Mattroberts' Heated Bed
There is nothing new here, just a record of my successful bed design.
I've used 3mm thick aluminium sheet for the bed for two reasons: Firstly it conducts heat very well, so I don't get hotspots. Secondly, at 3mm thick the bed will not bend.
The heater elements are simply aluminium clad wire-sound resistors. I used 14 resistors of 10 ohms each to build a ~200W heater (farnell part number 9506829).
As you can see from the photos, I attached the resistors by using counter sunk M3 screws and nuts. (Unlike Adrian, I drilled the holes on the resistors to 3mm and used nuts).
(When placing the resistors, be careful not to cause interference with the y-bar clamps, motor or idler).
In order to make sure that the bed fits on my standard Mendel: I cut all 20 of the long M4 bolts to be as short as possible. And, it looks like there is plenty of room.
This bed, even without insulation, reaches 120 degrees in less time than it took me to make a cup of tea (i.e. it is fast enough not to be a problem).
Currently, I use a sheet of 316 stainless steel (1.5mm thick). I ordered a thick piece in order to guarantee its flatness (see left photo). In order to increase the grip (and therefore reduce warping) I sanded the stainless with some 150grit sandpaper, the result is a durable bed that has worked for every object I've tried so far (one example in the rightmost photo).
For reference: the underside of the bed was 130C and the ABS was extruded at (a claimed) 260C. The first layer needs to be squashed into the bed in order to maximise the mechanical grip.
Other Build Surfaces
Originally I used kapton tape on a removable sheet of aluminium - which works well enough, but the tape needs replacing every so often.
Since then, I've tried borosilicate glass (which grips far far less than stainless steel and kapton). With glass you have to sand it, otherwise the ABS doesn't stick. If you wish to repeat this experiment: the cheapest way to trial borosilicate glass is to purchase a few microscope slides.