A call for help/ideas to develop the Heated Bed

Aluminum whilst a wonderful material (It is a wonderful conductor, and I really do love the stuff) is badly prone to warping/bending when subjected to a sneeze. It is so soft it has screwed up my lathe (badly) a couple of times.

(yeah I know a bit of an exaggeration, but true non the less)

You can either roll it flat (supposes you know how bent it is in the first place) or mill it flat. Either way it has got to be thick (Torry Amos, professional Widow, Its gotta be big)

To be honest with aluminum (I have a Land Rover and regularly have to unbend something that has got bent of the back of a slight breeze) You are on to a struggle if it wasn't milled out of a solid lump anyway.

So for dimensional stability, don't do aluminum.

Cheers

aka47

PS I look forward to being shot down in flames about this.

---

Necessity hopefully becomes the absentee parent of successfully invented children.

i have the same love hate relationship with aluminum you have there aka47 its a great materal but it also has its flaws.

---

[mike-mack.blogspot.com]

Well, there is aluminum, and then there is aluminum. Heat treated aluminum, know in the states as 6061 T6 aluminum is heat treated to be very strong and stiff. Even in thickness of 3mm or less, it is very hard to bend, and forcing a 90 degree bend may cause it to crack. If you can find some 6061 T6 in 6mm thickness, I think you will never have a bent plate again!

Mike

Hmm I wonder what the heat treating process is and if it could be done in the kitchen oven.

---

Necessity hopefully becomes the absentee parent of successfully invented children.

Although I'm still waiting for my power resistors to arrive, the 2mm piece of aluminium I'm using is very flat, and very strong. Its 30x30cm, but I can't deflect the centre by more than about 1mm, and that takes some force. It sounds like it might be heat treated, as it looks a different colour, and is much much stronger than other sheets I've used. I'm pretty sure its flat enough to use the whole surface once each corner is adjusted to the right height, but if not I plan on epoxying a nut to the bottom centre, then using a bolt from the mdf to get the centre to the right height. Because its so thin, the temperature may be less even, so I plan on using 12 power resistors, up to a maximum of 100W, with a thermocouple for thermal control. I'll post my results on the blog, hopefully tomorrow, but if all goes as planned it will be a fairly cheap and easy design(£5 epoxy, £5 aluminium, £5 thermocouple+amp, £10 resistors, £2 mosfet, so <£30 for the whole thing, excluding fastenings)

If aluminum is a problem, I'll also throw out the idea of using slate. It might sound like an odd suggestion, but slate is extremely flat, very machinable (can be drilled and tapped easily, and cut with a bandsaw or a hacksaw), and is a cheap and widely available material. Its coefficient of thermal expansion is less than half that of aluminum, so that becomes less of a problem.

Slate is also anisotropic, so it will conduct heat better in the plane than perpendicular to the plane - this means that the heat will spread more effectively across the bed than it would for an isotropic material of equivalent conductivity. This is one reason it's used in heated flooring. However its thermal conductivity is still much less than aluminum or even stainless, so it would require a distributed heating element like a nichrome wire grid.

I have often heard of heat treated aluminum, but was never taught the secrets. Here is what I found with a little Googling:

Racing Pistons T6
They talk of a two phase process. Phase one is heat to 920F (493C) for 9 hours to allow the copper in the alloy to fuse with the aluminum. The water quench down to room temp. The second phase is to 'age' the alloy at 350F (177C) for 10 hours to allow the AlCu crystals to grow, but still remain small.

Ebay seller talking about his aluminum rod this one talks of solution heating (copper fusing with aluminum) at 990F (532C), and precipitation heating (growing crystal size) at 320F for 18 hours.

engineer class questions talks about the precipitation heating being 525C.

Looks like you are going to need a VERY hot kitchen oven to do the first step!

Mike

---

Team Open Air
Blog Team Open Air
rocket scientists think LIGHTYEARS outside the box!

... for 'hard' aluminium i buy different alloys with some percents of Mg, Pb and Si - there are some really sturdy materials with excellent milling and dimensional stability ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

On the Hacklab.to Makerbot I have gone with a acrylic sheet (~1/2 inch i think) on top of an aluminum heater block powered by kaptop taped nicrome on the bottom. The assembly is held away from the wooden makerbot by a nutted bolt that attaches to a lower acrylic assembly for mating with the makerbot's build platform interface.

Heres the important bit:
To get acrylic not to warp you need it to be thick and to bolt it to the metal. That way an acrylic sheet was able to last me for the majority of my Mendel parts (which translates into hundreds of prints on the makerbot due to it's various issues + small build platform).

I'm not saying acrylic is perfect, but that is one way to get it to work with a heater reasonably well.

Also resisters are more expensive, harder to source in specific wattages/resistances/heat sink configurations, bulky and if you press 1 side against the aluminum you are leaving over 3/4 of the surface area for convection.

Also, from my understanding, induction heating would only work for ferromagnetic print materials.

