Preheating/Soaking your chamber not only makes sure that your bed is up to temp, but also makes sure that the frame etc is. Its very noticable when printing without and soaking. As the chamber gets hotter, the frame expands which causes nice "curved" prints.
Yea, my Voron belts are quite (too) loose when cold, but after a good 30min-1hr heat soak they are noticeably tight (and properly tensioned). Printing without a heat soak on mine would be a great way to achieve terrible print quality. 450mm of aluminum extrusion expands quite a lot.
Wait could this be the cause i always have too much squish on the front or the back of the pei? And could bed mesh solve that? If i level when preheated to printing temps?
I also find it interesting that on Bambu Printers they run ABS/ASA at 90C and it has some of the most constant adhesion I have seen. I guess compared to the 80-85C the top of many printers may be reaching 90C would be a step up. They also use thinner beds so regardless of location the thermistor (should) be closer to accurate.
@@ModBotArmy That is interesting. I don't have one, but on my Voron 2.4, 110c will get the top of the bed to 100c. I have wondered what impact the magnetic sheet as well as the double sided PEI have on all this. They just might be causing some of that temperature difference.
Good stuff! This is the reason we use 115C bed temps for our VORON parts profiles. Most surprising is how consistent your results were across different bed setups.
Try insulating the top of the bed, and re-run your tests. Remember the top is a radiator and will lose heat to the atmosphere. A layer of plastic will insulate the top plate, and your readings will most probably be where you expected. Physics is a great tool when understood. I tend to use preheating for the first layer or so then turn it off to prevent melting of the first layers with PLA.
Great vid Daniel! that lines up with our testing on the V0 bed as well. The biggest difference on the V0 was made by using the thermistor that screws into the bed itself at the center. I think LDO's edge-to-edge heater accomodates this too, and it gives readings within 1 or 2 degrees of the surface of the bed.
Thanks man :). I need to source a bed that has a top mounting channel for a thermistor or a screw in type like you are describing that is getting the reading deeper into the bed.
I have a cork plate under my heater mat, so the heater is sandwiched between the cork and the aluminium. It makes a huge difference compared to typical designs where the heater mat and the plate can lose a lot of heat to the air below the build plate.
@@RocktCityTim It was just just a 5mm thick 100x50cm cork plate sold on Amazon as a pin board from which I cut a square to fit my print bed. I clamp the whole sandwich together with the clamps that my CR-10 uses to clamp together glass bed and aluminium plate.
In my v2.4 I stuffed glas wool wrapped in caption between bed and seperation plate. And I'm used to give always higher bed temp than advised. Would be interesting to find out how the producers measure their "advised" bed temp ;-)
It can help, but it depends on what you want it to do. If it's open air, then you'll just be losing heat to the air, true. In an enclosure? Your bed is the primary heating element for it. Indeed, both the Nevermore and the Filter actively harvest heat from the bed to increase and maintain chamber temperatures. For an enclosure, this insulation makes it worse.
Slightly unrelated, my bed is glass with a garolite plate on top. Preheating makes a huge difference to surface level, even with only a 60 degree bed temp.
Preheating the bed is as important to corexy as it is to bedflingers, just for a different reason. The mostly cheap, thin beds have a tendency to warp, so preheating before a mesh is basically essential. I even have a start gcode for heat soaking my chamber to 40c, just to be extra sure. A part coming loose mid print on a full plate is the last thing i want. Fantastic video as always.
Glad to see I'm not alone with this kind of behavior. My first big prints in ABS were nightmares because of that. I replaced the thermistor once with no success. Since I simply raise temperature by 15 degrees.
this lines up with some of what I've been seeing on my 2.4. I heat soak and run the initial layer at 100C and subsequent layers at 110C. I had tried bumping it up in a bid to improve bed adhesion. apparently was spot on the issue.
I've had a cheap 200W fan heater in my old printer enclosure for years, initially because the bed heater didn't quite have enough power to maintain 110'C for ABS at room temperature. I quickly noticed the drop in warping, so preheating with bed and heater was done for every ABS print.
For future videos: Keep in mind that thermal cameras do not do well with reflective materials. This may be part of the reason the temperatures appeared uneven at 6:28
Insulation makes sense only from outside for enclosed printers. I stuck 12mm thick yoga mattress pieces on all my side, back and top panels and the doors. With 110C bed can crank up the chamber to real 60-65C in 2 hours in the shed. Will see what results will have during winter.
When using thermocouples for surface measurement, it's a good idea to add thermal paste. I do such measurements often so have gone a step further and picked up some very thick (5mm) thermal pads and I stick the thermocouple inside the pad. Without proper thermal coupling both the surrounding air and heat loss to the thermocouples wires impact the repeatability of measurement (up to 5deg in my experience). There are proper probes for surface measurement but I've found thermal compounds/pads to be just as accurate for a fraction of the price.
I expected a diff, but 15-20C was wider than I expected. I was told early on the preheat for PETG, so I've always set my nozzle and beg temps about 20 minutes before I start printing. I'm now experimenting with PETG-CF and PLA-CF filaments on my SV06 Plus using Klipper and I think I'll push my bed temps to 95C so that the face can get to the expected 85C for the prints.
I think why the temperature is peaking at 85°C is because the printer thinks it has reaches 100°C and is reduing the power to the bed. I was thinking, if there was a way to attach the temperature sensor above the heated bed, between the top surface of the bed and the magnetic sticky layer, the temperature control would be more accurate.
But if all of us are getting 85C beds when we think we're using 100C, it means that 85C is actually a good temperature for bed adhesion. Heating the top to a true 100C might be too hot.
As already mentioned, the heat cameras may give a wrong temperature depending on the material. Matt black plastic gives a much higher temperature than shiny aluminum. This is the εr that is shown on a lot of the cameras. My IR thermometer was delivered with matt black stickers for that reason. Most have a fixed 0.95 value, more versatile cameras will have a configurable setting to allow the right temperature shown for different materials. But there is also the direct cooling of the thermocouple due to draft over the build plate. Since the thermocouple is not welded to the plate with a thermal compound, the slightest bit of draft will result in the top of the thermocouple being cooled down. Adding a small dab of thermal compound does wonders. I just tested this (2 probes on the bed, one with and one without thermal paste) and the one with the thermal compound heats up faster and goes all the way to 90+ while the other sticks around 82 C.
Given that the print bed is a giant radiator, I'm not surprised it's a lot cooler at the top than at the bottom. That's how radiators work, after all. Remember that some of that lost heat is transferring into your plastic, which is after all the point. Anyway, my main thought is this: regardless of the actual measured temperature of the bed, if you have a system that works with your printer, with your preferred filament, in your environment - don't fret the numbers. Whenever I'm baking something in the oven, I have to set the temperature about 25F less than the recipe calls for. Likewise, when I'm printing ABS on my X1C, I let it "bake" until the chamber temp is reading about 45C and don't take the parts out until the chamber has cooled to under 35C. This seems to work, and I don't care what the actual measure temperature on the bed is, I just know what settings and process I need to have good prints and unburnt food.
