Heated bed super basics

Dear all,

I have a read lots of different solutions for heated bed, but I would like to try some super basic home made solution, which copies the functioning of the heated nozzle. Would you be able to advise if I am in the right direction?

1) If I use a stiff surface as top (ie glass or ceramic, commercial) and I pass a nichrome wire on the bottom of it (trying to cover as much surface as possible) and attach this to RAMP 1.2/1.3/1.4 I should be able to heat without no problem, right?
2) I imagine the nichrome wire should have around 7ohms (same of the print head) to be able to heat quick enough
3) I can control temperature simply adding a thermistor below the bed and connect it to the electronics
4) To ensure even distribution of the heat, I could add some sort of heat conductive tape (like the metal tape for chimneys) on the bottom of the glass bed, this ensuring the avoidance of hot and cold spots which would lead to cracking

How far I am from reality?

Many thanks for any advise you have.
Syncra

1) RAMPS 1.3 and up have a separeate power input and fuse to run the heated bed. Earlier versions are limited to 5A, enough for the machine but not a heated bed.

2) While the target temperatures are lower, the bed is much larger and loses heat more quickly. Not sure how much power you need. I did work this out once and built an uncontrolled heat bed that runs straight from the PSU, but I can't find my notes. Since you are talking about a controlled bed it doesn't matter so much. I think you want about half the resistance of the heater, which should come in the form of as many parallel loops of nichrome as you can fit on to spread the heat as much as possible.

3) That and tell your firmware what to do with it. It should be a matter of tweaking config.h, or it might come pre configured.

4) I use an aluminium bed so I taped my nichrome straight on. I'd prefer to have a metal layer between the heater and the glass, glass is a fairly lousy conductor of heat so you will get hot spots if you just attach the heater straight to it.

I've made two based of this excellent concept, and they work very well: [reprap.org]

And here are a few more options: [reprap.org] A clothes iron might be the cheapest and simplest way, if you don't just use a hair-dryer.

---

--
-Nudel
Blog with RepRap Comic

Here's Adrian Bowyer's latest bed, which I think will be my next one: [reprap.org]

I am using a similar design to Adrian's...pretty much the same, except I use (10) 10ohm 10W Vishay/Dale aluminum housing resistors (data sheet says they are 13.5W but they are labeled as 10). I copied Arcol's blog: [blog.arcol.hu]

My current bed only has 9 resistors on it (I am doing 10 on my build in progress). I get 130C @12V when all the resistors are wired in parallel. I have a standard design Sells mendel bed that is 5mm thick with a sheet of glass on top and wool felt as an insulator underneath. I am adding a resistor because I find it takes about 15 minutes to heat to 130C and I would like that to be a bit faster.

Also, another benefit of a resistor type bed is that you can rewire and run mains power (120VAC), something that you cannot do with a PCB type heater.

---

----------------------------------------------------------------------------------------------------------------------------
www.Robosprout.com RepRap Mendel parts and accessories.
Message if there is anything you need...I have more than what is listed on my site.
Located in the Spokane, WA / Coeur d'Alene, ID area.
Contact: Robosprout@gmail.com Flickr: [www.flickr.com]

Wow you guys use a lot of resistors. I use 4x 2.2ohm arcol aluminium housed resistors wired in parallel running 60w at 12v and they easily get my bed which is 5mm thick aluminium up to 55-60C for PLA. I don't use ABS but I could go hotter by just upping the volts

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

All info are very useful!!

Unfortunately I do not have much clearance underneath the bed and I do not want to rise it very high to accommodate power resistors (I am working on an extended Huxley which has only 100mm z travel and would prefer to not diminish this!), so I cannot use power resistors (which seems like making life easier, blah!). Looks like I am forced to a thin layer of nichrome wire, I think.

A good alternative seems the tape used in US for the heating reptile cages, but not sure I can use that in Europe with 220V. I will visit a zoo shop as soon as I can to see what other ready-made heating alternatives they have in-house.

Thanks for all inputs, and have a good New Year!!!

If you are using 4 x 2.2ohm in parallel that means you have a .55ohm load. If you are putting 12V to it you are getting 21.81A of current from your PSU. Which would give you 261.8W, or 65.45W per resistor.

