MakerBot Industries

Trying to revive this thread as I am having the same problem with tall prints with flat sides where the sides end up uneven and not quite flat. Things I'm going to try:

• tightening the belts
• active cooling
• Lower print temp

Im currently printing with PLA.

I think I found it.

I looked in the gcode (every time I have a problem, it seems like the root can be found there!) and it appears that my HBP temp was only 100C. When I was using the MK6+, it was 120C. I just manually changed it and the part stuck like glue!

Also, note to anyone else out there, using the 'chamber' settings to control bed temperature doesn't seem to be effective :(. That is what I was modifying. I just wasn't going back to the gcode to verify that it was actually changing! I think I have it now, though.

Thanks, all, for the suggestions!

J

I have tried reducing this z height, but the first layer or two end up smashed. This isn't a huge problem, but I don't think that the first layer should need to be smashed. The problem still exists as well, it just sticks better when I do this. Thanks for the suggestion.

J

Other than the mentioned sticking problem (which I don't think is related to the extruder), everything is great! Assembly is extremely fast. It will probably take you longer to put it on the 'bot than it will to assemble it, and most of that time is just switching out wires and ensuring that everything is in working order under the hood. I'm loving the fast warm-up time. That saves me from having to pre-heat nearly as much.

J

I replace the kapton on my heated bed about every 10 prints. I am careful to never touch the build surface itself during this process. I also only remove parts using tweezers on the vast majority of parts in order to avoid touching the build surface. This is very easy on small parts, but more difficult on large parts. As the parts that I'm building get larger, I will usually just replace the surface after every build just in case. It really sucks to not spend the 5 min to replace the surface and then have things mess up 2 hours into a build.

I know you didn't necessarily mean 'grubby' personally :). I agree, fingerprints = bad.

My bed is leveled, but I have this sneaky suspicion that the surface of the bed is cooler than it once was. I was originally printing well on the surface of the hbp pcb, but the parts end up having the ripple from the underlying traces. Once I received the aluminum, I used the thermal paste provided, but it didn't seem like much. I have been considering moving the platform to a higher temperature, but I also don't want to damage the pcb. The bed may not be up to temperature, but I am confident that the underlying pcb is. Any suggestions on keeping the heat in?

Thank you for your time,

J

Hello all !

I have just finished my new Makerbot. Very happy with what the community has done for all of us.

I would like to cast some ideas for the future :
Long time ago, there were only black and white printers. This is where we are here with the makerbot. Today, everyone has a (paper) colour printer, so why not going this direction and think about making real "in depth" coloured objects with our machines.

I think this is possible, let me tell you how this could work:
I talk here about mixing 3 ABS rolls together, typically one Blue, one Red, one Green, and pushing them into the same extrusion head.
In the case of paper or display units, those 3 colours are mixed, this is how it works. In the case of plastic extrusion, I think we need to add another colour, that would be a neutral, lets say white.
So the idea is to push those 4rolls in the same extrusion head, each roll being pushed individually by one stepper motor. We need the software to balance those colours, driving the motors to individual speeds to make the sum of them 100%.

That is the idea. Of course, this is a big mod. It will involve the design of a new extrusion head, much bigger with 3 or 4 motors pushing together. Much heavier, but there are solutions here also. One solution is to be inspired from how the ultimaker works, which means the extruder motors are static, therefore the extrusion head is much lighter.

Think about it, geeks, and you can start a new design. We need this head with a special shape, so each roll movement has no consequence in the others roll movement, the mixing of colours should be as close as possible to the extrusion head.

We can keep a single heater, should work the same, same amount of heat has to be generated at the end… You get it, so we can control the colour of the "pasta" going out of the extruder. We need to design this head with specific angles, so the ABS has no tendancy to flow back.

Software also has to be implemented. It means we need an addition in the Gcode, something like a new "C: R,G,B" where values of R,G,B are a percentage of colours, the sum would be 100(%).

For the electronics, our current cards have one extra axis not used yet. Today, we use only A for the extruder. There is a B axis, for a second extrusion motor. Not enough but an extension card would have to include A,B,C axis (trichrome) in the case of 3 rolls, or even the 4th roll (D), which seems like the good solution for me.

Yeah, easy to say, harder to do…
I have not all the time necessary to try this, althought I will try sometime to make some mixing and see what happens.

The mechanical part seems feasible, I would really like to know how the ABS mix together. Even if mixing is not perfect, from a certain distance, the objects should look OK.
Software, Well. Seems doable also, but need quite a lot of time.

I am sure someone will build that. Some big investment, but the reward is even much bigger. Think about a whole new kind of objects we could make, real arts. Nothing like any plastic today which are all "monochromatic".