Oh 100% I was not trying to state that you should be chasing higher temps especially if you are having success. More that if you are not having success and you have tried other things, it may be worth looking into the temps you are actually hitting and not what you think you are.
Great video as always. My 2.4 & V0 both had Precise Printer Parts build plates that have thermistor slots in the top. On the 2.4 I read both the top & bottom of the bed surface. It is wild how different the temps can be when a print starts. I personally always bake beds, even when printing PLA I find it helps bed adhesion.
I need to look into one of those beds. It would be interesting to see how close it can get the top of the bed to correct temp measuring the top of the bed. Are you using a macro typically or just doing a manual preheat and then when you see the thermistors within a certain range sending a print?
@@ModBotArmy I generally start pre-heating the bed when I go to slice a design. If it is a material that needs chamber heating I'll do it earlier of course. Otherwise, I've got the thermistor on top of the bed set as my "Bed Temp" sensor, so slicer settings go off that surface temperature & not the underside. That one I just monitor for my own information.
For me this makes sense, knowing how temp. sensors work in general. However, the real question is - if filament printing bed temperature is stated, let's say at 100c by manufacturer, is it actually tested on 85c, as most of printers will suffer from this behaviour, or should we actually aim for true 100c with 115c bar on printer? In general, you do what works on your printer, guidelines are what we get from manufacturer.
Very interesting. For your thermistor installation, I would use an thicker insulated tape with some foam. Otherwise the sensor is being cooled from the top side while being heated from the bed side.
A few others have mentioned this. Kapton is a thermal/electrical insulator which is why I chose it. It also was the main adhesive method used for thermistors on most printers before they began embedding them. That being said, I dont doubt there are better options out there for at least this testing method. I may look into a few alternatives and see if there really ends up being much difference in the results.
I added a 1 minute pause to my starting g code to wait for temperature to settle at all corners of my bed. I use garolite bed and its quite warped if corners are cooler than center, so i really need more time to start printing on an even surface.
@modbot What i did is machine a small hole right below the surface on the build plate roughly 1mm down from the surface and use that to pid tune , What is happening is the thermal sencor on bottem is cutting the heating early so the surface never reaches the correct temp, your temp sencor needs to be as close to the top of the build plate as you can get it so the tuneing is calbrating for the surface temp not the bottem temp
Some other things you can do are mount fans under your bed to push the hot air around the chamber while preheating for more even distribution. You can also put your hotend at 160 and your part fans at max speed during preheating to work as a little chamber heater. Preheating ensures everything expands and contracts to where its going to be for the entire print. You want to preheat, then mesh, so that your mesh doesnt only take into account the warping of your bed, but actually the frame itself as well. You dont want the geometry of the frame to change from heating after youve already started your print.
Yes! Anything that's not another: printer that I've been sent free of charge, but all the thoughts are my own, etc, etc, anyway, you can order it from my affiliate link below.
I found exactly the same thing with my Voron Trident, I have to set the bed temp at 125 to get 110 on top. This also meant I needed to get a new thermal fuse as the old one was only rated to 125. If I try and bed level before properly soaking then the readings can constantly change
Great idea to test this out. Very interesting results. I wouldn't just apply a blanket +15C to your bed temp though. I have a suspicion that the real relationship is going to be more of a logarithmic plot, opposed to linear. That is, the higher the temp, the larger the delta. The bed is a heatsink.
I remember adjusting the temp table on my v2.4 to shift the 100c resistance value to what is really is when the top reads 100c, it's pretty simple to do. I only have this issue on my v2.4 since I could not get any insulation under bed. My other printers which have insulation actually read 2c higher on the top than target by thermistor. When you get into the 100c range on an non insulated bed, the variance is pretty drastic, so tweaking the temp table would do some good.
My Tiny-M had 20 degree difference from the probe to the bed surface. I just borrowed a thermal camera so I can do some testing. so to get to 100 for abs printing I set the bed to 120.
I had an old printebot with the factory thick bed heater. I found that once I did a 10min heat soak on the bed, I had much better results. This tracks with what I had observed. On a side note. When I had seen people saying "I printing with X degrees on the bed." You have to remember your setup vs theirs.
I’m a bit surprise you guys don’t have any bed insulation below the heater. That can help and even save a bit of money on the ole power bill. At work I have two Modix 120X 3D printers. We replaced the bed with a 1 inch thick MIC6 plate for amazing flatness. I let that heat up for an hour before printing and check it with the FLIR camera.
I'm really glad you posted this. I got a Gulf Coast robotics bed to replace the heatbed on my Ender 3 S1 Pro (which shorted out). Noticed I had adhesion issues while using the same temperatures while printing PLA on the new bed, so I got an IR thermometer and discovered the same phenomenon that you've pointed out here. I assumed that the thermestor that came with the new bed was bad... seems that's normal based on what you're showing me. 🤨
On my Ender 3 Max, the bed has a pretty significant droop in the center. After 15 minutes of pre-heating at 60C, the droop is gone and I have a nice flat bed, corner to corner. At the same time, the nozzle and heater block is expanding and heading down towards the bed. So, for the first 15 minutes of heating, I've got a bed expanding upwards and a nozzle expanding downwards. After 15 minutes there are no measurable changes in bed or nozzle, even after hours of soaking.
This is interesting. I hadnt even considered the expansion factor in this case. It makes sense which is why you want to do a mesh at temps but someone else mentioned frame expanding and in your case the hotend itself.
On my old Replicator the build plate was so small there was about a 1-2c delta. Still, I put a wool rim around the bed and an air gapped film of Kapton tape. That allows the heat to move around and even out under the bed 😉 After i did that the thin little bed never warped after and i could print ABS to the edges 🥰 i used to have to bend it back every few months 😅
I've noticed on my cr6 that I'll sometimes have issues in the first layer, so I'll cancel print and restart right away, cleaning off the material while the machine does its startup shenanigans. And it'll usually do much better the second time, I was suspecting it was the bed surface temp and I try to remember to heat up the bed and nozzle manually as I'm preparing a print. I've seen a lot better results for the early layers this way. It is something I'll probably integrate into the firmware once I replace the factory setup.
I had a problem regarding nozzle temperature that was ultimately solved by changing the thermistor type. After a certain point the differential between reported and externally measured temperature grew too large and visible artifacts and extruder clicks followed.
It seems like this temperature differential (between the print surface and heating pad thermister) are accounted for in filament manufacturer temp recommendations. If you set your temp high enough so that the print surface is at the recommended temperature, you will see every indicator that your bed temp is way too high. So... It would be cool if you could revisit this and try printing with an actual measured bed temp that falls withing the 'recommended' bed temp for a given filament - my guess is the print would be pretty horrible with many sings that you're printing with just way to high of a bed temp.
Seems like we do need chamber heating to get better control of temperature. Of course it needs to be done safely, and it doesn't help that some groups don't even allow discussing it.