The reason I use 9-10 resistors is because my resistors are rated for 13.5W...so I can safely dissipate 120-135W without exceeding the rating of the resistors (even though the bed is a very large heatsink so they can do a bit more).

My setup with 10 x 10ohm resistors gives a 1ohm load @12V gives 12A of current for 144W total (14.4W/resistor).

If I were to run mains power and run 9 series resistors I would have a 90 ohm load @120VAC which would give a total of 160W or 17.7W/resistor (10 resistors would give the same 144W as the 12V in parallel)

---

----------------------------------------------------------------------------------------------------------------------------
www.Robosprout.com RepRap Mendel parts and accessories.
Message if there is anything you need...I have more than what is listed on my site.
Located in the Spokane, WA / Coeur d'Alene, ID area.
Contact: Robosprout@gmail.com Flickr: [www.flickr.com]

For what its worth. I also use 10 2.2 ohm resistors in a series parallel combination to give an even heat on a 5mm alli bed with glass on top.

I think this gives me about 0.9R at 12V around 10.8A which is about 130 W

Edited 1 time(s). Last edit at 01/05/2012 02:59PM by rcs.

If you are using nichrome, you want a much lower resistance than for your heated nozzle. The thermal mass and the rate of radiation for the bed is far higher than the nozzle. The RAMPS bed power terminal is capable of putting out 10A of current, as opposed to 2A for the nozzle. If you're running it all off the same power source this will be 12V, resulting in 120W of power going into the bed. To draw all this current you need around 1.2 ohm for your heater. Since 1.2ohm is a pretty short length of nichrome, what I did was run several ~6 ohm lenghts in parallel to spread the heat around. The end result is a bed that can heat up to 110C in around 15 min.

Thanks jsadusk, this is good advice.

Still, how many is "several ~6 ohm lenghts in parallel"? I imagine there is a calculation behind?
And do you use a separate transformer to feed ramps for the bed? Is it a 12V with >120W, I imagine?

Thanks

For parallel resistors of the same value, just divide by the number of legs, so 5 lengths of 6 ohms in parallel = 6/5 = 1.2.

I powered the entire RAMPS off of one PC power supply. It was a pretty beefy one, I think 450W. I put the bed power input and the main power input on two separate hard drive rails, however with a supply that big I doubt it matters. If your power supply can provide the combined wattage of both heaters, it should be fine to wire it to both.

Edited 1 time(s). Last edit at 01/04/2012 08:01PM by jsadusk.

At the moment I am using a simple 60W laptop charger and works ok without heated bed.... Do you think if I put another one in parallel (maybe a litlle more powerful than 60W) it could work?
Otherwise maybe you are right, it's easier and safer to just replace the current power supply with a proper PC power supply. I imagine I can find info on which cable are the 12V easily on the internet.
Many thanks,

No don't wire power supplies in parallel. One will tend to have a slightly higher voltage and supply most, if not all, the current.

---

[www.hydraraptor.blogspot.com]

That's good advice Nophead, and I understand in case you connect two power supplies in parallel to the same mosfet. But what about RAMPS 1.3-1.4, where there are two separated input for power? I thought this was done appositely to run to separated power supplies?
Thanks again

Yes that is fine because it does not connect the two supplies to each other.

---

[www.hydraraptor.blogspot.com]

The separate power input for the bed was done in order to handle the large current (10A or more) needed to heat the bed. The + side goes directly to the + bed terminal only. It can be run with the same power supply as the other power terminal if it has the current rating or it can be a separate power supply. So, yes, connect a power supply with higher rating on the bed terminals.

On RAMPS 1.2 or earlier, you can do the same thing by connecting one side of the bed heating resistor directly to the + terminal of another +12V power supply terminal and connecting the other side to the - terminal of the bed mosfet in order to handle the large current. You also need additional ground wires between the power supply and RAMPS.

Hi brnrd, is this "You also need additional ground wires between the power supply and RAMPS." only for the case of RAMPS 1.2 or also .3 and .4? In this case, do you have a web reference on how to do this?
Thanks

The ground wire is always needed to complete the circuit if you're using a separate supply for the bed. The additional ground wires are also needed to handle the current for the bed if you're using the same power supply for everything. The Ramps wiki has the wiring diagram for each version.