/wayToMuchTimeMessingWithHeaters

Edited 2 time(s). Last edit at 03/08/2010 06:14PM by theOtherRob.

theOtherRob Wrote:
-------------------------------------------------------
> On the Hacklab.to Makerbot I have gone with a
> acrylic sheet (~1/2 inch i think) on top of an
> aluminum heater block powered by kaptop taped
> nicrome on the bottom.

Have fun with the acrylic bending in all directions and breaking up due to thermal stress.
It works but large, raftless prints at 80-120°C bed-temperature will not work for a long time before replacing it.
There is a reason everyone tries to insulate the acryplic RepMan from the various heated bed designs and does not just tape a heating pad below the exchangable, acrylic printing bed.

---

-------------------------------------------
* homeprototype free 3d design repository
* Blog
* Google+

Another data point: With the nichrome cut down to 13Ω (0.9Amps / 11 Watts)the temperature of my 1/2" slab of aluminum gets up to 60°C... barely... after about half an hour. ABS did not stick to the Kapton. Definitely need MORE POWER. :-)

Edited 1 time(s). Last edit at 03/11/2010 10:02AM by JohnWasser.

Yes you need about 100C for ABS to stick to kapton and about 1kw / square meter to hold that temperature. At least twice that to get there in a reasonable time.

---

[www.hydraraptor.blogspot.com]

Just saw a program that said the output of the sun was 1kw / m^2. So if you live in a hot country just paint the bed black and leave it in the sun!

---

[www.hydraraptor.blogspot.com]

I guess it is becoming obvious why commercial machinery uses a heated build chamber.

Small is better then.

You should be able to get the hot plate hot for less power if you insulated anything you didn't intend to print on. (including underside and sides and around the build space.)

How much more efficient is anybodies guess. Any body any good at thermodynamics. I am sure it should be predictable through a bit o math.

---

Necessity hopefully becomes the absentee parent of successfully invented children.

nophead Wrote:
-------------------------------------------------------
> Yes you need about 100C for ABS to stick to kapton
> and about 1kw / square meter to hold that
> temperature. At least twice that to get there in a
> reasonable time.


OK. 100x100mm = 1/100th sq m.

1/100 x 1,000 W = 10W

So minimum of 10W (0,83A @ 12V)) and preferably 20W (1.7A @ 12V) or more.

That would <14Ω (preferably 7Ω or less) of Nichrome. I think I'll try 6Ω (2A) just like the extruder barrel.

Thanks for the data!

That cant be... i have an alu-plate 200x120mm x4mm
i need 2,6 Ohm to get 70° C


JohnWasser Wrote:
-------------------------------------------------------
> nophead Wrote:
> --------------------------------------------------
> -----
> > Yes you need about 100C for ABS to stick to
> kapton
> > and about 1kw / square meter to hold that
> > temperature. At least twice that to get there in
> a
> > reasonable time.
>
>
> OK. 100x100mm = 1/100th sq m.
>
> 1/100 x 1,000 W = 10W
>
> So minimum of 10W (0,83A @ 12V)) and preferably
> 20W (1.7A @ 12V) or more.
>
> That would <14Ω (preferably 7Ω or less) of
> Nichrome. I think I'll try 6Ω (2A) just like the
> extruder barrel.
>
> Thanks for the data!

---

Mfg
Wolfgang

Aller Anfang war schwer - Wolfstrap

Reprap-fab.org

Erfahrungen kann man leider schlecht weiter geben, die muss man selber machen.

Stoffel15 Wrote:
-------------------------------------------------------
> That cant be... i have an alu-plate 200x120mm x4mm
> i need 2,6 Ohm to get 70° C

24000 sq mm = 0.024 sq mm
0.024 * 1000W = 24W to maintain 100°C
2.6Ω at 12V = 4.6A at 12V = 55W

You're right. The math does not work out in your case.

I only needed about 40W to keep a 200x200 sheet at 100C. That is where I took the figure from. I assumed it was proportional to area but perhaps the perimeter
has more losses than the centre.

It was also supported by four insulating pillars, so a more lossy insulator could put the power up. For example placing it on a sheet of something more thermally conductive than air.

---

[www.hydraraptor.blogspot.com]

Remember, just because you are using 55W of electricity, that doesn't mean you are converting all of that electrical power into heat. There will always be some conversion loss, and if your conversion loss is higher than in nophead's setups, you will need to compensate.

Cefiar Wrote:
-------------------------------------------------------
> Remember, just because you are using 55W of
> electricity, that doesn't mean you are converting
> all of that electrical power into heat.

actually it all ends up as heat.. we usually try to do useful work with the power on its way to becoming heat, but sometimes we just want the heat.

in this case, we want all the heat to go into the bed, which isn't particularly difficult. the tricky part is holding it in the bed without losing heaps to the environment shortly afterwards. that's where the talk of enclosed chambers started

---

-----------------------------------------------
Wooden Mendel
Teacup Firmware

Triffid_Hunter Wrote:
-------------------------------------------------------
> in this case, we want all the heat to go into the
> bed, which isn't particularly difficult.

Not always, but anything that will block it (ie: any insulating material, air gaps, etc) will create an inefficiency in getting it into the plate. nophead's done a very good job of avoiding that in his designs, but I've not seen anything Stoffel15 has done in regards to a heated bed (I did look at his wiki page), and it could be that there is a problem in his design so that all his heat is not getting into the aluminium plate he is trying to heat.