Daniel: Nice effort! A couple of suggestions: Just as a reality check, it would be useful to run your thermocouple bed tests with a layer of insulation covering the top of the bed (ie: also covering the thermocouples.) That greatly reduces dissipation from the top of the bed, greatly reducing the heat flow through the bed (bottom to top), and thus also the temperature drop bottom to top. The result should be that the bed heats uniformly throughout its thickness, and hence your thermocouples should read the same temperature as the set point. This will validate your thermocouples and instrument and also .checks that the printer's thermocouple or thermistor is sensible. Your tests in this video using thermocouples only on the top surface don't actually verify that the printer's sensor is getting up to the set temperature, even if the printer thinks so. Next, in the comments there are suggestions to attach the thermocouples with thicker (heat) insulating tape, on the principle that then the thermocouples won't be cooled from above. However I think that's a mistake: If you place insulating tape on top of a thermocouple that will reduce dissipation from that neighborhood of the bed, creating a local hot spot, so your measurement will not be representative of the rest of the surface. Instead, the tape you are currently using exposes the top of the thermocouple and local bed to approximately the same air conditions as the rest of the bed, so the temperature is representative. Anyhow, nice to see folks performing experiments like this and trying to get at some ground truth!
Im curious on how much the magnet is influencing the temperature reading as it is the least conductive material in the assembly. That could really justify the added cost of a bed with integrated magnets like the prusa mk3 has
I generally presoak when printing large parts or on the v0.1 with its thick build plate.. Now let me see if I change my mind on this at the end of the video. :)
I have found with klipper their thermistor settings are a bit hit and miss. You need to do some tweaking to get things right. Would be interesting to see if you done the same with a marlin machine, I have read they are more accurate.
Could you perform a test with the printer upside down? I think you would see a lower difference between the two sides. Putting the control thermistor on the opposite side (of the part surface), and without the build surface between the aluminum and the part, will always give you a lower temperature. Putting the control thermistor on the top make would provide better control. Unfortunately it would interfere with the build area. I wonder what it would be like if the thermistor were placed at a corner and the control temperature dropped by a couple of degrees. I think this would give a closer target temperature under the build area. A further question is, how does a part over the build area affect the temperature. But at that point I would expect that it would be less of a concern after a few layers hade been completed.
There is a big problem with the methodology here which put in question the remarks about top-bottom plate temperature. Thermistors have a much faster response to temperature changes when compared to thermocouples. Also, regardless of the probing type, if you use more than one metrology equipment, it is required to have them calibrated, otherwise risking a few degrees of error between them, up to tens of degrees if using uncalibrated IR meters to probe metal surfaces. I would suggest using the same 4x thermocouple meter to measure top and bottom.
Aluminum is quick to heat, but it's also quick to dissipate heat. I feel like insulating the bed would go a long way to improve this issue, but would add weight, complexity and still not be an end all solution.
Hello, I am generally learning a lot from your videos but I see some hickups in this one. Temperature measurements is an area filled with many traps. For one, your thermocouples basically have one or two tiny point of contact with the PEI, and have an hugely bigger area of contact with the kapton tape which are flexible and coated with glue. So the thermocouples are in the middle of a thermal divider, the ground being the ambiant temp, the supply voltage being 100c, the top thermal resistant being a few points and a little bit of air, the bottom thermal resistance being a few square mm of kapton in series with air convection thermal resistance. Bare thermocouple like that are not good at measuring surface temperature. Thermal imagery is the final boss in term of temperature measurement. Many people do not understand the tremendous impact of emissivity, background temperature, specular and diffuse lambertian or non lambertian reflexion (i think the last one are the causes some inhomogeneity you saw)
You might be right. I chose Kapton because it is a thermal insulator and prior to embedded bed thermistors most printers I tested out had them attached using Kapton tape as well. Maybe this being a good choice was an incorrect assumption on my part. Any suggestions on a better insulator for the thermocouple?
Would love to see a video comparing popular production printers to see the difference. I was definitely not expecting so much difference in temperature. Given these results I would expect there to be a significant difference between different production printers too. Great video for provoking some head scratching.
@ModBot This might all be fine and correct, but I wonder if there was a significant flaw in your method. Attaching a thermistor with tape to the top of the bed exposes the bulb on the end to a sector maybe 60° or so exposed to the bed surface while the remaining 300° is exposed to the surrounding air & thin tape. My guess is the thermistor is averaging the reading down. I would like to have seen if a tiny amount of insulation between the thermistor tip and the tape, like the size of a q-tip or something, might give you a more accurate measurement of the actual bed temperature instead of a kind of average of the bed and air. I mean, you saw with a camera how much cooler just the tape was than the bed and that whole surface was literally attached 100% to the bed - the thermistor has even less contact with the flat bed surface. Of course extruded plastic will exhibit similar cooling as it's built up, but at least it would be good to know for sure the temperature of the bed surface and I'm not positive that measurement method was accurate. Again, I might be wrong. And thanks for the video. It's still good info and illustrates that the bed is constantly losing heat at the surface and makes sense to keep that in mind.
You might be right. I chose Kapton because it is a thermal insulator and prior to embedded bed thermistors most printers I tested out had them attached using Kapton tape as well. Maybe that was an incorrect assumption on my part. I took the outer surface of the Kapton being cooler being due to its insulating properties but I will make no claim that is definitive. Any suggestions on an insulator?
@@ModBotArmy D'oh. I didn't consider that the tape looked cooler because it might be insulating the heat some off the top. Slap forehead. Obvious now at least as a possibility. But in case you want to try a quick experiment to see if it makes any difference, I was thinking just a small piece of that insulating paper stuff we used to wrap hotends back in the day. Just taping a little of it over the top (not surrounding it of course) might give an idea whether it's a thing or not. I know kapton is good for witstanding heat, but I never thought of it as an insulator, so it might be worth a shot.
I would have loved to see you measure the underside of the bed with a separate thermocouple so you could compare the printer reading to you measurement setup.
You just confirming data from time when beds were heated by bunch of power resistors and frames were made from all thread or mdf boards. You NEED isolation layer under the bed. Its not optional if you want higher temp.
Id be interested in knowing how open builds cool down by putting the sensors a little above the bed to see how much a difference there is and the cool speed and stuff
I was kind of hoping you'd come to the conclusion you did. I've always wondered just how strong the difference was, and I'm especially curious what you'd find with the Nevermore and/or bed fans running as well, to see if those help or hinder the balancing of temperatures on the build plate.
