1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

MAX Bed Temp for the R2?

Discussion in 'General Questions' started by SteveDjarrell, Nov 30, 2017.

  1. SteveDjarrell

    SteveDjarrell Active Member

    Joined:
    Jan 9, 2015
    Messages:
    101
    Likes Received:
    40
    Guys, what is the highest setting you think I can safely use with the Heated bed on the R2? I am looking at trying to print some Delrin and I would like to get up to 130C but I am not sure if the stock bed is capable of that temp. I know the setting in the firmware limits it but just not sure what the safe max would be if I wanted change it. Also just heating up to 85C takes forever so I can only image how long it would take.

    Also, I know Delrin degrades when heated above 230C and I will have it fully enclosed and vented outside.
     
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Not going to happen. I do not know what the max they designed for is, but that stock heater is not going to get you to 130c. You can't even do that on the R1+ :)

    (note: you could mod the R1+ to do that, but stock -- no)
     
  3. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    You won't get the bed over 110°C because the firmware limits the bed temp to that temperature. That is if you can actually get it that high in the first place. The PEI sheet should actually help you run at a lower bed temp than specified and PEI will bond to the nylon if you don't actually use a release agent on the bed. @Geof has a lot of experience with Nylon and I think he uses glue stick (PVA) so nylon doesn't stick too well.

    @mark tomlinson beat me to it.
     
    mark tomlinson likes this.
  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Great to know :) I was not sure how high the R2 would actually get... I knew 130 was not going to happen...
     
  5. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    Why are you after Delrin? Link to the OEM where you bought it? I use alot of Nylons. Lower bed heat, higher extruder heat a bit on the slow side with no issues. Glue stick like @WheresWaldo mentioned on the bed.
     
  6. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Also don't forget that hysteresis will actually not allow you to get to 110°C, because it will fail if it moves above 110° for any even small period of time. At least anecdotally I have observed that the bed or hotend for that matter can only reliably reach a temperature a degree or two lower than the firmware maximum.
     
  7. Kilrah

    Kilrah Well-Known Member

    Joined:
    Apr 18, 2017
    Messages:
    498
    Likes Received:
    332
    110 is the firmware max above which it cuts, but the max you can actually set is 100. And you can reach that easily with the firmware set to allow full 255 PWM for the bed heater (although it's long).
     
  8. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    @Kilrah That is anecdotal information and may not represent actual performance. The fix of changing the PWM duty cycle should not affect maximum temperature as you think it does, what it would affect is the speed at which you can attain maximum temperature.
     
  9. RSC

    RSC New Member

    Joined:
    Nov 25, 2017
    Messages:
    20
    Likes Received:
    6
    My new R2 was quite happy to accept a 110ºC build plate temp. And it got it there relatively quickly. But every time I tried to print at this temp I got the "communication error" failure within a few layers of starting the print. Dropped to 100ºC and the printer runs error free. Having the FW limit the max without accounting for the inherent hysteresis that's going to occur is IMO a significant oversight. The printer should limit the max it lets you set to one that includes some "deadband".
     
  10. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Talk to the Marlin devs. You're just preaching to the choir.
     
  11. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
  12. Kilrah

    Kilrah Well-Known Member

    Joined:
    Apr 18, 2017
    Messages:
    498
    Likes Received:
    332
    That's incorrect, and is not even what I was talking about anyway.
    I was saying that (at least when I checked which was in June when I received my R2, could have changed with updates) the LCD interface did not let you enter temps >100°C, and when specifying a bed temp >100°C in Gcode it was truncated to 100°C.
     
  13. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    So I was responding to this message directly above mine and particularly this statement which I know you know is not relevant to whether or not the bed can reach 100°C.
    We already have a lot of reports that the bed cannot reach 100°C even after the PWM firmware change. PWM is duty cycle not the actual power that is delivered to the bed. So the old default was full power for 250 of the 500 cycles per second that the hardware PWM is set to, it is not half power. Changing it to 255 just changed it to full power for 500 out of 500 cycles, result is faster heating not more heating.
     
