I have also taken two videos that I will upload now.

[habitforming11]

video of initial load, and of first filament swap

https://drive.google.com/file/d/1B92-2giY3KKgaqNLeZ_n7MUPwQ8ieYHt/view?usp=sharing

https://drive.google.com/file/d/1BDlPciolJ8HjPprmJjbjyLOnv1qH1oCx/view?usp=sharing

[TheZeroBeast]

So...... It appears to me like you need to do your Bowden calibration as the MMU2S is fighting to push more down a nonexistant tube. Check out my wiki post, the calculated values differ for my FW.

Hopfully

[habitforming11]

I can certainly do that again, you're correct that it's probably out of spec at least some. I had to replace that tube a week or two ago as the old one got chewed up from all the disassembly/reassembly.

However, I still see the MMU extruder motor getting dragged around during printing the actual part, not only during filament changes. Shouldn't the selector index over to the next lane, so that the in-use lane can pass freely through MMU?

[TheZeroBeast]

I’m not sure what would happen if the mmu2s tries to feed more and goes over the load. My guess is the load never acts on the final command from the printer due to not being at that stage and never disengages the filament.

Try doing the cal at least twice to ensure you’re filament is at the middle of the bondtech gears 👍

[habitforming11]

I had assumed the IR sensor trigger would tell the MMU that filament reached the print head anyway. I'll calibrate the length and see if the result is any different.

[habitforming11]

disregard

[TheZeroBeast]

Lane one is your one and only for cal 😊👍

[AbeFM]

It appears, to save time, to send filament most of the way at high speed, then it monitors the IR sensor.

              Thanks!
                      -Abe.

Sent from my "smart"phone, please excuse brevity and Swype-oes

On Wed, Oct 9, 2019, 5:56 AM Robert McKenzie notifications@github.com wrote:

Lane one is your one and only for cal 😊👍

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/TheZeroBeast/TZB-MMU2S-Firmware/issues/135?email_source=notifications&email_token=AAAKLEPGWHIDQFY57GGPVY3QNT62NA5CNFSM4IWLH4QKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEAVX2JQ#issuecomment-539720998, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAKLEJ7EPCAXOZNNU4WESTQNT62NANCNFSM4IWLH4QA .

[habitforming11]

From your wiki description, it sounded like it stayed at lane 4. Once I actually got to the printer & did it, the process was quite clear.

And with that, I believe the printer is now functioning properly. I guess the MMU extruder was getting dragged along because of existing ridges in the filament from the bondtech. Once it went through 2-3 tool changes, the funny noises stopped and the MMU extruder was no longer getting dragged around anymore.

Now I need to spend some time getting tips dialled in for my matterhackers white pla, and I might actually be in range of my first successful MMU print!

THANK YOU @TheZeroBeast

[TheZeroBeast]

WOOOO!!!!! Thanks great, good work. I'll release the MK2.5S code soon👍

[habitforming11]

What is the logic behind the "heatbreak jam/clog" message? I am getting that dragging effect on the MMU extruder again, and routinely getting heatbreak jam/clog notification. On inspection nothing is apparently amiss.

[TheZeroBeast]

Double check your IR_SENSOR is giving solid readings as if it's tripping on/off while loading it'll report a Jam.

Jam logic goes as follows within the load window (ie. filament has done major load and now both MMU2S and Printer are extruding so printer to grab)

1. Printer is turning extruder no filament in gears yet (IR_SENSOR must not give false positives)
2. Filament goes into gears which flags the IR_SENSOR
3. Filament is then extruded into the heatbreak a known amount but not to the hotzone
4. Filament is retracted a little less than the extruded distance in step 3
   • If there was a jam in the heatbreak then step 3 would slip and step 4 would release filament
5. Check if gears still have filament, is the IR_SENSOR still tripped. If not report error and await human intervention

I have never used the stock tower extruder design, the one I did up is very solid. In either case you have to adjust your extruder to get solid IR_SENSOR readings for this and Stock FW.

