glass bed not reaching temperature at the surface

Hi. I have a heat-spreader plate directly under the glass and I know it is reaching 100 degrees because water applied to the exposed edge of the plate will boil.
But the surface of the glass bed is way below that. I pushed the to bed to 130 and I still couldn't boil water on the bed surface.
Is this normal and if so how do others adapt?
thanks

Glass is a thermal insulator and a poor choice for a bed surface. The only reason people use it is because they can't find anything else that is flat enough to print on. Or I should say they don't want to pay for a material that is both flat and thermally conductive such as cast aluminum tooling plate. It's more important to save $5 on the bed plate and slather it with glue, ABS juice or hairspray to make the print stick than it is to just spend an extra $5 and get something that actually works.

I get around 15 to 20C temperature drop across my 4mm glass bed. For ABS printing, I set the indicated bed temperature to 140C, but that's partly because I am using a silicone heater with a thermistor embedded within it, so the thermistor reads a higher temperature than the temperature of the aluminium bed substrate.

It takes a lot of energy to boil water because of the latent heat, and before it gets to boiling point it is evaporating quickly, taking latent heat with it. With the aluminium at 130C, I am fairly sure the top of your glass will be above 100C when you don't have water on 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].

@digital_dentist:
Thanks. I occasionally see a metal bed but they are so rare I haven't given it much thought except:
..'compared to glass, metal is soft and during a nozzle crash will both scratch and catch on the plate; also when I need to scrape something off with a metal blade (which is preferred because it harder than plastic and won't melt on a hot bed'').
..'Most metal expands like crazy, which also means potential warping (of the plate) due to uneven heating and cooling, plus mounting issues, whereas borosilicate glass basically has no expansion'
..'Metal would certainly heat up faster and more evenly than glass, which would be nice'
..'I wonder what the 'sticky' conditions are for PLA and ABS?'
..'Glass does not conduct electricity, which is sometimes a blessing with respect to nearby electronics.'

These are just my main thoughts, plus some some other small ones. Metal balanced-out poorly compared to the glass so I didn't see a reason to pursue it. Let me know please what other thoughts I should have for a fair comparison.

@dc42: thanks for telling me there is no magic formula; I will just turn up the heat then. Your argument makes sense about the latent heat. Another way to look at is that, metal and glass may have similar heat capacities but the insulative properties of glass don't allow the heat to flow into the water fast enough to boil before it evaporates ie the surface of glass drops temperature quickly and recovers poorly. Radiated heat and boundary conditions of the glass wrt to the air would keep the surface of the glass lower in temp that a similar-sized metal plate.
One fallout from this analysis is that a half-printed ABS part would insulate itself under the part but not around it; there would be some temperature gradient around the borders of the printed part with the highest temperatures under the middle of the whatever is being printed. I wonder how much this might contribute to parts curling off the bed? If so then there is a point towards a metal bed.
What I need is a diamond plate!
According to this, it is the surface area vs volume that matters, so I should be able to put a small container of water on the bed and it will eventually boil, if it is insulated well enough and has a relatively small area of exposed water.
BTW, did I notice that you had a silicon-wafer bed? Why choose that?

Edited 1 time(s). Last edit at 06/22/2015 01:56PM by shadowphile.

Quote
shadowphile
BTW, did I notice that you had a silicon-wafer bed? What choose that?

If you are referring to my video, it's nothing as fancy as that! It's 5mm aluminium, painted black with stove/barbecue paint (cured for a few hours at 170C using the bed heater), with 4mm glass on top of that. So the reflection you see is from the top of the glass. Painting the aluminium black makes the IR sensor work better, because otherwise the very strong reflection from the aluminium interferes with the weak reflection from the top of the glass.

Edited 1 time(s). Last edit at 06/22/2015 01:40PM by dc42.

---

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
shadowphile
@digital_dentist:
Thanks. I occasionally see a metal bed but they are so rare I haven't given it much thought except:
..'compared to glass, metal is soft and during a nozzle crash will both scratch and catch on the plate; also when I need to scrape something off with a metal blade (which is preferred because it harder than plastic and won't melt on a hot bed'').
..'Most metal expands like crazy, which also means potential warping (of the plate) due to uneven heating and cooling, plus mounting issues, whereas borosilicate glass basically has no expansion'
..'Metal would certainly heat up faster and more evenly than glass, which would be nice'
..'I wonder what the 'sticky' conditions are for PLA and ABS?'
..'Glass does not conduct electricity, which is sometimes a blessing with respect to nearby electronics.'

If a nozzle crash occurs with an aluminum plate, it may scratch. Scratches are not an issue- you're covering the plate with a layer of kapton tape anyway. If a crash tears the tape and scratches the bed underneath, you replace the tape and you're good to go. What aluminum won't do is shatter into tiny shards that will end up in your bearings, skin, and maybe eyes.

Aluminum tooling plate is cast and expands evenly with temperature, more evenly than extruded plate. It remains flat. The fact that it's thermally conductive means the temperature is relatively even throughout the bulk and surface of the plate.