  14. RSC

    RSC New Member

    Joined:
    Nov 25, 2017
    Messages:
    20
    Likes Received:
    6
    Not looking to have an argument, but.... On full is on full and the power dissipated when it's on full is indeed the same no matter how long it's on for (ignoring things like inductance). But the bed heating system can't respond instantaneously to input power so how long the power is on (duty cycle) actually matters to thermal behavior in a very fundamental way. The bed will heat at a rate and to a maximum dictated by the average electrical input power (as set by the PWM duty cycle) and the rate the build plate is dissipating heat into the ambient environment.

    By way of a metaphor, light a candle and then hold your hand over the flame for .001 seconds (low duty cycle), say, once every 5 seconds. Now hold your hand over the flame for 4.999 seconds (high duty cycle) once every 5 seconds. The flame is just as hot in each case but what your hand feels (= build plate) is the average. How hot your hand gets and how quickly it gets there is a function of how much time it spends over the flame (how long the PWM is "on") and the temperature when it's not over the flame. Duty cycle.

    The difference between having your hand in the flame 2.5 seconds of every 5 seconds vs. putting it in the flame and keeping it there is the difference between 250/500 PWM (50% duty cycle) and 100% on. I think you'll agree this is the difference between 1st and 3rd degree burns.

    So if changing the PWM from a 50% duty cycle to full-on doesn't make a difference, there is another issue that is limiting the system's performance. Probably something is current-limiting, so when power goes on full the voltage being delivered drops. You might see this if the power supply was a little underrated and relied too heavily on output capacitance (and it didn't have enough).

    As I mentioned earlier in this thread, my R2 bed quite happily goes to 110ºC. And I'm pretty sure it it let me input 110 from the touch screen. And the build plate gets there fairly quick (maybe 7 minutes. Much faster than my Replicator could). My only problem is that it doesn't stay within the too-narrow deadband and the printer faults for high temp eventually. So for my machine anyway, there doesn't appear to be any inherent limitation in reaching 110ºC. And given the rate at which my build plate heats (fairly quick), I'd guess if I modified the FW to let it, it has the operating margin to get to something comfortably above 110ºC...
     
  15. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Just was being technically clear. Your metaphor does not apply the same way because humans are different than machines. We experience the world differently than mechanical and electrical systems.

    The code that runs RoboLCD will take any input of Bed temp over 100°C and return a value of 100°C.. If you don't want to believe me then look at the code in manualcontrol.py and look for yourself. I'll even direct you to the correct line (87-90). So even if you modified the firmware you would not be able to use the LCD to heat the bed over 100°C. Now you can use the web interface or G-code to get the bed to the maximum that the firmware allows, within the limits of the power deliverable through the Robo Mainboard..

    Just out of curiosity does your bed attach to the wiring with pogo pins?

    Anyone that starts a post with "Not looking to have an argument but ..." is doing exactly the opposite, You obviously do want to argue so why not just be truthful about it. Since you do apparently want to argue, then send me a PM and we can argue to your hearts content..
     
  16. RSC

    RSC New Member

    Joined:
    Nov 25, 2017
    Messages:
    20
    Likes Received:
    6
    Well, I do love to argue, but no, I'm not trying to argue. However, if I don't agree with what you've said, there's no way for me to articulate my disagreement without being contrary. Which is why I extended you the courtesy of a pre-apology before explaining why I disagree.

    As for my metaphor, well, a metaphor is not an exact model. It's not meant to be taken literally. That being said, heat is heat. True, different materials absorb and conduct and radiate heat differently, but they're all governed by the same rules of thermodynamics. If you don't like the "hand" metaphor, then replace the hand with a 3D printer build plate. Everything else in my metaphor stays the same.

    The plate heats up a little each time it's exposed to the flame, it cools down a little when the flame is removed (=PWM ON and PWM OFF). It gets hotter only if the heat lost when the plate is removed from the flame (PWM OFF) is less than the heat gained when the plate is in contact with the flame (PWM ON).

    How quickly it gets hotter is a function of how small a delta there is. If it heats up 5º and cools down 4º each PWM cycle it's going to take a lot longer to heat up than if it only cools down 1º each cycle.

