More powerful heated bed?

Hi,

I have 3 machines that use the 200x200 MK2a heated bed. I use 1/8" borosilicate as the build surface. The bed is powered by ramps 1.4/arduino 2560 combo running marlin.

Is there anything I can do to:

1. Speed up the heating process (I know this directly requires more wattage input into the glass)
2. Improve the temperature distribution of the bed

Basically, I get prints that curl at the edges of the bed. I have tried squashing the bed closer to the heater. I was thinking of smearing thermal paste on the bed/glass interface. That seems very messy though.

I dont want to use aluminum because I need the bottom surface of the print to have a nice finish and I am afraid aluminum will cause a grainy rough bottom surface.

Is there any way I can modify the firmware somehow to boost power to the existing bed? Or is it just moot at this point and I should replace the bed with something else?

Thank you

You could use thinner glass. Mine is 1.6mm ( approx. half the thickness of yours ) It is from cheap picture frames.
I also took the MDF-backplate and clamped it underneath the heatbed. Isolation helps alot!

Build a hull around the printer to avoid a fresh breeze cooling down the edges.

Finally, you could raise the voltage ( upto 16V if you have a 16V polyfuse ) to speed up the heating. ( not recommended )
-Olaf

there are two kinds of solutions here:

1) Increase wattage to the bed

2) Decrease losses.

The cheapest ways to achieve 1) are: thicker wires and a better MOSFET with a small heatsink.

The cheapest ways to achieve 2) are: a 2mm cork sheet glued to the bottom of the MK2 heatbed and make sure the MK2 heatbed is installed correctly, i.e. with the traces up.

I was reading somewhere that installing the MK2 upside down puts down more heat. Is that true?

The thin glass from picture frames, is that stuff going to break when heated and cooled? That glass is not tempered.

The machine is encased in a box and the electronics have a dedicated cooling fan, so it has plenty of power and no cold breezes.

The bottom of the bed currently has fiber insulation. The sides have wood insulation and the glass is clamped down on it. I dont see how cold air could be getting between the glass and MK2.

Not sure what else to do, I am positive that the heat distirbution is screwing up my prints. I can put my hand on the bed 1" away from the edge when the thermistor is reading 120C... Thats a 50-60C differential!!!! This is not right!

I need to prevent that. Should I install a thin layer of aluminum on my heater and then the glass on top of the aluminum? Kind of like how pots or copper on the outside but have an aluminum core, helps distribute the heat. What do you think?

I use 1.5mm thick aluminium between the bed heater and the glass plate on my Cartesian printer with a 200x210mm bed, although 2mm aluminium would be better. Another advantage of spreading the heat in this way is that you can use ordinary float glass instead of borosilicate.

Also measure the voltage on the bed heater terminals and compare it with the voltage going into the RAMPS. If there is more than about 0.3V difference, find out where the biggest voltage drop is (i.e. cable, mosfet or polyfuse) and do something to reduce it.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

If you don't want heat distributed evenly, use glass.
If you want it to take longer to heat up, use glass.

If you want heat distributed evenly, use aluminum. Put a layer of Kapton tape on it. The bottom surface of prints will be glass smooth.
If you want it to heat up faster, putting a thermal insulating layer - i.e. glass- between the print and the heater is NOT going to make it heat up faster. Use aluminum.

That's how it works until someone comes up with thermally conductive glass.

Use two power supplies and a relay, one for your ramps 1.4 and everything but the heat bed and the other for your heat bed. I get to 105c in just a few minutes. I use 1/4 inch glass, it heats up evenly enough that I have no curling when using hairspray and using the raft technique on larger parts. Get some Ceramic Fiber Heat Insulation Blanket to reduce heat loss from under your heat bed. This works for me very well, obviously everyone has their favorite things to use. Digital Dentist's way works too, probably better than mine but in case you have most of the things I use anyway and if you are worried about thick glass, don't be just get more power to that heat bed.

I have Ramps 1.4/Arduino 2560 combo like yours. I went from a Mk 2b bed with 6mm glass on 12v, which was slow, to a Mk 3 aluminium bed. It was even slower, until I fitted a 24v PSU just for the bed, as lunarkingdom.

It is controlled separately via a SSR. I reduced the 24v voltage down to give 15 amps through the bed to keep it within the PSU rating.

Situation was suddenly transformed! It heats from ambient to 110 deg C in about 2 to 3 minutes. I can start printing now within 5 mins of turning the printer on.

I agree with the digital dentist. I print on Kapton tape on top of the aluminium. I also only print ABS.

Quote
VTE
I was reading somewhere that installing the MK2 upside down puts down more heat. Is that true?

No. That on the other hand explains why you are having such poor heat transfer between the heatbed and the glass. You do have your heatbed upside down, don't you?

Quote
VTE
The thin glass from picture frames, is that stuff going to break when heated and cooled? That glass is not tempered.

Float glass is much more likely to break from thermal shock than borosilicate glass. Use borosilicate glass, 3mm is a common thickness and works well.

