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

How hot can the bed get

Discussion in 'General Questions' started by drbanks, Mar 15, 2018.

  1. drbanks

    drbanks Active Member

    Joined:
    Dec 29, 2017
    Messages:
    141
    Likes Received:
    40
    Not asking for Robo's official specifications, but actual feet-on-the-ground experience: How hot can I reasonably get the bed heated to?

    I was using the new PID tuning wizard the other day, and the amount of time it took to get the bed to 80 degrees seemed to be awful long. Long enough that the PID tuning routine timed out when even though the target was 80 degrees, it was inexplicably showing around 84. Seems to imply that the heater is weak, and the regulation of the heater is poor.

    I'm used to using 100 degrees with ABS, and I'm starting to doubt the printer can even do that (Never mind the issue of it not being enclosed). And other filaments I see might want it to go even higher.

    It has no problem getting to the 60 degrees I use for PLA, but above that, the Delta-Temp/Delta-T slope really starts leveling out.
     
  2. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    The BuildTak should allow you to use ABS at 80°. I have played a lot with the bed heater and I am pretty sure mine is properly PID tuned and I routinely cannot exceed about 85°Cwithout a disconnect error. I have gotten the bed to 100°C but that is restarting the printer after a temperature fail while the bed is still hot. What I mean is if it fails, then reconnect and try again immediately, then when it fails again repeat the process. Robo's answer to the weak heater so far has been to widen the temperature ranges and to lengthen the time before a failure is detected. Not what I personally would do. This is not new to the R2, the R1 series also had a woefully undersized heater, even worse on that platform the heater only serviced approximately 67% of the build area. At least the aluminum bed on the R2 acts as a heat spreader.

    By the way, with BuildTak and /or PEI you can probably get away with PLA bed temps closer to 40-45°C
     
    drbanks likes this.
  3. Jerome Helbert

    Joined:
    Aug 19, 2017
    Messages:
    82
    Likes Received:
    29
    I had problems getting to 80C with my version 1 bed, but the version 2 bed seems considerably more powerful, I've gotten it between 105 and 110 while playing around.

    I actually had to use the octoprint web interface to set the bed temp higher than 100. The LCD wont accept inputs larger than 100C.
     
  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,672
    Likes Received:
    7,316
    Yea, even the R1 bed peaked at about 110 (if you were patient). That is about all you can get from the heaters they are using.
     
  5. Jeff Lastofka

    Jeff Lastofka Member

    Joined:
    Feb 21, 2016
    Messages:
    62
    Likes Received:
    26
    I scanned mine with an IR thermometer and got about 43C pretty uniformly over the surface of the bed set at 60C. I haven't tried other temperatures yet. It's sticking so strongly on PLA that I'm thinking of dropping the temp some anyway. On my R1+ the temperature was so non-uniform that I added heat tapes around the periphery underneath and ran them to a lamp dimmer for supplemental heat. I'd set it somewhat lower than what the temperature required and then let the controller make up the difference. If the printer weren't running at the moment I'd look under to see if there's room on this one for the same treatment. Not for uniformity. Just to give the heater a boost.
     
  6. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    @Jeff Lastofka That is the difference between using aluminum (which conducts heat very well) and a Borosilicate glass bed (glass is an insulator and heat tries to stay localized). The heater on the R2 pretty much covers the whole bed and it is a circuit board based heater. I have not found an easy way to swap it out for something more robust. I doubt that you can actually achieve anything close to 100° C for the bed without an error and disconnect message. But that is why we have PEI (version 1) or BuildTak (version 2) beds, they work with a lot less heat. ABS should stick with as little as 80° C, I routinely run PLA set to 40° C, Please note that with the diffusion offered by the bed surface your IR thermometer will likely read lower than the actual bed temps.
     
  7. Jeff Lastofka

    Jeff Lastofka Member

    Joined:
    Feb 21, 2016
    Messages:
    62
    Likes Received:
    26
    If the error is coming from their sensor getting above some temperature. adding more heat wouldn't stop that. If the error were from the system complaining about too much load on the heater circuitry, then supplemental heat should help. I'm familiar with the glass/aluminum/conductivity thing (retired mechanical engineer) but the IR temperature reading from the surface would be correct as long as the emissivity of the material is fairly close to the value set in the thermometer. A matte black surface like the Buildtak would be just about the standard value (around .97 I think I remember) used most everywhere I expect. I used a Raytherm thermometer I have in the kitchen. I also have a nicer Fluke one I can try later. I can let everything stabilize at room temperature and make sure the thermometer is reading the surface correctly as a double check.
     
  8. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    The error is an undersized heater speced at the maximum current draw of the mainboard it takes too long to heat and errors out of the temperature runaway protection.

    Also the surface isn't technically matte black, it is textured hence the scattering of the IR. But the Fluke should be a very close read, cheap Chinese ones measure all over the place.
     