    How hot the plate ultimately gets is a function of the rate heat is being input (which is dictated by the duty cycle per above) vs. the rate at which it's being dissipated (a function of the environment). The plate will heat until equilibrium is reached. Assuming the ambient temperature stays the same, equilibrium (the plate's steady-state operating temperature) will be different depending on how much average energy is being put in to the plate, how long the plate is exposed to the flame vs. not exposed. Less heat energy in per unit time means equilibrium is reached at a lower temperature. More energy in per unity time means a higher temperature. Apply the heat source 100% of the time, the plate heats up until the rate it is dissipating heat equals the rate heat is being put in from the flame at which point it gets no hotter.

    Replace the candle flame with the resistive heating element of the build plate because it doesn't matter how the heat is generated. Watts are Watts. So now instead of getting Watts from exposure to the candle flame we're switching the resistive heating element off and on. The behavior is the same. PWM duty cycle dictates how quickly the plate heats and how hot it gets.

    PWM motor speed control is another analogy (but not a metaphor because it's directly analogous), although you can also apply this to a lamp dimmer. The motor's speed can't change instantaneously. So when the PWM pulse turns on the motor starts to speed up. When the PWM turns off the motor starts to slow down (analogous to heating up and cooling down with the build plate PWM). The speed the motor runs at becomes a function of the ratio of on vs. off, AKA duty cycle. Feed an electric motor a 50% duty cycle voltage and it sees an average voltage of 1/2 of the full voltage and it runs at 1/2 the speed it would run at with 100% duty cycle.

    The motor is doing Work when it turns, and Work can be expressed as Watts. And this is exactly the same Watts as the Watts a heated build plate dissipates when it's being fed a PWM signal... so the rules that apply to PWM control of a motor's power dissipation apply equally to a resistive element heated build plate. Motor power dissipation is a function of the PWM duty cycle powering the motor, same as for a build plate heater.

    If you still don't accept what I'm saying, do some searches for "PWM control of Resistive Heating Elements" and see what others have to say on the topic.

    My R2 was ordered last week. I am guessing it is the latest rev HW. The build plate is wired, it does not use the pogo pins. If what I'm saying about my printer is at odds with a printer that's got pogos, I could easily imagine ways that the pogos might explain the differences. I used to do ATE bed-of-nails test fixtures (100s of pogos) and pogos just aren't all that reliable. Running higher current through a pogo was also a no-no. They tend to make point contact with whatever they're touching, and this can produce high contact impedances which limit current and cause undesirable heating at the contact point.
     
    mark tomlinson likes this.
  17. RSC

    RSC New Member

    Joined:
    Nov 25, 2017
    Messages:
    20
    Likes Received:
    6
    And just to be sure I wasn't delusional (about this, anyway) here you can see my print bed heating to 110ºC. It started at about 24ºC and was this hot in about 45 seconds... I didn't wait for it to go all the way to 110ºC since after confirming you could indeed program it to 110ºC it didn't seem necessary. I know I set my ABS prints to 110ºC and the print won't start until the plate gets to temp. Since those prints started, and since you can definitely set the plate to 110ºC, the plate must also be able to get to 110ºC.

    [​IMG]
     
  18. SteveDjarrell

    SteveDjarrell Active Member

    Joined:
    Jan 9, 2015
    Messages:
    101
    Likes Received:
    40
    Thanks for the comments guys. Most i already knew. Reason i am going to dial in delrin is that is what is being requested for a prototyping contract i have. Marine equipment manufacturer, most of their product line is made of it. I would prefer to use 910 but they want to see or in delrin. Might hit you up for some tips though Geof.

    Also guys, i was just asking about the bed temp. No need to argue Fact is i would prefer to never take it above 100.

    Sent from my SM-N950U using Tapatalk
     
  19. RSC

    RSC New Member

    Joined:
    Nov 25, 2017
    Messages:
    20
    Likes Received:
    6
    No worries. It's not "arguing", it's "rational discourse". A productive discussion. At worst a heated (pun) debate. But not "arguing"... at least, I don't see it that way. I hope no one else does, either. If they do, my apologies, this was not my intent.
     
    mark tomlinson and SteveDjarrell like this.
  20. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    @RSC yes the screen will show what temp was set in your sliced G-Code or set in the web interface. I already mentioned that, but try setting a temp over 100°C on the Robo LCD. It will revert back to 100°C. Robo intended that the LCD be the primary means of input but thankfully they do not filter out settings coming from OctoPrint or sliced G-Code.
     
    SteveDjarrell likes this.

Share This Page