Quote
VTE
...
Not sure what else to do, I am positive that the heat distirbution is screwing up my prints. I can put my hand on the bed 1" away from the edge when the thermistor is reading 120C... Thats a 50-60C differential!!!! This is not right!
...

Fix your heatbed orientation first, then come back with your new measurements. Also make sure the heatbed thermistor measures the temperature of the borosilicate glass (heatsink thermal compound is your friend...). Oh yes, one last thing: I would not try to measure temperatures with the palm of my hand, bad idea. A $15 infrared thermometer is recommended.

Quote
AndrewBCN

Quote
VTE
I was reading somewhere that installing the MK2 upside down puts down more heat. Is that true?

No. That on the other hand explains why you are having such poor heat transfer between the heatbed and the glass. You do have your heatbed upside down, don't you?

Quote
VTE
The thin glass from picture frames, is that stuff going to break when heated and cooled? That glass is not tempered.

Float glass is much more likely to break from thermal shock than borosilicate glass. Use borosilicate glass, 3mm is a common thickness and works well.

Quote
VTE
...
Not sure what else to do, I am positive that the heat distirbution is screwing up my prints. I can put my hand on the bed 1" away from the edge when the thermistor is reading 120C... Thats a 50-60C differential!!!! This is not right!
...

Fix your heatbed orientation first, then come back with your new measurements. Also make sure the heatbed thermistor measures the temperature of the borosilicate glass (heatsink thermal compound is your friend...). Oh yes, one last thing: I would not try to measure temperatures with the palm of my hand, bad idea. A $15 infrared thermometer is recommended.

Hi, I said I was just reading somewhere, I never run it that way.

Here is where I read it.

[reprap.org]

Scroll down to where it says "Sides" and you will see it.

I am already using 1/8" borosilicate as was mentioned in original post.



Lunarkingdom, which power supply should I get for the heated bed? Can I run 24V through the MK2a or must I get the MK2b?

Would I just wire up the relay into the ramps green terminal block, so it flicks the relay on to the power supply when it gets signal to do so?

I need to compile a list of parts to get to make this happen.

1. 24V PSU - how much current we talking? 10-15 amps?

[www.amazon.com]


2. Relay - i am thinking any automotive relay would work, any other suggestions on this part?
3. MK2b heated bed @ 24v
4. I already have 1/4" borosilicate and fiber insulation backing it. That part is done.

Any other suggestions are greatly appreciated and I think you all for the help.

I forgot to ask, where is a good supplier of milled aluminum plate?

Thank you,

Edited 1 time(s). Last edit at 05/23/2015 08:36PM by VTE.

Quote
VTE
...
Hi, I said I was just reading somewhere, I never run it that way.

Here is where I read it.

[reprap.org]

The wiki sometimes contains information written three or more years ago and that has not been updated. Thanks for pointing out that bit of information which I thought was not very clear. I have updated it, hopefully it will not lead to such confusion anymore.

Just to clarify this again: the MK2A and MK2B PCB heatbeds, when used together with a piece of borosilicate glass as recommended nowadays, should be mounted with the side with copper traces UP (i.e. in contact with the glass).

Quote
AndrewBCN

Quote
VTE
...
Hi, I said I was just reading somewhere, I never run it that way.

Here is where I read it.

[reprap.org]

The wiki sometimes contains information written three or more years ago and that has not been updated. Thanks for pointing out that bit of information which I thought was not very clear. I have updated it, hopefully it will not lead to such confusion anymore.

Just to clarify this again: the MK2A and MK2B PCB heatbeds, when used together with a piece of borosilicate glass as recommended nowadays, should be mounted with the side with copper traces UP (i.e. in contact with the glass).

Hi, thank you for the quick response.

I am mounting them up. Here is the issue that I noticed, the PCB is not perfectly flat, the middle is bowed up and the corners are about 0.5mm away from the glass, this is where the problems seem to stem from. Poor conduction due to little or no contact. I was thinking bonding, squashing, gluing, or somehow fixing the heater to the glass. I feel like that is the only way I can get the even distribution of temperature.

I dont even think the 24V PSU/relay method will help distribution, it seems that it will help warm up time but do little to nothing for distribution. Distribution has more to do with the assembly, how the parts allow heat transfer between them etc... Not the amount of power input.

I am getting desperate at this point. I started spec'ing out 24V PSU's, milled aluminum beds, thermal pastes, fastening mechanisms and all kinds of other ways to improve the even temp distribution.

I think that all that may not be necessary if I can get just find the one or two things that are really keeping the distribution from propogating throughout the glass.

The MK2a cant be that bad? I am certain when affixed properly it should be an even heat, most of the time, problems like this are user error. Throwing money at it wont solve anything.

Should I try to move forward with the MK2a and standard boro glass and just work out the basic kinks like bed/warmer contact area and insulation etc...??

Thank you

Edited 1 time(s). Last edit at 05/23/2015 10:47PM by VTE.

Quote
VTE

Hi, thank you for the quick response.