ABS sticks to kapton tape at about 105C and PLA sticks to painter's tape applied over the kapton tape (assuming you print with ABS and PLA) at a lower temperature. You may have to adjust the Z axis zero point a little when switching from ABS to PLA printing because of the added thickness of the painter's tape.

If your wiring is such that it would be hazardous to have a metal bed plate near it, you need to fix the wiring.

Quote
the_digital_dentist
If a nozzle crash occurs with an aluminum plate, it may scratch. Scratches are not an issue- you're covering the plate with a layer of kapton tape anyway. If a crash tears the tape and scratches the bed underneath, you replace the tape and you're good to go. What aluminum won't do is shatter into tiny shards that will end up in your bearings, skin, and maybe eyes.
Aluminum tooling plate is cast and expands evenly with temperature, more evenly than extruded plate. It remains flat. The fact that it's thermally conductive means the temperature is relatively even throughout the bulk and surface of the plate.
ABS sticks to kapton tape at about 105C and PLA sticks to painter's tape applied over the kapton tape (assuming you print with ABS and PLA) at a lower temperature. You may have to adjust the Z axis zero point a little when switching from ABS to PLA printing because of the added thickness of the painter's tape.

If your wiring is such that it would be hazardous to have a metal bed plate near it, you need to fix the wiring.

Nozzle-crash: makes sense. Was hoping to get to a minimum-maintenance surface. I've been doing only unheated painter's tape for the PLA just fine, and hairspray-on-glass at 100 C for the ABS. Replacing the tape or swapping it out for PLA/ABS switching is getting tedious (I suppose a stack of pre-cut bed-size sheets of painter's tape would be ok. I actually bought some 2" wide painter's tape to make this easier but it is a different brand and the grip is TOO strong.

I was hoping to get to a naked bed approach for both PLA and ABS. Right now I am running a PLA job on 60 deg naked glass to test a tape-free approach. The switching to ABS would be a lot easier.
Aluminum plate: agreed, it certainly has some major thermal benefits. One of things I do like about a naked glass bed is the nozzle can do a hard crash-n-drag without marring anything (I assume the steppers or the belts skip).
Breakage of course is a danger although I have not come across any anecdotes of it happening, yet.

Wiring: My system has been in the redesign/commission stage since last October so I am still thinking like a prototyper: keep everything available for troubleshooting while everything is powered up. The fewer conductive materials I have around the better. I'm a little clumsy so I have to take accidents into account.

I think for now I will just overheat the glass bed to see if I get less ABS curl and look into a metal bed later.

Thanks for your feedback, it has given me good material to contemplate!

What's the thickness of this glass plate?
How much of the sides of the thermal spreader are exposed?

glass is 3mm, heat spreader is 2.6mm thick aluminum. only the vertical edge is exposed.

I use 5mm thick aluminium with 4mm float glass on top. To switch from PLA to ABS, I replace the naked glass bed by one coated with ABS juice. I haven't tried hairspray yet.

Another thing I like about using glass beds is that if a print sticks too well, I can remove the bed and put it in the fridge. Meanwhile I can start another print on a different bed.

I never got on well with Kapton tape, I found it too tricky to apply.

Edited 1 time(s). Last edit at 06/23/2015 03:26AM by dc42.

---

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].

Applying Kapton tape

The first couple time it heats up, small bubbles will form and you will push them toward the nearest edge of the tape with the edge of a piece of plastic. It will be bubble free after a couple heat/cool cycles.

If you use 5 mil kapton tape, it will be much tougher than the typical 1 or 2 mil stuff and will last for months before it has to be reapplied. I bought a roll of this almost 3 years ago and have been using it to replace the tape on my printer and a few printers at the makerspace and I still have enough to go a couple more years. Using strips instead of a big sheet allows you to replace just the damaged strips when the inevitable happens.

I don't use a heat spreader at all, only 3mm glass directly on my heater. Maximum temperature variation across the surface is less than 10 degrees. For PLA I use blue tape but for ABS I use Wolfbite, no Kapton or other tape. Note that the temperature.
The temperature under the print will be higher than the bare surface as there will be less cooling with the print keeping air away from the surface.

Mike

Odd... I thought i added a reply to this last night that was more detailed then this is going to be, but I'm with Leadinglights. Don't use the thermal spreader if you're just using a standard PCB heating element. The glass plate is sufficient on it's own.

IF you have a specialty heated bed, like a custom silicone heater, or are using wire wound resistors as heating elements, or if you're using your own custom NiChrome wire heater, then a thermal spreader is probably a good idea. Most PCB heaters are not too bad, in terms of even heating. Not to say they don't have hot spots, but the hot spots are not the same as those other sources. Silicone heating pads may even be a bit better then PCB's but they're a bit higher end then I've wanted to go just yet.

As long as your glass plate isn't full of nics and dings, it's probably not going to shatter from uneven heat.