Be aware - Actual buildplate temp lower than temp setting

Topics around mechanical design, controller and electronics. Mods & hacks welcome.
EldRick
Posts: 264
Joined: Thu Jun 22, 2017 5:55 am

Be aware - Actual buildplate temp lower than temp setting

Postby EldRick » Thu Jan 11, 2018 9:52 pm

Be aware that while the heated aluminum plate under the buildplate glass may reach the temperature you set, the actual temperature of the surface on which you print will be substantially lower.

I set the buildplate temperature to 100C with the cover on the machine, and waited an hour for temps to completely stabilize. I pointed my pyrometer (which I believe to be accurate to around +/-3C) at the glass buildplate with a good coat of gluestick on it. The temperature read within +/-1C, at various locations on the plate, except near the Home position, where the printhead fan apparently cools the area by several degrees. So I lowered the bed a couple of inches and waited another half-hour before measuring.

Unfortunately, the reading was 87C, thirteen degrees lower than the 100C I had set. When I used a plate with Buildtak on it, the temp only read 84C, as the Buildtak material insulates the heater even more.

So the message is, if you are having bed adhesion problems, it is very possibly because you don't have the buildplate as hot as you think you set. I found that when I set the buildplate temp to 106C using gluestick instead of the recommended 100C, I got substantially better print adhesion with ABS and PC without any problems with removal when cooled.

I'll use this information in a couple of days, when I receive the Ultem I ordered. I'll also modify the end-of-print gcode to lower the bed a couple of inches after printing, for more even bed temps on the next print.

BTW, here is a good source for Ultem - better than Mcmaster.com:
http://catalog.cshyde.com/viewitems/3d- ... /ultem-pei
Their Ultem comes with 3M adhesive pre-applied, and you can request both precut exact size (331mm X 339mm for the N2) and whether the shiny or matte side is up, in addition to thickness. I chose .020" (=.5mm) because I hope it will be easier to apply to the glass without bubbles vs. a thinner sheet.

User avatar
Casale8
Posts: 178
Joined: Fri Oct 20, 2017 5:17 pm
Location: The Real Upstate New York

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Casale8 » Fri Jan 12, 2018 4:18 pm

Add materials between the source of heat and there will be dissipation. An engineering given.

I have purchased a bunch of temperature indicating labels for the platform. I place them on the buildtak in various locations during print to observe the difference as I need. McM: https://www.mcmaster.com/#stickers/=1b3guoj

I also took a gauge and noted that the bed varied at random locations ±.3° max, so I know that the heating is fine in my case.
-SCC
“One man's "magic" is another man's engineering. "Supernatural" is a null word.” -R.A.Heinlein

EldRick
Posts: 264
Joined: Thu Jun 22, 2017 5:55 am

Re: Be aware - Actual buildplate temp lower than temp setting

Postby EldRick » Fri Jan 12, 2018 4:46 pm

Once it stabilized, heating on mine was very even, but quite low compared to the set-point.
Clearly it needs correction, and I'm glad I took the time to investigate.

Vice Chief
Posts: 63
Joined: Wed Aug 23, 2017 5:04 am
Contact:

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Vice Chief » Tue Jan 16, 2018 2:29 am

I saw the same thing with a piece of glass painted black, measured with a thermal camera, Casel8 is right, I set my bed to 108C to hit a desired temp of 100C. There is a +/- 5 deg C variation across the bed.

zemlin
Posts: 294
Joined: Sat Oct 21, 2017 2:02 pm

Re: Be aware - Actual buildplate temp lower than temp setting

Postby zemlin » Tue Jan 16, 2018 4:23 am

I get a good ABS print with my bed temp set to 105C. Does it matter that the surface may actually be 100C?

firesped
Posts: 836
Joined: Mon Mar 21, 2016 9:23 pm
Location: Tucson, AZ

Re: Be aware - Actual buildplate temp lower than temp setting

Postby firesped » Tue Jan 16, 2018 6:41 am

bed adhesion is far more complicated then just reaching a target temperature. your surface needs to be level or the printer has to have sensors to compensate for it not being level. When you heat metal, it bends and warps. the more you heat it, the more it bends and warps. using clips to keep the glass in place will cause the glass to also bend and warp with the heated bed.