Great video sir! It would be interesting to see what moving the bed thermistor to the top side of the bed would make to controlling/realizing actual commanded temps. I would think that this would drive the top bed surface temp to be actual commanded temp where it matters?
very interested to see longer soak times and the effects on bed tempps. Also, curious to see the delta between thick aluminum beds and the mk52 with respect to delta between thermistor and measured temp at high temps
Thin beds usually have a 2-3c delta, glass is about 5c. I have a specific print profile that has a 5 minute heat soak time after the bed reaches temp to then start heating the extruder. I also picked up a small 60w heater and temp controller years ago that I was going to use. But that has seemed to work well enough that I haven't set it up 😅 *i do a lot of ABS printing, small and large (285x285 and 315mm high)
So you are comparing your probes reading with the printer beds built-in reading ? If you want any relevant measurement, you should have had the same type probe on top and bottom, so you could also compare it to how the printer read temperature - I am pretty sure that is not calibrated !?
Hi Daniel, I know in Voronland they recommend limiting heating rate to 75 - 80% of heater power, curious if changing or removing this limit impacts the result of this testing? Cheers
My $0.02: My Ender 3 with a bed sensor would report varying bed heights while heating for about five minutes. After that, the bed would cease warping and expanding measurably.
I would not have expected that much of a gap. Could improvements in print quality / bed adhesion be had by increasing the setpoint such that the surface was the desired setpoint? I was also considering the X1 Carbon.... I'd be curious how that unit would fare in a similar test. I've been really impressed by their engineering and wonder if they've already taken that into account. Thanks again for the content... much appreciated.
I feel like it would have skewed the results though. Although I am sure the powder coated plate transfers heat well compared to the aluminum or magnet sheet I still wanted it on the very top.
Most printers are firmware limited to ~100-110C. Are the silicone heaters able to go higher or will they melt down? It would be interesting to see how much higher your set point has to be to get the 100C surface temp. Does this mean that ABS prints well at 85C bed temp abs we’ve all just assumed it is 100, is it worth chasing a 100C surface temp?
I personally print at 95 with Hatchbox abs. But if my bed is actually 70 C, that’s really interesting. When I set my bed to 100 it’s likely at the glass transition temp because the part deforms when I take it off early. Maybe there’s something mixed in that drops the temp/makes it print even more reliably? Really interesting question for how many printers would show the same/similar results
None of the printers have bed fans. There is a Nevermore on the 2.4 that could be used as one but in this case no I just tested heat coming off of bed with the only real variable being heater types, bed thicknesses, and enclosed vs not.
What I find most irritating is that the final bed temp is nowhere near the setpoint. My profiles are a lie! We're probably all printing way colder than we think. I usually preheat my X1C until the chamber temp shows 40°C or so. It would be interesting to test production machines (like Bambu printers, wink wink, nudge nudge).
Good info for the video, but poor title. Betteridge's law of headlines is an adage that states: "Any headline that ends in a question mark can be answered by the word no."
Much larger difference than i thought. I thinit would have been worth an extra hour of your time to have let it soak for the hour. Just for the science if it.
Would be great if you try to understand what temperature you have to set in firmware to get real 100 degrees on top. The reason why top temp capped to 85 - probably PID control, when firmware think in reached 100 - it reduces amount of heating a lot
I imagine part of it is PID and that printer lowering heat. I still would have guessed that the heat would have eventually evened out or at least closed its range. One concern someone brought up is how hot the silicon heaters can reach before its an issue. If there is a 15C spread and you are printing at 115 on the bottom to get 100 on top other factors such as adhesive on the bed will need to be considered.
Yeah? So what? Your results are very similar to my own experience with my Voron 2.4r2 that has the same honeybadger heater. When discussing the issue on Discord, I was told that the variation between the top and bottom thermistor readings was a function of the fact that the bottom thermistor is directly touching the heater, and the kapton tape was not a very effective thermal bond to couple the thermistor with the flex plate (which is not well bonded with the bed plate beneath it), all of which explains a large portion of the variance even after soaking. The IR measurements a not very reliable due to the reflective nature of the PEI plate, so a mat black surface would allow for more accurate IR camera readings. The issue with this video, is that you never explained what to do about the variation. Nobody seems to know how to best quantify the print surface temperature compared to the bed sensor reading so you can deterministically adjust your filament temperature slicer settings.
Great test. This is the reason for Prusa having the resistive heater just under the springsteel. It can give some weare, when you remove the springsteel thou. Thomas Salandere have made a temperature video, about the unifomity if heat spread of a headbed. ua-cam.com/video/yPy0zDpmc5U/v-deo.html Thanks for sharing your experience with All of us 👍😀
Aluminum is an extremely efficient heat-sink so it would make sense that on the opposite side of the plate from the heater it would be cooler, the aluminum is doing what it does really well, dissipating heat.
Might not be the case with all materials. There are a lot of very sensitive filaments out there that will warp not matter what if perameters are not in a very tight range.
Hello and a good day toooo all the people out there! Is there somebody who have a FIBTIP Nozzle, because they promised it's the last you need! Indestructable, non-stick technology With fibtip's unique ceramic diffusion coating you will get unrivaled and everlasting performance against even the most abrasive filaments (copied from there side). Some tried the Nozzle or hey a Video Idea 💡!?! I wish all the best to all and enjoy your day wherever you are😁😁😁👍👍‼️‼️
Preheating/Soaking your chamber not only makes sure that your bed is up to temp, but also makes sure that the frame etc is.
Its very noticable when printing without and soaking. As the chamber gets hotter, the frame expands which causes nice "curved" prints.
I love hearing the few creak noises I get when preheating. Reminds me it's working.
@@condorman6293 yup absolutely
Yep, absolute must for sequential printing
Yea, my Voron belts are quite (too) loose when cold, but after a good 30min-1hr heat soak they are noticeably tight (and properly tensioned). Printing without a heat soak on mine would be a great way to achieve terrible print quality. 450mm of aluminum extrusion expands quite a lot.
Wait could this be the cause i always have too much squish on the front or the back of the pei?
And could bed mesh solve that? If i level when preheated to printing temps?
This is why for ABS, many set the bed temp for a print to 110c or 115c. Those temps will get the top of the bed to 100c.
I also find it interesting that on Bambu Printers they run ABS/ASA at 90C and it has some of the most constant adhesion I have seen. I guess compared to the 80-85C the top of many printers may be reaching 90C would be a step up. They also use thinner beds so regardless of location the thermistor (should) be closer to accurate.
@@ModBotArmy That is interesting. I don't have one, but on my Voron 2.4, 110c will get the top of the bed to 100c. I have wondered what impact the magnetic sheet as well as the double sided PEI have on all this. They just might be causing some of that temperature difference.
If you are using 110c the chamber temp gets higher. Less warping. Using Nevermore fan mod increases the chamber temp to around 50c.
Good stuff! This is the reason we use 115C bed temps for our VORON parts profiles. Most surprising is how consistent your results were across different bed setups.
You use, not all. Mine is calibrated on top surface
Try insulating the top of the bed, and re-run your tests. Remember the top is a radiator and will lose heat to the atmosphere. A layer of plastic will insulate the top plate, and your readings will most probably be where you expected. Physics is a great tool when understood. I tend to use preheating for the first layer or so then turn it off to prevent melting of the first layers with PLA.