Voila, thats about it. Enough to make some of you dream about it… And let me know if someone goes this direction…Its a "free" patent.

Cheers
Franc

How do you like the MK7 otherwise? I have been doing a wait and see on it.

Why is the relief valve closed at the end of the print? When both the relief valve and pressure valve are closed, the frostruder continues to frostrude. Seems like by just getting rid of the last line of code it doesn't have that problem

Also, let me add when I did the firmware testing, I just used a CD rom audio cable and a jumper wire to take the "sense" pin both high (+5V) and low (GND) just to make sure it wasn't the motherboard preference issue of inverted endstops. The bottom line, nothing happens when you "trip" the estop to the motherboard.

I'm kind of in shock we went this long and nobody noticed. Only with MK7 being a hack and thus people getting a lot of false trips, did this even come to light.

So, some testing this morning showed that firmware 2.7, 2.8, and 3.0 did not sense the e-stop.

Either you don't have the estop board soldered correctly, your wiring is wrong, or the firmware version doesn't support the estop correctly.

So, apparently it is that last option I listed, the firmware is what's broken.

I'm pretty sure this worked when MK6+ first came out (hence the safety board) but don't remember how that lined up with firmware versions timewise. I'm thinking 2.4 and 2.5 might have proper e-stop support.

I don't think any of us users test out the safety feature after flashing firmware (just something I took for granted). The idea behind the board was a failsafe that didn't depend on the firmware, and honestly, it won't hurt a thing if it does just sit there and skip steps.

Feel free to post a bug at MakerBot. This was supposed to be a feature, and honestly, I don't have a good way to go back several firmware revisions and see how they implemented it. I searched all of the current code and only found where the pin was assigned, but not where any action takes place based on the pin status (thus why it doesn't seem to do anything).

I never have problems on my MK6+ systems, but they are a far different design in how and where the safety is mounted. The MK7 safety is not properly designed and is really a hack to take what worked on MK6 and stick it on MK7. What we need is a proper sensor device (not one rated at 140C) that really is rated to trip at the correct temperature. It's just a matter of a part change-not even so much a system change.

And well, the firmware-good luck with that. Maybe if enough people complain. From informal testing, it goes back to 2.7 and now I would have to dig to prove 2.4 or 2.5 worked so being that it's gone this long and nobody even said a word, they may not implement a bot shut down.

In summation of your questions
1. You fixed the sensor by adding another layer of insulation
2. It's the firmware and right now, there is no easy fix
3. The 10 minute shutdown may be a sequence issue (or you could have a faulty supply). There is nothing but a dead short jumper on the motherboard that turns on the supply (no active electronics) so again, it's either backfeed from USB, or the power supply itself.

Also.. the instructions clearly say,

If you’ve wired everything up correctly, the red FAULT LED will not be lit. If anything goes wrong, you’ll see that light go on and your heater will be stopped automatically.

It doesn't say the whole bot shuts down, just the heater, which is exactly what mine does. So, once again, can you please confirm that the entire bot should shut down in the event of a heater fault? Maybe mine is working just fine, and this is just a weird design flaw IMO. I think extruding nothing and having your filament stepper grinding on filament potentially for hours is sort of a bad design, which is why I think mine is set up wrong somehow.

As far as the heater shutting down, that definitely happens. Just the bot keeps on truckin' as if nothing ever happened otherwise. Maybe this is how it is designed. Yours shuts completely off?

ha.. you are correct, I was going off of memory at work. Fail.

It is a 4 pin on the motherboard and it is in the e-stop plug correctly. I was referring to the other plug/setup from memory. My mistake.

I checked and my system is plugged in correctly. I added the ceramic tape and tried overheating it at 250 degrees and it wouldn't trip, so i think that problem is solved, and when I disconnect one of the leads on the sensor at the cutoff board, it does click and go red, but I haven't tried this during a print yet to see if the whole bot shuts down. I'll have to double check my firmware and see. It's weird that it only overheated during a 100% fill print.

And it's interesting that you mention the pauses.. because my bot connected via USB to a 6 core 24gigs of ram beast of a machine pauses frequently during prints. Maybe 1-2 times, maybe 4, maybe none. Seems like it doesn't like me using the machine while printing, which I can understand, and I'll accept the pauses.

I could use the SD card, but that just isn't practical when I hit cancel and re-slice often.

Thanks for the help again, I'll keep checking those paths you put me down.

You also might want to check your initial layer height (and your Z home position). If its a bit too high things will tend to want to stick less

Anyone else having trouble with the new RepG adding an extra toolhead during the g-code generation? It seems to work as intended on my windows version (only toolhead 0 gets created), but I just tried it on a mac and for some reason it keeps making a second toolhead (toolhead 1) and then complaining about it when I try to execute it…

???? It should be a 4 pin connector period-unless the motherboard was changed and I didn't know about it.

