When using a heated bed with your printer, I have seen claims of running temperatures of 90c throughout the print.
That seems like a fairly high power use to keep a large slab of, say, aluminium at 90c for long print times (ie multiple hours).
Is there a common 'sweet spot' for operating temperature?
Does it depend on material?
Is a heated bed required?
Heatbeds have two purposes:
The bit about surface energy is straightforward. Most materials are stickier when hot than cold. In comparison, pure mechanical-bonding bed surfaces like fibrous painter's tape and perfboard don't particularly benefit from bed heat.
Warping is a bit more complicated. The basic cause of warping is when the previous layer is allowed to cool and thermally contract before the next layer is deposited. When you stick hot, expanded material on top of cold, contracted material, large shear stresses are generated when the fresh material cools and contracts. Those inter-layer shear stresses then accumulate over many layers into large-scale bending stresses that try to lift the edges of the print off the bed.
So, to prevent warping, we should minimize the amount that the previous layer is allowed to cool before the next layer goes down. But we DO need it to cool solid so the print doesn't sag in a mushy mess. This is a balancing act: cooling the plastic solid without over-cooling it. The optimal temperature for the print is right around the glass point of the plastic: this is the temperature at which the plastic becomes fully solid and thermal contraction stresses start to accumulate.
The extruder pumps more heat into the print as it deposits molten plastic and radiates a little bit of heat. So we want to set the heatbed temp a little bit below the glass point to ensure the print is able to cool solid. Now, this gets a bit difficult, because everyone's print bed temperature sensor is different. What matters is bed surface temp. Many people have to set their bed temp quite a bit higher than the actual surface temp. It's just something you have to calibrate via print results. The exact filament glass point (Tg) also depends on the blend.
There are other schools of thought, for example printing the first layer onto a surface much hotter than Tg for good adhesion, and then dropping the bed temp to a value somewhat below Tg to allow the print to solidify. That works fine too.
But, with all that said, it's important to understand that the heatbed only keeps the bottom of the print warm. A centimeter up from the build plate, the print is typically much closer to ambient temp than it is to the bed temp. Heated build chambers are thus much more effective for large prints. But heatbeds are still quite effective, because they allow building a strong, warp-free foundation that resists warping stresses induced by the cooler zones higher up in the print.
Googling "what temperature for different filaments" gives a few good links, but the top link looks golden.
https://filaments.ca/pages/temperature-guide
They have temperature guides for both extrusion temperatures and heated bed, as well as suggestions for better adhesion. I'm not that experienced, but their information is similar to information I've seen elsewhere.
Their suggestion for PLA is 215-235 degrees Celsius and a bed temperature of 60 to 80 degrees. That sounds a bit hot to me but every brand (and type) of filament will perform best at different temperatures. I've had problem getting nice bridges at 210 degrees, but had excellent results at 190.
For ABS they say 230 to 240 degrees with the bed at 80 to 100.
Use these values as a starting point, when they fail you make an educated guess about what went wrong and adjust (one parameter at a time) til it works for you. Find a calibration object you like and trim the temperatures so it prints the best your printer can. At this point in the technology experimentation is a large part of making things work.
One thing that's important to remember is that the temperature that you set your bed to and the temperature that your bed gets to is not the same thing. Depending on the construction, assembly and quality of your heater the actual temperature can differ anything from almost nothing to twenty degrees or more.
First you say 90c, which means you are talking about ABS likely. I state this as some PLA printers do not even use heated beds, and instead use elaborate rafts only. (Makerbot)
Now that I have gotten that out of the way I wanted to point out to you that it really isn't using that much power. 110 watts (based off this fellow)
http://reprage.com/post/39698552378/how-much-power-does-a-3d-printer-use
8 hours. 0.11 USD.
http://energy.gov/energysaver/estimating-appliance-and-home-electronic-energy-use
I am sure this number is not 100% spot on. That said having had used a power monitor when I first started out I can confirm getting very cheap results. Actually I calculated the costs of powering my 1 meter by 1 meter silicon heater. It was not that much.
This will not apply to every printer. Some are built better than others.
Last for Min Temp. You will pick the temp that works for the specific plastic. Every plastic is different. Materials with high pigments usually need higher temps for extrusions. That said, generally PLA is 60. ABS is 90-100. When it is wrong you will see more (almost always see some) heat warping.
I also want to mention that heat chambers give the best result and would not require you to use the bed as much. You don't see these on printers for two reasons. Heaters can catch fire, and because of patents blocking people from selling (nothing stopping you though. Look into just using a Box without an active heater if it is still a concern