Great vid Daniel! that lines up with our testing on the V0 bed as well. The biggest difference on the V0 was made by using the thermistor that screws into the bed itself at the center. I think LDO's edge-to-edge heater accomodates this too, and it gives readings within 1 or 2 degrees of the surface of the bed.
Thanks man :). I need to source a bed that has a top mounting channel for a thermistor or a screw in type like you are describing that is getting the reading deeper into the bed.
I have a cork plate under my heater mat, so the heater is sandwiched between the cork and the aluminium. It makes a huge difference compared to typical designs where the heater mat and the plate can lose a lot of heat to the air below the build plate.
Are you just using standard craft-store self stick cork, or something special?
@@RocktCityTim It was just just a 5mm thick 100x50cm cork plate sold on Amazon as a pin board from which I cut a square to fit my print bed. I clamp the whole sandwich together with the clamps that my CR-10 uses to clamp together glass bed and aluminium plate.
In my v2.4 I stuffed glas wool wrapped in caption between bed and seperation plate. And I'm used to give always higher bed temp than advised. Would be interesting to find out how the producers measure their "advised" bed temp ;-)
It can help, but it depends on what you want it to do. If it's open air, then you'll just be losing heat to the air, true. In an enclosure? Your bed is the primary heating element for it. Indeed, both the Nevermore and the Filter actively harvest heat from the bed to increase and maintain chamber temperatures. For an enclosure, this insulation makes it worse.
Slightly unrelated, my bed is glass with a garolite plate on top. Preheating makes a huge difference to surface level, even with only a 60 degree bed temp.
Preheating the bed is as important to corexy as it is to bedflingers, just for a different reason. The mostly cheap, thin beds have a tendency to warp, so preheating before a mesh is basically essential. I even have a start gcode for heat soaking my chamber to 40c, just to be extra sure. A part coming loose mid print on a full plate is the last thing i want. Fantastic video as always.
Glad to see I'm not alone with this kind of behavior. My first big prints in ABS were nightmares because of that. I replaced the thermistor once with no success. Since I simply raise temperature by 15 degrees.
this lines up with some of what I've been seeing on my 2.4. I heat soak and run the initial layer at 100C and subsequent layers at 110C. I had tried bumping it up in a bid to improve bed adhesion. apparently was spot on the issue.
I've had a cheap 200W fan heater in my old printer enclosure for years, initially because the bed heater didn't quite have enough power to maintain 110'C for ABS at room temperature. I quickly noticed the drop in warping, so preheating with bed and heater was done for every ABS print.
For future videos: Keep in mind that thermal cameras do not do well with reflective materials.
This may be part of the reason the temperatures appeared uneven at 6:28
Good to know. I did notice some slightly odd behavior that may have been related to this then. I will keep this in mind.
Insulation makes sense only from outside for enclosed printers.
I stuck 12mm thick yoga mattress pieces on all my side, back and top panels and the doors.
With 110C bed can crank up the chamber to real 60-65C in 2 hours in the shed.
Will see what results will have during winter.
When using thermocouples for surface measurement, it's a good idea to add thermal paste.
I do such measurements often so have gone a step further and picked up some very thick (5mm) thermal pads and I stick the thermocouple inside the pad.
Without proper thermal coupling both the surrounding air and heat loss to the thermocouples wires impact the repeatability of measurement (up to 5deg in my experience).
There are proper probes for surface measurement but I've found thermal compounds/pads to be just as accurate for a fraction of the price.
I expected a diff, but 15-20C was wider than I expected. I was told early on the preheat for PETG, so I've always set my nozzle and beg temps about 20 minutes before I start printing. I'm now experimenting with PETG-CF and PLA-CF filaments on my SV06 Plus using Klipper and I think I'll push my bed temps to 95C so that the face can get to the expected 85C for the prints.
I think why the temperature is peaking at 85°C is because the printer thinks it has reaches 100°C and is reduing the power to the bed. I was thinking, if there was a way to attach the temperature sensor above the heated bed, between the top surface of the bed and the magnetic sticky layer, the temperature control would be more accurate.
But if all of us are getting 85C beds when we think we're using 100C, it means that 85C is actually a good temperature for bed adhesion. Heating the top to a true 100C might be too hot.
I know there's a pull-up resistor value for the hotend in Klipper, but is there a similar offset value for the bed?
As already mentioned, the heat cameras may give a wrong temperature depending on the material. Matt black plastic gives a much higher temperature than shiny aluminum. This is the εr that is shown on a lot of the cameras. My IR thermometer was delivered with matt black stickers for that reason. Most have a fixed 0.95 value, more versatile cameras will have a configurable setting to allow the right temperature shown for different materials.
But there is also the direct cooling of the thermocouple due to draft over the build plate. Since the thermocouple is not welded to the plate with a thermal compound, the slightest bit of draft will result in the top of the thermocouple being cooled down. Adding a small dab of thermal compound does wonders. I just tested this (2 probes on the bed, one with and one without thermal paste) and the one with the thermal compound heats up faster and goes all the way to 90+ while the other sticks around 82 C.
I am surprised it took you this long to look into this… Since this is very very old news, but it is very nice to see you cover it.
Given that the print bed is a giant radiator, I'm not surprised it's a lot cooler at the top than at the bottom. That's how radiators work, after all. Remember that some of that lost heat is transferring into your plastic, which is after all the point.
Anyway, my main thought is this: regardless of the actual measured temperature of the bed, if you have a system that works with your printer, with your preferred filament, in your environment - don't fret the numbers. Whenever I'm baking something in the oven, I have to set the temperature about 25F less than the recipe calls for. Likewise, when I'm printing ABS on my X1C, I let it "bake" until the chamber temp is reading about 45C and don't take the parts out until the chamber has cooled to under 35C. This seems to work, and I don't care what the actual measure temperature on the bed is, I just know what settings and process I need to have good prints and unburnt food.
Oh 100% I was not trying to state that you should be chasing higher temps especially if you are having success. More that if you are not having success and you have tried other things, it may be worth looking into the temps you are actually hitting and not what you think you are.
Great video as always. My 2.4 & V0 both had Precise Printer Parts build plates that have thermistor slots in the top. On the 2.4 I read both the top & bottom of the bed surface. It is wild how different the temps can be when a print starts. I personally always bake beds, even when printing PLA I find it helps bed adhesion.
I need to look into one of those beds. It would be interesting to see how close it can get the top of the bed to correct temp measuring the top of the bed. Are you using a macro typically or just doing a manual preheat and then when you see the thermistors within a certain range sending a print?
@@ModBotArmy I generally start pre-heating the bed when I go to slice a design. If it is a material that needs chamber heating I'll do it earlier of course. Otherwise, I've got the thermistor on top of the bed set as my "Bed Temp" sensor, so slicer settings go off that surface temperature & not the underside. That one I just monitor for my own information.
Great video, I always presoak with my vorons at 110*C for 20min and has worked perfectly.