It should also be pointed out that the borosilicate glass build plate is a high heat resistant material. so the glass build plate is not really intended to actually hit 100C itself. This is also why letting it have time to actually get to temperature helps with printing.

Raise3D attempts to keep the bed level are by installing a massive number of screws that have counter screws to keep it in place. This is a very time consuming process to re-level after it goes out of level from the factory install.

some people have switched to 4 point leveling. I think one owner has put in a new 3 point heated bed. personally I switched to a 4 point leveling system. I recently switched over to wing nuts versus using thumb screws. I was not able to turn the thumb screws without releasing tension on the bed. the wing nut for this bed actually let me make controlled movements to level it. The technique that I learned to level it works. 100C was sufficient for me to print ePC with no problems at all.

Stating that there is a major issue with the heated build plate heating element on all Raise3D printer is not really valid. If you thing you have an issue with your printer, you should contact Raise3D tech support and work out the problem.
RL name: Michael Nolen
printers:
raise3D N2 kickstarter Early Bird
Trinus Deluxe (running smoothieware on Azteeg X5 GT board)
Monoprice Maker Select v2

EldRick
Posts: 264
Joined: Thu Jun 22, 2017 5:55 am

Re: Be aware - Actual buildplate temp lower than temp setting

Postby EldRick » Tue Jan 16, 2018 4:24 pm

Stating that there is a major issue with the heated build plate heating element on all Raise3D printer is not really valid.

Did someone say that?

I, at least, just said "be aware" that the buildplate heater temp is not equal to buildplate surface temp. So, to get to the usually-recommended 100C buildplate temp for ABS, you will need to set 105-110C.

This does seem to me to be a significant issue with buildplate adhesion, but I'm pretty sure it would afflict any 3D printer with a heated glass plate (including all Raise3D printers).

Squenz
Posts: 140
Joined: Fri Dec 01, 2017 11:18 pm
Location: Stuttgart/Germany

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Squenz » Sun Jan 21, 2018 11:16 am

I have to chime in on the issue. The temparaures i measured on my N1 HBp are 8 to 10 degress Celsius lower than the setting. I got around 100 C on the bare plate, with one high spot with 102. I snapped pics with my Testo FLIR camera with different plates on top of it and heat loss makes things even more difficult to work with. I need it hot, stock standard ist not good enough for me.

Since i'll start printing Nylon, PA12 to be specific, soon, i need more HBP, if the HBP would reach the advertised 110 C wouldn't be a bad start, but i'd like to reach more, say 120C on the bare plate.

I have some ideas to get there, but all of them incorperate things which are undiserable.

I hope someone from from Raise will also chime in and give us answers what can be done to remedy this. I'm not shy to get out allen keys, screwdrivers and soldering iron and certainly have all necessary tool at hand, but since we're talking about a high-power Output of the main board, i'm not sure how to get where i need to without a bit of guidance and someone who knows the hardware inside out and can make use of ohm's law.
Last edited by Squenz on Sun Jan 21, 2018 5:37 pm, edited 2 times in total.

Iamkar33m
Posts: 48
Joined: Fri Dec 01, 2017 2:53 pm

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Iamkar33m » Sun Jan 21, 2018 4:45 pm

I posted this in another thread the other day, I wanted to see the effect of using thermal silicone pads under the heat bed (as to not have to use the binder clips to secure the bed). Here was my finding:

Image

For each table, each cell represents the area of the bed from which I took a 5-second (max hold) reading with my infrared/non-contact thermometer with the top of the table being toward the back of the printer.

1st table (upper left): Removed the stock glass build platform and pre-heated the bed to 100°c (as indicated by the printer's sensor) for 15 minutes.

2nd table (upper right): Replaced the stock glass build platform & secured with the provided clips. Pre-heated the bed to 100°c (as indicated by the printer's sensor) for 15 minutes.

3rd table (bottom center): Removed the stock glass build platform and installed 4x thermal pads on the aluminum bed. Re-installed the stock glass build platform & pre-heated the bed to 100°c (as indicated by the printer's sensor) for 15 minutes.