Note: This is actually what I was trying to refer to in my previous post (and failed badly).

> tricky part is holding it in the bed without
> losing heaps to the environment shortly
> afterwards. that's where the talk of enclosed
> chambers started

Has anyone thought of putting a lid over a heated bed during the initial heating? Could this help reduce the initial time to get it up to temp?

> Has anyone thought of putting a lid over a heated
> bed during the initial heating? Could this help
> reduce the initial time to get it up to temp?

I didn't remember to do such think but I will tomorrow :-)

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

I think its important to be able to able to maintain working temperature with a small fraction of maximum power input, so you're not waiting for ages. I'm using 12 1ohm resistors in a 3x4 array, 1.33ohms in all. At 12V, the 108W still takes a minute or two to get to 55 degrees, I'd recommend even lower resistances if possible. It will actually save considerable amounts of energy if you have a greater maximum power, because you're not sitting at <working temp for as long, so less heat is being lost.

james glanville Wrote:
-------------------------------------------------------
> I think its important to be able to able to
> maintain working temperature with a small fraction
> of maximum power input, so you're not waiting for
> ages. I'm using 12 1ohm resistors in a 3x4 array,
> 1.33ohms in all. At 12V, the 108W still takes a
> minute or two to get to 55 degrees, I'd recommend
> even lower resistances if possible. It will
> actually save considerable amounts of energy if
> you have a greater maximum power, because you're
> not sitting at heat is being lost.

12V/108W from a PC PSU? I was afraid it would not work... maybe there could be some problem, I don't know...

James, don't you want to test this hardware? [objects.reprap.org]

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

My heated bed doesn't differ too much from that one, mine is 2mm aluminum, uses power resistors, a thermocouple and MAX6675, and a STP55NF06 (55A at 60V for only 70p from rapid, I like to massively overspec mosfets so I can forget about overheating them)

casainho Wrote:
> 12V/108W from a PC PSU? I was afraid it would not
> work... maybe there could be some problem, I don't
> know...


15-24V(manually adjustable) 120W from a universal laptop power supply
for 40eur (in an expensive shop).
Works well.

---

-------------------------------------------
* homeprototype free 3d design repository
* Blog
* Google+

james glanville Wrote:
-------------------------------------------------------
> My heated bed doesn't differ too much from that
> one, mine is 2mm aluminum,

You didn't bloged yet as you promised, right? ;-)

Your 300X300 aluminium sheet is for Mendel? in what printer are you using it?

Could you get me a sheet of 4mm, of the size for Mendel? looks like is relative cheap... I can't find on local shops such an aluminium as yours. Did you bought it already flat?


uses power resistors, a
> thermocouple and MAX6675

Why thermocouple and MAX6675? because you had them in stock? -- because that TSIC101 temperature sensor should cost $5 while the MAX6675 + thermocouple should cost $30, no?

, and a STP55NF06 (55A at
> 60V for only 70p from rapid, I like to massively
> overspec mosfets so I can forget about overheating
> them)

Good :-)

Are you using Arduino?

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

MarcusWolschon Wrote:
-------------------------------------------------------
> casainho Wrote:
> > 12V/108W from a PC PSU? I was afraid it would
> not
> > work... maybe there could be some problem, I
> don't
> > know...
>
> 15-24V(manually adjustable) 120W from a universal
> laptop power supply
> for 40eur (in an expensive shop).
> Works well.

And I am using a lab regulated PSU, outputting ~100W and my 150x150x15 aluminium takes LONG to get hot... for PLA, to get the 55ºC it is relative quick, but for the ABS at 110ºC, takes a long time :-(
(15mm thickness to make sure that aluminium do not get bended)

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

I haven't had time to blog about it yet, I will asap. My printer is almost a darwin, at least in terms of dimensions. I got the aluminium from www.aluminiumwarehouse.co.uk but I think there's a pretty large minimum order, and I had to go collect it as their delivery isn't great. I had a free sample MAX6675, and the thermocouple was only a few pounds, and I'd like the flexibility to try reflow soldering on it, and the TSIC101 would burn out at those temperatures. I use an arduino for extruder and bed control, with a 16x2 lcd so I can see what's going on. I'd post my code, but its probably useless as most of it is just dealing with the timing idiosyncrasies of my lcd controller.

james glanville Wrote:

I had a
> free sample MAX6675, and the thermocouple was only
> a few pounds, and I'd like the flexibility to try
> reflow soldering on it, and the TSIC101 would burn
> out at those temperatures. I use an arduino for
> extruder and bed control, with a 16x2 lcd so I can
> see what's going on. I'd post my code, but its
> probably useless as most of it is just dealing
> with the timing idiosyncrasies of my lcd
> controller.

At least blog about this, since looks like you are the first with a kind of 12V/100W heated bed, controlled with Arduino and having a LCD, almost my idea that wrote on that wiki page.

We can try to reuse your hardware, like the same mosfet, etc.

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]