Always very nice to see your Voron Switchwire. I love my Switchwire and enjoyed the build soo much
For me this makes sense, knowing how temp. sensors work in general. However, the real question is - if filament printing bed temperature is stated, let's say at 100c by manufacturer, is it actually tested on 85c, as most of printers will suffer from this behaviour, or should we actually aim for true 100c with 115c bar on printer? In general, you do what works on your printer, guidelines are what we get from manufacturer.
Very interesting. For your thermistor installation, I would use an thicker insulated tape with some foam. Otherwise the sensor is being cooled from the top side while being heated from the bed side.
A few others have mentioned this. Kapton is a thermal/electrical insulator which is why I chose it. It also was the main adhesive method used for thermistors on most printers before they began embedding them. That being said, I dont doubt there are better options out there for at least this testing method. I may look into a few alternatives and see if there really ends up being much difference in the results.
I added a 1 minute pause to my starting g code to wait for temperature to settle at all corners of my bed. I use garolite bed and its quite warped if corners are cooler than center, so i really need more time to start printing on an even surface.
@modbot What i did is machine a small hole right below the surface on the build plate roughly 1mm down from the surface and use that to pid tune , What is happening is the thermal sencor on bottem is cutting the heating early so the surface never reaches the correct temp, your temp sencor needs to be as close to the top of the build plate as you can get it so the tuneing is calbrating for the surface temp not the bottem temp
Very good video! Clear concept, real measurements, no crazy conclusions. Great work!
Some other things you can do are mount fans under your bed to push the hot air around the chamber while preheating for more even distribution.
You can also put your hotend at 160 and your part fans at max speed during preheating to work as a little chamber heater.
Preheating ensures everything expands and contracts to where its going to be for the entire print. You want to preheat, then mesh, so that your mesh doesnt only take into account the warping of your bed, but actually the frame itself as well. You dont want the geometry of the frame to change from heating after youve already started your print.
Now this is the type of content we need!!!.... Good job Modbot!
Keep it coming!
Yes! Anything that's not another: printer that I've been sent free of charge, but all the thoughts are my own, etc, etc, anyway, you can order it from my affiliate link below.
I found exactly the same thing with my Voron Trident, I have to set the bed temp at 125 to get 110 on top. This also meant I needed to get a new thermal fuse as the old one was only rated to 125. If I try and bed level before properly soaking then the readings can constantly change
Great idea to test this out. Very interesting results. I wouldn't just apply a blanket +15C to your bed temp though. I have a suspicion that the real relationship is going to be more of a logarithmic plot, opposed to linear. That is, the higher the temp, the larger the delta. The bed is a heatsink.
Yup, that’s exactly why we run 115C bed on all of our profiles for VORON parts.
I remember adjusting the temp table on my v2.4 to shift the 100c resistance value to what is really is when the top reads 100c, it's pretty simple to do. I only have this issue on my v2.4 since I could not get any insulation under bed. My other printers which have insulation actually read 2c higher on the top than target by thermistor. When you get into the 100c range on an non insulated bed, the variance is pretty drastic, so tweaking the temp table would do some good.
My Tiny-M had 20 degree difference from the probe to the bed surface. I just borrowed a thermal camera so I can do some testing. so to get to 100 for abs printing I set the bed to 120.
I had an old printebot with the factory thick bed heater. I found that once I did a 10min heat soak on the bed, I had much better results. This tracks with what I had observed.
On a side note. When I had seen people saying "I printing with X degrees on the bed." You have to remember your setup vs theirs.
It's a great camera, have one here too. Very handy and works REALLY good for what it is!
I’m a bit surprise you guys don’t have any bed insulation below the heater. That can help and even save a bit of money on the ole power bill. At work I have two Modix 120X 3D printers. We replaced the bed with a 1 inch thick MIC6 plate for amazing flatness. I let that heat up for an hour before printing and check it with the FLIR camera.
I'm really glad you posted this. I got a Gulf Coast robotics bed to replace the heatbed on my Ender 3 S1 Pro (which shorted out). Noticed I had adhesion issues while using the same temperatures while printing PLA on the new bed, so I got an IR thermometer and discovered the same phenomenon that you've pointed out here. I assumed that the thermestor that came with the new bed was bad... seems that's normal based on what you're showing me. 🤨
On my Ender 3 Max, the bed has a pretty significant droop in the center. After 15 minutes of pre-heating at 60C, the droop is gone and I have a nice flat bed, corner to corner. At the same time, the nozzle and heater block is expanding and heading down towards the bed. So, for the first 15 minutes of heating, I've got a bed expanding upwards and a nozzle expanding downwards. After 15 minutes there are no measurable changes in bed or nozzle, even after hours of soaking.
This is interesting. I hadnt even considered the expansion factor in this case. It makes sense which is why you want to do a mesh at temps but someone else mentioned frame expanding and in your case the hotend itself.
On my old Replicator the build plate was so small there was about a 1-2c delta. Still, I put a wool rim around the bed and an air gapped film of Kapton tape. That allows the heat to move around and even out under the bed 😉
After i did that the thin little bed never warped after and i could print ABS to the edges 🥰 i used to have to bend it back every few months 😅
I've noticed on my cr6 that I'll sometimes have issues in the first layer, so I'll cancel print and restart right away, cleaning off the material while the machine does its startup shenanigans.
And it'll usually do much better the second time, I was suspecting it was the bed surface temp and I try to remember to heat up the bed and nozzle manually as I'm preparing a print. I've seen a lot better results for the early layers this way. It is something I'll probably integrate into the firmware once I replace the factory setup.
I had a problem regarding nozzle temperature that was ultimately solved by changing the thermistor type. After a certain point the differential between reported and externally measured temperature grew too large and visible artifacts and extruder clicks followed.
Interesting. The Prusa XL does a heat soak during warmup before starting the print for about 8 minutes.
It seems like this temperature differential (between the print surface and heating pad thermister) are accounted for in filament manufacturer temp recommendations. If you set your temp high enough so that the print surface is at the recommended temperature, you will see every indicator that your bed temp is way too high. So...
It would be cool if you could revisit this and try printing with an actual measured bed temp that falls withing the 'recommended' bed temp for a given filament - my guess is the print would be pretty horrible with many sings that you're printing with just way to high of a bed temp.
Seems like we do need chamber heating to get better control of temperature. Of course it needs to be done safely, and it doesn't help that some groups don't even allow discussing it.
Some groups wont discuss chamber heating? That seems odd to me.
@@ModBotArmy active chamber heating, eg adding active heaters to vorons. They discourage it for safety.
Daniel: Nice effort! A couple of suggestions: Just as a reality check, it would be useful to run your thermocouple bed tests with a layer of insulation covering the top of the bed (ie: also covering the thermocouples.) That greatly reduces dissipation from the top of the bed, greatly reducing the heat flow through the bed (bottom to top), and thus also the temperature drop bottom to top. The result should be that the bed heats uniformly throughout its thickness, and hence your thermocouples should read the same temperature as the set point. This will validate your thermocouples and instrument and also .checks that the printer's thermocouple or thermistor is sensible. Your tests in this video using thermocouples only on the top surface don't actually verify that the printer's sensor is getting up to the set temperature, even if the printer thinks so.