As you can see, even taking the readings on the bare aluminum bed (without build platform) already shows a difference of 12-21°c off from the printer's internal thermistor.

Squenz
Posts: 140
Joined: Fri Dec 01, 2017 11:18 pm
Location: Stuttgart/Germany

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Squenz » Sun Jan 21, 2018 5:09 pm

Thank you for getting in here.

I read your posting and it was somewhat a reason for me to get into this deeper. My PA12 gives me a hard time in regard of adhesion, and i wanted to get some data to see what's really going on.

Using these thermal pads might be good idea in regard of keeping the bed nice and flat without thermal distortion of the alumnium transferring to the glass plate, but you're producing an airgap inbetween both and air is a terrible thermal conductor. Therefore, the more air you can get outside the gap the better the heat transfer will be. So, your findings didn't surprise me at all. But thank you so much for taking your time to get something togehter and bringing it up.

In regard of your data of cold spots, well, 15 Minutes is not good enough for proper heating of the bare aluminium. Your results show that. If you need every Kelvin in the HBP to get your job done, you need to give the machine at least 30 minutes to heat up. That's not only true for the HBP, but goes also for the bulit chamber. If everything is nicely heated, i can print even the badest warping material without thinking about it, as long as i can get it sticking to the built plate at all.

Fire it up and give the HBP and the chamber a good time to warm up and everything will be fine. I saw people mounting a hair dryer or an electric heating gun to speed things up, not my cup of tea, but different strokes for different folks.

For me, it does not matter much if there are hot or cold spots, in the end you can't beat physics, even with the best design possible. But the HBP even doesn't reach advertised temperature and that's what something needs to be fixed. Very soon.

There are several things which can be the problem even on a machine working 'within specs'. Not enough power of the heating element could be one, the sensor might be different to the specs and the firmware doesn't allow more than the 'measured' 110C beeing part of the matter also.

If the bare plate doesn't reach the designated temperature, there could be an offset parameter in the firmware installed, which would allow the user to adjust the beast. If you say, give me 100 and you measure actual 90, tell the computer the offset is ten Kelvin and be a happy camper. IF the heating element is designed with some room to spare, if the heating element is on 100 percent power all the time and there's still a 10 degree delta, there's not enough power available. That could mean, the resistance of the element is to high, so not enough power is created to reach the set point or the heat transfer between the element and alumnium is not good enough to heat the aluminium while the element is happily cooking himself at high temperature. There also could be a current limiter on the main board which doesn't allow enough power to created. There are also a number of things which would go wrong on the machines.

That's where somebody with a more intimate knowledge would be welcome to give some facts and ideas how to reach advertised temps and a bit more than that.
Last edited by Squenz on Sun Jan 21, 2018 5:35 pm, edited 1 time in total.

EldRick
Posts: 264
Joined: Thu Jun 22, 2017 5:55 am

Re: Be aware - Actual buildplate temp lower than temp setting

Postby EldRick » Sun Jan 21, 2018 5:13 pm

"I get a good ABS print with my bed temp set to 105C. Does it matter that the surface may actually be 100C?"
What matters is if you set it to 100C, thinking that you will get 100C at the surface, and instead get much lower temps, which will ruin adhesion.

Squenz
Posts: 140
Joined: Fri Dec 01, 2017 11:18 pm
Location: Stuttgart/Germany

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Squenz » Sun Jan 21, 2018 5:43 pm

EldRick wrote:"I get a good ABS print with my bed temp set to 105C. Does it matter that the surface may actually be 100C?"
What matters is if you set it to 100C, thinking that you will get 100C at the surface, and instead get much lower temps, which will ruin adhesion.


You're right, actual values don't mean much as long as you reach your goals.

But in my case, true 120C seems to be the normal for the PA12 nylon on a garolite surface and my machine is far away from that, doesn't even reach the temps it should get to.

Squenz
Posts: 140
Joined: Fri Dec 01, 2017 11:18 pm
Location: Stuttgart/Germany

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Squenz » Mon Jan 22, 2018 9:00 am

In addition to the conversation, i monitored the power consumption of the machine last night. The HBP seems to have an ED between 40 and 50 percent at 110C. Meaning, if the HBP is heated up to target temp, it shut's off more than half of the print time.