I think you are confusing the heated bed thermistor going to the extruder controller.

The e-stop plug is on the motherboard properly. The fourth pin hangs over the end like instructed, and no, it's not plugged into the wrong thing that the instructions clearly say not to do.

Um-no that's clearly not right
http://www.makerbot.com/docs/safety-cutoff-kit-1-0-rev-d/

The Estop is motherboard to safety board via a 4 pin locking CD-rom cable. No pin shifting!

I know it's confusing but there is a reason the whole bot is configured that way. The motherboard runs the bot, the extruder board is designed to offload the processing tasks of the feedback loop for temperature from the motherboard. The estop has got to be on the motherboard not the extruder controller, because the whole idea was a safety check in case something got "stuck" on the extruder controller.

The 2 board system is key in how gen4 outperforms other single board solutions in overall motion control. For example, turning on temp reporting on a single board solution is known to cause pauses in the printed part which results in little nubs all over the print.

Hey thanks for the reply. The safety cutoff board soldering was just about the easiest thing I've done, so I'm doubting that, but willing to check. The fact that the LED lights up correctly and the hot end shuts down tells me something is set up right. The test of removing one of the leads and hearing the click and seeing the red LED works just fine too.

I will also add another strip of ceramic tape, eventhough it's built exactly how the guide says to, and even looks almost identical. So I don't think anything was done "wrong" per se.. just maybe their instructions are a bit on the safe side.

I have the option in the motherboard prefs set to safety cutoff kit. Is this wrong?

I am running a mk7 RepG 0027 alpha and the corresponding firmware. Thought that was rather current.

The e-stop plug is on the motherboard properly. The fourth pin hangs over the end like instructed, and no, it's not plugged into the wrong thing that the instructions clearly say not to do.

My gen4 interface kit shows the temps properly. As does RepG.

So to clarify.. I should see the entire bot shut down when the temp sensor is tripped, correct? Not just the extruder head.

I'll double check everything on the underside of the bot later. Possibly the connector got shimmied out of the way. That's possible considering it's a 4 pin being slid over jumper 3 pin connectors.. Not the best design.

Step 1, clean the heated bed (only when cool) with pure acetone (from a hardware store) and a paper towel to remove your fingerprints which are what is keeping the ABS from sticking.

Common mistake by people when they get their grubby hands all over the print surface while assembling. You will be amazed at what a difference it makes. (By grubby, I mean we all leave invisible fingerprints and don't even think about it because they a nearly invisible).

Also ensure you bed has been "leveled" in reference to the nozzle in the 4 corners. This is very sensitive at well.

As a rule, I try to clean the surface every dozen prints or so just to "make sure".

"Are your doing something wrong?"

Yes, you don't have enough insulation between the temp switch and the core.
Fix this first by putting another piece of the ceramic tape between the switch and the core. DON'T insulate the switch (another layer on top). This way, the switch has a chance to disipate any extra heat to the air, and only sense when the core truely is too hot.

Either you don't have the estop board soldered correctly, your wiring is wrong, or the firmware version doesn't support the estop correctly. Reverse input polarity to the safety relay (from the Extruder controller), or other issues could easily cause the motherboard not to be able to see the relay trip. Even setting the wrong endstop preference in the motherboard options could affect this.

The not powering on for 10 minutes is a function of the power supply-not anything with the bot. USB is known to backfeed the supply, thus the supply is in fault mode until the capacitors discharge, at which point you can turn it back on.

It a bad design, but your shutdown procedure is causing the issue. ALWAYS, ALWAYS unplug USB, then power down the supply. Reverse order is important too. POWER on the supply first, then plug in USB. Never do it in any other order. Failure to turn on often happens when the power supply is turned off before the USB is unplugged. This can cause the power supply to not turn on for up to 10 minutes (as you are experiencing) because USB charged up capacitors in the supply, so it thinks it's in a fault state. The capacitors have a bleed resistor in the circuit to slowly discharge them, but it takes time.

So in summary, your build of the extruder is one problem, the safety board might be another, and your procedure is likely the third problem.

For those without a front bearing on their MK6 or MK6+, this indicator

http://www.thingiverse.com/thing:10116

is actually a pretty nifty way to see that the extruder gear is doing its job. It also helps differentiate between skip steps and reversal (they sound very similar).

I came back to the thread because I was having more problems. The set screw the holds the feed gear on the stepper had come loose. I cleaned my nozzle 3 times before I thought to check the extruder feed. It was working some, but longer prints were pooping out. I think when the motor got warm, the gear slipped until finally it worked all the way loose. I tightened it and added some glue to the post to help it stay put. I was able to do a 2hr print no problem.