Next, in the comments there are suggestions to attach the thermocouples with thicker (heat) insulating tape, on the principle that then the thermocouples won't be cooled from above. However I think that's a mistake: If you place insulating tape on top of a thermocouple that will reduce dissipation from that neighborhood of the bed, creating a local hot spot, so your measurement will not be representative of the rest of the surface. Instead, the tape you are currently using exposes the top of the thermocouple and local bed to approximately the same air conditions as the rest of the bed, so the temperature is representative.
Anyhow, nice to see folks performing experiments like this and trying to get at some ground truth!
Great info! That's a much bigger temp difference than I expected.
I get much better bed mesh results with an inductive (EZ ABL) sensor after giving it a 3m heat soak in my start gcode.
Precision printer parts sells a bed with thermistor mounted on top right under the print surface.
Did you try attaching the external sensors right next to the printer's sensor, just to confirm it's not a sensor calibration issue?
Im curious on how much the magnet is influencing the temperature reading as it is the least conductive material in the assembly. That could really justify the added cost of a bed with integrated magnets like the prusa mk3 has
you forgot to measure temperature under bed, to check if printer thermistor reading was correct at all or not
I generally presoak when printing large parts or on the v0.1 with its thick build plate.. Now let me see if I change my mind on this at the end of the video. :)
And yup.. I will keep doing what I do. :)
I have found with klipper their thermistor settings are a bit hit and miss. You need to do some tweaking to get things right. Would be interesting to see if you done the same with a marlin machine, I have read they are more accurate.
Could you perform a test with the printer upside down? I think you would see a lower difference between the two sides.
Putting the control thermistor on the opposite side (of the part surface), and without the build surface between the aluminum and the part, will always give you a lower temperature. Putting the control thermistor on the top make would provide better control. Unfortunately it would interfere with the build area. I wonder what it would be like if the thermistor were placed at a corner and the control temperature dropped by a couple of degrees. I think this would give a closer target temperature under the build area.
A further question is, how does a part over the build area affect the temperature. But at that point I would expect that it would be less of a concern after a few layers hade been completed.
There is a big problem with the methodology here which put in question the remarks about top-bottom plate temperature. Thermistors have a much faster response to temperature changes when compared to thermocouples. Also, regardless of the probing type, if you use more than one metrology equipment, it is required to have them calibrated, otherwise risking a few degrees of error between them, up to tens of degrees if using uncalibrated IR meters to probe metal surfaces.
I would suggest using the same 4x thermocouple meter to measure top and bottom.
Aluminum is quick to heat, but it's also quick to dissipate heat. I feel like insulating the bed would go a long way to improve this issue, but would add weight, complexity and still not be an end all solution.
This video is remarkable. Very eye opening
love to see if insulating the bed heater changes the variance.
Hello, I am generally learning a lot from your videos but I see some hickups in this one. Temperature measurements is an area filled with many traps. For one, your thermocouples basically have one or two tiny point of contact with the PEI, and have an hugely bigger area of contact with the kapton tape which are flexible and coated with glue. So the thermocouples are in the middle of a thermal divider, the ground being the ambiant temp, the supply voltage being 100c, the top thermal resistant being a few points and a little bit of air, the bottom thermal resistance being a few square mm of kapton in series with air convection thermal resistance. Bare thermocouple like that are not good at measuring surface temperature. Thermal imagery is the final boss in term of temperature measurement. Many people do not understand the tremendous impact of emissivity, background temperature, specular and diffuse lambertian or non lambertian reflexion (i think the last one are the causes some inhomogeneity you saw)
You might be right. I chose Kapton because it is a thermal insulator and prior to embedded bed thermistors most printers I tested out had them attached using Kapton tape as well. Maybe this being a good choice was an incorrect assumption on my part. Any suggestions on a better insulator for the thermocouple?
Would love to see a video comparing popular production printers to see the difference. I was definitely not expecting so much difference in temperature. Given these results I would expect there to be a significant difference between different production printers too. Great video for provoking some head scratching.
How about adding the Sovol family into this?
@ModBot This might all be fine and correct, but I wonder if there was a significant flaw in your method. Attaching a thermistor with tape to the top of the bed exposes the bulb on the end to a sector maybe 60° or so exposed to the bed surface while the remaining 300° is exposed to the surrounding air & thin tape. My guess is the thermistor is averaging the reading down. I would like to have seen if a tiny amount of insulation between the thermistor tip and the tape, like the size of a q-tip or something, might give you a more accurate measurement of the actual bed temperature instead of a kind of average of the bed and air. I mean, you saw with a camera how much cooler just the tape was than the bed and that whole surface was literally attached 100% to the bed - the thermistor has even less contact with the flat bed surface.
Of course extruded plastic will exhibit similar cooling as it's built up, but at least it would be good to know for sure the temperature of the bed surface and I'm not positive that measurement method was accurate.
Again, I might be wrong. And thanks for the video. It's still good info and illustrates that the bed is constantly losing heat at the surface and makes sense to keep that in mind.
You might be right. I chose Kapton because it is a thermal insulator and prior to embedded bed thermistors most printers I tested out had them attached using Kapton tape as well. Maybe that was an incorrect assumption on my part. I took the outer surface of the Kapton being cooler being due to its insulating properties but I will make no claim that is definitive. Any suggestions on an insulator?
@@ModBotArmy D'oh. I didn't consider that the tape looked cooler because it might be insulating the heat some off the top. Slap forehead. Obvious now at least as a possibility. But in case you want to try a quick experiment to see if it makes any difference, I was thinking just a small piece of that insulating paper stuff we used to wrap hotends back in the day. Just taping a little of it over the top (not surrounding it of course) might give an idea whether it's a thing or not. I know kapton is good for witstanding heat, but I never thought of it as an insulator, so it might be worth a shot.
put 2 layers of corc ontop of the thermo couples. you are going to get a split reading between the bed and the air.
I would have loved to see you measure the underside of the bed with a separate thermocouple so you could compare the printer reading to you measurement setup.
I drilled a hole in the side of my aluminum bed and put a TC in it, and it runs like 5c less and trails like 10 to 15c less when heating up.
I was surprised at your results. I had never really thought much about this, but could solve a lot of problems with higher temp warp prone material
You just confirming data from time when beds were heated by bunch of power resistors and frames were made from all thread or mdf boards. You NEED isolation layer under the bed. Its not optional if you want higher temp.
Id be interested in knowing how open builds cool down by putting the sensors a little above the bed to see how much a difference there is and the cool speed and stuff
I always preheat for oddly enough PLA. My bed is set to 75c in order to get the surface to reach 60c consistently across the entire surface.
I was kind of hoping you'd come to the conclusion you did. I've always wondered just how strong the difference was, and I'm especially curious what you'd find with the Nevermore and/or bed fans running as well, to see if those help or hinder the balancing of temperatures on the build plate.