So, HBP element power shouldn't be a factor in the issue.

Hopefully someone from R3D will come in and give us some ideas.

sunraise
Posts: 11
Joined: Mon Aug 14, 2017 8:01 am

Re: Be aware - Actual buildplate temp lower than temp setting

Postby sunraise » Mon Jan 22, 2018 9:01 pm

I've had this issue too, the difference in temperature was way over 10 deg, without the glass on. It wasn't about the bed not being heated properly, as I took the measurements withing minutes and after about two hours of heating the bed in enclosed cover.

Then it just started to work allright - seriously - i measured it using the FLIR gun. temp variations 1-4 degrees accross the whole plate, with temp 100 on 100c settings.

What did i do ? I yanked the cables a little, just to make sure they are connected properly on both sides. And a power off.

not kidding you, give it a try.

good luck !

User avatar
Julia Truchsess
Posts: 232
Joined: Mon Mar 21, 2016 1:30 pm
Location: Connecticut
Contact:

Re: Be aware - Actual buildplate temp lower than temp setting

Postby Julia Truchsess » Tue Jan 23, 2018 2:33 pm

If you're really serious about measuring bed surface temperatures, I'd suggest you tape thermocouples directly to the surface being measured. I'm not saying your non-contact measurements are wrong, but I'd have a lot more confidence in the data if it was via contact, as emissivity and other factors can significantly affect non-contact measurements.

firesped
Posts: 836
Joined: Mon Mar 21, 2016 9:23 pm
Location: Tucson, AZ

Re: Be aware - Actual buildplate temp lower than temp setting

Postby firesped » Wed Jan 24, 2018 2:19 pm

just to point out this information.

A temperature gun measures the temperature of one spot. The FLIR E40sc thermal imaging camera measures the temperature of 19,200 spots. ... Both spot pyrometers and thermal cameras work according to the same principle; they detect infrared radiation and translate it into a temperature reading.

can we note one thing here.

they detect infrared radiation and translate it into a temperature reading.

That means you are not measuring the heat of the HBP, you are measuring the heat it is radiating as infrared.

What we are looking for is this.

Heat transfer
Heat transfer is the process of heat energy flowing from a source at a high temperature to a load at a lower temperature. The three forms of heat transfer are conduction, convection, and radiation (infrared.) Conduction occurs when there is a transfer of heat energy due to a temperature difference within an object or between objects in direct physical contact. Convection is the result of a transfer of heat energy from one object to another via a moving fluid or gas. Radiation heat transfer can occur by infrared, ultraviolet, microwave and radiowaves. Infrared (electromagnetic radiant infrared energy) is the transfer of heat energy via invisible electromagnetic energy waves that can be felt as the warmth from the sun or a downwind fire or other hot object.

pulled from here. https://www.infraredheaters.com/basic.html

But we are not using Infrared heat transfer. we are using Conduction or a combination of Conduction and Convection (when we use thermal pads correctly with the bed.

When we use thermal pads placed in the 4 corners and cut into triangles. we use the air between the HBP and the Glass bed to heat the glass bed via Convection. but the thermal pads also heat the glass plate via conduction as it is direct contact. The advantage of using thermal pads in this way is it allows the glass plate to float above the HBP, so then the HBP bends, the glass does not. When you clip the glass to the HBP, the glass will bend as the HBP bends. Metal bends when you heat it, this fact is absolutely true. This is why raise3d has put so many retaining screws and counter leveling sinks into the Z Carriage plate to attempt to keep the HBP level.

So if you are using a IR gun, you are measuring the heat that has transferred into the air. realistically if it was showing 110 C, you would have a far bigger problem as your printer is going to start melting.
RL name: Michael Nolen
printers:
raise3D N2 kickstarter Early Bird
Trinus Deluxe (running smoothieware on Azteeg X5 GT board)
Monoprice Maker Select v2


Return to “Hardware”

Who is online

Users browsing this forum: Google [Bot] and 3 guests