  9. Jeff Lastofka

    Jeff Lastofka Member

    Joined:
    Feb 21, 2016
    Messages:
    62
    Likes Received:
    26
    Ah, so the error is the controller saying it's having too hard a time driving the heater. In that case, adding supplemental heat should help. Like, say a heat lamp over the bed while it starts up.....
    Or an oxyacetylene torch used carefully*
    -Jeff

    *just kidding. It's not possible to use an oxyacetylene torch carefully. At least, not indoors next to a $1500 printer.
     
    mark tomlinson and WheresWaldo like this.
  10. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Robo's solution is to extend the ranges (both lengthening the time and widening the temperature range) in firmware, so it errors out less. Kind of kludgy.
     
  11. Jeff Lastofka

    Jeff Lastofka Member

    Joined:
    Feb 21, 2016
    Messages:
    62
    Likes Received:
    26
    When I get around to trying ABS [ don't like the smell much:-( ] I'll just try 80C with the Buildtak first, and then probably use a heat lamp to jump start it if I want to go higher.

    Have people tried preheating to the PLA setting at 60C and then changing the setting to 100C to give the controller a head start? Or program another material at 80C. Preheat to that second. Then third, preheat to the final 100C.

    I think I like supplemental heat better than changing preheat targets two or three times. Just hold a heat lamp in there for X seconds and then fire up the system. (Perhaps a poor choice of words there.)
     
  12. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,672
    Likes Received:
    7,316
    Generally I avoid ABS (there are better materials, it has no properties that make it better other than price since it can be had dirt cheap at times)

    http://taulman3d.com/how-to-choose.html


    Don't limit yourself
     
  13. Jeff Lastofka

    Jeff Lastofka Member

    Joined:
    Feb 21, 2016
    Messages:
    62
    Likes Received:
    26
    I just checked with a good IR thermometer and the bed's right on target. I was wrong with the kitchen unit I used earlier. Need to toss that out. I now get 61-62C over the whole surface set at 60C.
    I see McMaster-Carr has a nice 6x6" silicone heater sheet for around $62 to $84 depending on thickness. 180W total, which a guy could stick under the bed and run to a light dimmer. Use that to heat the bed a little lower than your set point, then turn on the controller and let it make up the difference. I have an email in to them to ask about the thickness of the $62 version. The $84 version is very thin, like .010" or so.... They're adhesive backed and rated up to 150C.
     
  14. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Just be careful adding more heat to the bottom of the bed. The frame is made of injection molded ABS and mine is warped already from the limited heating of the existing bed. More heat may just make that frame warp more.

    Just a tip, try PLA printing at 40°C bed, it should work fine.
     
  15. Jeff Lastofka

    Jeff Lastofka Member

    Joined:
    Feb 21, 2016
    Messages:
    62
    Likes Received:
    26
    (My impression from reading these posts is that the controller will hold 100C, but it has a hard time getting there. I haven't tried mine that high yet.)
    I'm not suggesting to add _more_ heat. I'm suggesting to leave the Robo controller turned off and set the new heat pad to whatever max Wattage the controller used to use until the set point is near. Then the new pad would be dialed back to a supplemental level and the controller turned on as usual. The supplemental level would be less than the usual Wattage the controller uses to hold the set point, thereby easing the load on the controller but letting it do the temperature regulating as usual.
    That's the way I run my R1+, but I did that for flattening the temperature profile. The controller offloading was just a bonus.
    Or you could just use the new pad for pre heat and then turn it off entirely, but I'd prefer to let it help the controller a little as long as it's there.
    I do want to try the 40C PLA bed. Seems it should be fine, like you say.
    It's fun having a decent product that actually works and we're just tweaking things to preference:) Better than trying to save something that's just not good in the first place. I'm sure we've all been there:-(
     
  16. Jeff Lastofka

    Jeff Lastofka Member

    Joined:
    Feb 21, 2016
    Messages:
    62
    Likes Received:
    26
    I ran the PLA at 40C and it seemed good. In the process of setting the temperatures down in S3D, and the LCD, and then again in Octoprint (what a conglomeration of places to make settings - you can tell we're not using Apple products here....) I was annoyed (again) by the Octoprint hard coded temperatures for startup and wipe. I finally decided to just comment out those lines. I preheat using the LCD when I'm cleaning off the Buildtak and dusting my IR sensors anyway. I don't think I need Octoprint sticking in some (often incorrect) temperature in the process before the print G-codes take over. I guess I'll find out soon if that was a mistake:) Other people with different workflows might not like it, but I'm hoping it will work for me.
     
  17. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,672
    Likes Received:
    7,316
    I suspect we all do :) eventually.
     
  18. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    I pretty much stripped all the startup code from OctoPrint and just use On Connection, Print Ends, Print Cancelled code, so the printer will get out of it's own way and turn of the lights. I use the lights to tell if the printer is actually printing since it is in another room but has a glass panels in the doors, lights out and I know it is done.
     

Share This Page