Great video sir! It would be interesting to see what moving the bed thermistor to the top side of the bed would make to controlling/realizing actual commanded temps. I would think that this would drive the top bed surface temp to be actual commanded temp where it matters?
I completely agree. There are beds with thermistor slots on the top. I think I will need to source one.
very interested to see longer soak times and the effects on bed tempps. Also, curious to see the delta between thick aluminum beds and the mk52 with respect to delta between thermistor and measured temp at high temps
Thin beds usually have a 2-3c delta, glass is about 5c. I have a specific print profile that has a 5 minute heat soak time after the bed reaches temp to then start heating the extruder.
I also picked up a small 60w heater and temp controller years ago that I was going to use. But that has seemed to work well enough that I haven't set it up 😅
*i do a lot of ABS printing, small and large (285x285 and 315mm high)
So you are comparing your probes reading with the printer beds built-in reading ?
If you want any relevant measurement, you should have had the same type probe on top and bottom, so you could also compare it to
how the printer read temperature - I am pretty sure that is not calibrated !?
Hi Daniel, I know in Voronland they recommend limiting heating rate to 75 - 80% of heater power, curious if changing or removing this limit impacts the result of this testing? Cheers
So ,just like a Revo, at 10-15c to get the temps we actually want ?
Should have added one underneath to since bed don’t use same thermistor as the tool you are using.
My $0.02: My Ender 3 with a bed sensor would report varying bed heights while heating for about five minutes. After that, the bed would cease warping and expanding measurably.
Mine behaves similarly. Always have to run the bed heater before probing or printing if I want to avoid headaches
I would not have expected that much of a gap. Could improvements in print quality / bed adhesion be had by increasing the setpoint such that the surface was the desired setpoint? I was also considering the X1 Carbon.... I'd be curious how that unit would fare in a similar test. I've been really impressed by their engineering and wonder if they've already taken that into account. Thanks again for the content... much appreciated.
Is there a Gcode to add on start up code to wait a few minutes?
You should have put the sensors under the build plate and let the magnet hold them tight.
I feel like it would have skewed the results though. Although I am sure the powder coated plate transfers heat well compared to the aluminum or magnet sheet I still wanted it on the very top.
@@ModBotArmy Debatable. ;)
Most printers are firmware limited to ~100-110C. Are the silicone heaters able to go higher or will they melt down? It would be interesting to see how much higher your set point has to be to get the 100C surface temp.
Does this mean that ABS prints well at 85C bed temp abs we’ve all just assumed it is 100, is it worth chasing a 100C surface temp?
I personally print at 95 with Hatchbox abs. But if my bed is actually 70 C, that’s really interesting.
When I set my bed to 100 it’s likely at the glass transition temp because the part deforms when I take it off early. Maybe there’s something mixed in that drops the temp/makes it print even more reliably?
Really interesting question for how many printers would show the same/similar results
Part 2 with bed insulation?
That's why i print ASA at 120ºC in my voron v0.
Hi Danie, I am Lydia from LOTMAXX.
stick your printer on its side to see whether hot air rising is causing the heat to pool under the hotplate?
Were the bed fans on during testing?
None of the printers have bed fans. There is a Nevermore on the 2.4 that could be used as one but in this case no I just tested heat coming off of bed with the only real variable being heater types, bed thicknesses, and enclosed vs not.
@@ModBotArmy I mentioned it when I saw the Nevermore on the 2.4 I'd love to see how that effects the testing.
What I find most irritating is that the final bed temp is nowhere near the setpoint. My profiles are a lie! We're probably all printing way colder than we think.
I usually preheat my X1C until the chamber temp shows 40°C or so. It would be interesting to test production machines (like Bambu printers, wink wink, nudge nudge).
Good info for the video, but poor title. Betteridge's law of headlines is an adage that states: "Any headline that ends in a question mark can be answered by the word no."
Yup, I found on my Ender 3 it worked much better when I let the bed soak for about 10 minutes.
So the thing is that basically ABS needa 85 degrees ans thays why we set the temperature to 100 😅
Much larger difference than i thought. I thinit would have been worth an extra hour of your time to have let it soak for the hour. Just for the science if it.
Maybe I can do it on my own at some point just to see. I may need to revisit with thermistor on top of bed and a different insulator.
Would be great if you try to understand what temperature you have to set in firmware to get real 100 degrees on top. The reason why top temp capped to 85 - probably PID control, when firmware think in reached 100 - it reduces amount of heating a lot
I imagine part of it is PID and that printer lowering heat. I still would have guessed that the heat would have eventually evened out or at least closed its range. One concern someone brought up is how hot the silicon heaters can reach before its an issue. If there is a 15C spread and you are printing at 115 on the bottom to get 100 on top other factors such as adhesive on the bed will need to be considered.
wow dude... this took you long.
¬¬
°
welp, you solved my warping issue... sheesh..
you didn't put a thermocouple on the bottom, relying on just the temp read by the printer's thermistor. of course it'll off! duh.
Wait. People havent been doing this?
Yeah? So what?
Your results are very similar to my own experience with my Voron 2.4r2 that has the same honeybadger heater.
When discussing the issue on Discord, I was told that the variation between the top and bottom thermistor readings was a function of the fact that the bottom thermistor is directly touching the heater, and the kapton tape was not a very effective thermal bond to couple the thermistor with the flex plate (which is not well bonded with the bed plate beneath it), all of which explains a large portion of the variance even after soaking.
The IR measurements a not very reliable due to the reflective nature of the PEI plate, so a mat black surface would allow for more accurate IR camera readings.
The issue with this video, is that you never explained what to do about the variation.
Nobody seems to know how to best quantify the print surface temperature compared to the bed sensor reading so you can deterministically adjust your filament temperature slicer settings.
Great test.
This is the reason for Prusa having the resistive heater just under the springsteel. It can give some weare, when you remove the springsteel thou.
Thomas Salandere have made a temperature video, about the unifomity if heat spread of a headbed. ua-cam.com/video/yPy0zDpmc5U/v-deo.html
Thanks for sharing your experience with All of us 👍😀
Aluminum is an extremely efficient heat-sink so it would make sense that on the opposite side of the plate from the heater it would be cooler, the aluminum is doing what it does really well, dissipating heat.
Should You Be Preheating Your 3d Printer? - No , mine works fine without .
Might not be the case with all materials. There are a lot of very sensitive filaments out there that will warp not matter what if perameters are not in a very tight range.
Hello and a good day toooo all the people out there! Is there somebody who have a FIBTIP Nozzle, because they promised it's the last you need! Indestructable, non-stick technology
With fibtip's unique ceramic diffusion coating you will get unrivaled and everlasting performance against even the most abrasive filaments (copied from there side). Some tried the Nozzle or hey a Video Idea 💡!?! I wish all the best to all and enjoy your day wherever you are😁😁😁👍👍‼️‼️