Also good to note that the above procedure is directly from Stock code, I haven't modified that in my FW.

[AbeFM]

https://photos.app.goo.gl/avo4wobHQvNUfKva8 this allowed me to fix my issue.

              Thanks!
                      -Abe.

Sent from my "smart"phone, please excuse brevity and Swype-oes

On Wed, Oct 9, 2019, 10:16 AM Robert McKenzie notifications@github.com wrote:

Double check your IR_SENSOR is giving solid readings as if it's tripping on/off while loading it'll report a Jam.

Jam logic goes as follows within the load window (ie. filament has done major load and now both MMU2S and Printer are extruding so printer to grab)

1. Printer is turning extruder no filament in gears yet (IR_SENSOR must not give false positives)
2. Filament goes into gears which flags the IR_SENSOR
3. Filament is then extruded into the heatbreak a known amount but not to the hotzone
4. Filament is retracted a little less than the extruded distance in step 3
   • If there was a jam in the heatbreak then step 3 would slip and step 4 would release filament
5. Check if gears still have filament, is the IR_SENSOR still tripped. If not report error and await human intervention

I have never used the stock tower extruder design, the one I did up is very solid. In either case you have to adjust your extruder to get solid IR_SENSOR readings for this and Stock FW.

Also good to note that the above procedure is directly from Stock code, I haven't modified that in my FW.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/TheZeroBeast/TZB-MMU2S-Firmware/issues/135?email_source=notifications&email_token=AAAKLEJPMQVOJ7EU52VWD4DQNU5JLA5CNFSM4IWLH4QKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEAWI5LI#issuecomment-539791021, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAKLENJPBQ4PY3H4XN6SQLQNU5JLANCNFSM4IWLH4QA .

[TheZeroBeast]

Nah, no need to overcomplicate it. Simple, simple

[habitforming11]

Ok, I had suspected the IR sensor was the issue, thank you for the confirmation. I wish there was a centralized location that explained all of these details. It sure would make troubleshooting a whole lot easier.

[habitforming11]

@AbeFM I see the issue in the video, but what fixed it?

[TheZeroBeast]

Hey @habitforming11, I'm working on doing up steps and videos for troubleshooting in my wiki.

The MMU2S handbook that is sent out with the device covers most basic troubleshooting but I find videos convey a lot more when troubleshooting.

TZB-FW also has a lot of handy troubleshooting features that really only videos can convey.

[AbeFM]

I just meant it's good to check if it's monotonic.

              Thanks!
                      -Abe.

Sent from my "smart"phone, please excuse brevity and Swype-oes

On Wed, Oct 9, 2019, 10:59 AM Robert McKenzie notifications@github.com wrote:

Nah, no need to overcomplicate it. Simple, simple

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/TheZeroBeast/TZB-MMU2S-Firmware/issues/135?email_source=notifications&email_token=AAAKLEJ3NPR7LD3T6ILUDJTQNVCHVA5CNFSM4IWLH4QKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEAWLHNQ#issuecomment-539800502, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAKLEMS4CHUBMTXISIA5ILQNVCHVANCNFSM4IWLH4QA .

[AbeFM]

Sorry for the late reply! Er, I printed a couple different flags, and ended up putting foil tape on the tip. The shape of the tip is important, but I didn't get anything definitive enough to publish.

On Tue, Oct 8, 2019 at 9:29 PM habitforming11 notifications@github.com wrote:

@AbeFM https://github.com/AbeFM I see the issue in the video, but what fixed it?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/TheZeroBeast/TZB-MMU2S-Firmware/issues/135?email_source=notifications&email_token=AAAKLEI4RORKVGU3AS6ZYE3QNVM3VA5CNFSM4IWLH4QKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEAWQWMY#issuecomment-539822899, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAKLEK22SGYEBQALND6AZDQNVM3VANCNFSM4IWLH4QA .