I am mounting them up. Here is the issue that I noticed, the PCB is not perfectly flat, the middle is bowed up and the corners are about 0.5mm away from the glass, this is where the problems seem to stem from. Poor conduction due to little or no contact. I was thinking bonding, squashing, gluing, or somehow fixing the heater to the glass. I feel like that is the only way I can get the even distribution of temperature.
...

I agree with you that proper heat transfer between the PCB heatbed and the borosilicate glass depends essentially on good contact. Air as we know is an excellent insulator so any gap between the PCB and the glass will act as a heat barrier. I use 6 or more bulldog clips to insure a good contact between the heatbed and the glass. However, I haven't really done a lot of testing with printing big objects, the largest things I print are usually 10cm wide at most, and I usually print them around the center of the bed.

When using a 12V power supply to heat up an MK2A heatbed it is essential to verify the resistance of the heatbed with a good multimeter. You really want a resistance no higher than 1.1Ohm with the heatbed cold. With that kind of resistance my heatbeds usually reach 65C within 5-8 minutes and 110C within 15-18 minutes; I consider these times acceptable.

Obviously if you feed the heatbed more power it heats up faster, but I am OK with the approximately 120W I am feeding mine.

If you´ve told us about the airgap from the beginning, this discussion would´t have been alot shorter, I guess... ( and less informative )

If I were in your shoes, I´d get a nice MK3 heatbed.
It is flat and has an aluminum side to print on or you could use kapton or the glass plate and bottom insulation you already have.

It also puzzles me, why we talk about borosilicate glass vs. float glass and argue with "radical temperature differences"? At the same time we discuss why the hole process is sooo slow.
Ofcourse I´ve heard of people putting their finished prints in the freezer to get it off faster, but that´s not common?

I´ll turn around my MK3 and use the blue painter tape for PLA to print this huge LCD-cover again, that tends to warp before. ( reducing the bed temp from 75° to 40°C helped alot, but not completely)
-Olaf

Quote
AndrewBCN

Quote
VTE

Hi, thank you for the quick response.

I am mounting them up. Here is the issue that I noticed, the PCB is not perfectly flat, the middle is bowed up and the corners are about 0.5mm away from the glass, this is where the problems seem to stem from. Poor conduction due to little or no contact. I was thinking bonding, squashing, gluing, or somehow fixing the heater to the glass. I feel like that is the only way I can get the even distribution of temperature.
...

I agree with you that proper heat transfer between the PCB heatbed and the borosilicate glass depends essentially on good contact. Air as we know is an excellent insulator so any gap between the PCB and the glass will act as a heat barrier. I use 6 or more bulldog clips to insure a good contact between the heatbed and the glass. However, I haven't really done a lot of testing with printing big objects, the largest things I print are usually 10cm wide at most, and I usually print them around the center of the bed.

When using a 12V power supply to heat up an MK2A heatbed it is essential to verify the resistance of the heatbed with a good multimeter. You really want a resistance no higher than 1.1Ohm with the heatbed cold. With that kind of resistance my heatbeds usually reach 65C within 5-8 minutes and 110C within 15-18 minutes; I consider these times acceptable.

Obviously if you feed the heatbed more power it heats up faster, but I am OK with the approximately 120W I am feeding mine.

6 Clips? Wont those things warp the boro glass? I use 2 clips because I am afraid I will bend the glass all out of shape if I clamp it down too hard. It takes about 15-20 min for me to hit 120C which is my ideal operating temp, I like a really hot bed because I print mainly ABS prints for 9 hours or more and sometimes I print very tall parts with low aspect ratios so they can topple if they dont stick to the bed. I use a very wide brim too, but the extra bed heat is necessary for my application.

I just wish I can have that same heat thats in the middle of the bed, out near the corner so I can print 9 or 10 very tall objects all on the same bed without worrying they will fly off after 5 hours into a 9 hour print.

Borosilicate glass is much harder than float glass and does not bend anywhere as much. Compared to the relatively flexible PCB, using 6 or more bulldog clips forces the PCB against the glass and not the other way around.

Why not give it a try in your case?

Quote
AndrewBCN
Borosilicate glass is much harder than float glass and does not bend anywhere as much.

According to [www.glassarmonica.com] it's the other way round - borosilocate will bend slightly more than soda lime glass (it has a lower Young's modulus). But the difference is small.

Even 3mm glass is hard to bend appreciably, but if in doubt you can use thicker glass. Stiffness increases as the cube of the thickness. I use 4mm float glass (on top of aluminium) on my 330mm diameter delta. I don't recommend float glass directly on a heater PCB because of the uneven heating; but it is fine on top of aluminium. I have been using float glass for 16 months without any problems. Just don't do anything silly like take it off the heated bed at 100C and run it under the cold tap straight away.

One disadvantage of glass is that because of its low thermal conductivity, it is difficult to get reliable temperature measurements on top of it.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42

One disadvantage of glass is that because of its low thermal conductivity, it is difficult to get reliable temperature measurements on top of it.

That´s right, but it also has low thermal capacity, that´s why it gets hot pretty quick and cold as well. ( compared to metal or wood)
There is a good reason, why we use silicon pads or grease to connect semiconductors with heatsinks... ( glass is almost the same)
-Olaf