Heated plate vs heated chamber

[Seamongrel]

Hi, new to 3d printers and i'm looking at the MakerBot Replicator Z18 and the MakerBot Replicator (5th Generation) but i'm not sure whats best, a heated plate or a heated chamber. I would be printing one off prototypes designed in Solidworks 2014 from small in size, to reasonable size items. Also i've noticed that the two printers use a different material for the base plate, will this matter?

Thanks in advance,

Seamongrel

[Ryan Carlyle]

You don't really need a heated plate OR heated chamber to print PLA. And no one knows anything about the Z18 yet because it isn't released.

Industrial-quality Stratasys printers use heated build chambers. However this technique is patented so all consumer-grade printers to date have used heated build plates.

The one real advantage to a heated build plate over a true HBC is that differential thermal contraction of the plate and printed part allows prints to pop free from certain build plate materials (like glass) upon cooling.

[David Kessner]

If you have a HBC and a removeable build plate then you can still remove the plate and use contraction to remove your print. 

-David K

[Ryan Carlyle]

Very true, you just may get less contraction. I run my build plate at 90-95C true surface temp with ABS. I would only run an HBC at 75-80C. Should be sufficient anyway.

[Christopher M.]

Don't forget that makerbot belongs to stratasys now, which allows them to use their patents. 

The controlled environment of a heated chamber reduces acts against warping thereby it should improve dimensional accuracy of bigger objects.

The problem is nobody nows how well the z18 will perform. The only thing we kinda now is that the Replicator 5 gen firmware needs a huge amount of work.

[joe90]

Well, as I have been having problems on my 2X with warping and cracking with large objects, i have built a heated chamber for it. I am currently testing it to see what the differences are.

[David Kessner]

One thing that has concerned me with putting my Rep2 (not 2x) in a heated chamber is that many parts of the Rep2 were not designed to be at an elevated temperature.  For sure the electronics can't take it much beyond 40C.  But I also wonder about the plastic parts-- the arms and carriage specifically.  The motors, too.  The big concern with the arms is that they might sag beyond the point where the level-control-bolts can adjust it back into level, and it might take minutes to months to sag.

For my heated chamber, I took great care to make sure that the motors and various parts don't get too hot, and I only run my chamber up to 40C.  Of course, I also have my entire Rep2 inside the chamber, including the Mightyboard.  

If you try to print ABS inside a chamber then you will go way beyond the 40C that I do.  If you did it "right" then you'd be up near 90C.  On one hand I want to warn you that you could destroy your Rep2 doing that.  But on the other hand, I want you to try and tell me what happens!  You'll have to use your own judgement on what's acceptable.

Another note to consider...  You want more than just an elevated temperature in your chamber.  You want a consistent temperature.  The more your temp varies the more issues you can have.  My chamber varies only about 3C during a print.  My chamber is very well insulated, so just the heat of the Rep2 running is enough to raise the temperature, and I have to do extra things to COOL it down to 40C.  

So, good luck and tell us all what you did and how it worked out!

-David K

[Ryan Carlyle]

Enclosing the build chamber is a lot lower risk than putting the entire machine "in an oven" because most of the electronics are below the bot outside the build chamber. With my R2x chamber at 55C, the mightyboard enclosure below the bot stays pretty close to room temp. You really just have to worry about the stepper motors (which are rated to ~50C ambient) and all the plastic structural parts.

Running your printer over maybe 40C WILL eventually warp your plastic parts. They're still plenty strong at this temp, but the repeated thermal cycling allows injection molding stresses to relax. The Z-arms and X-ends should be replaced with aluminum if you run your chamber hot. 

The Y-brackets in the upper corner of the bot stay slightly cooler because they can radiate heat through the steel frame. But I'm eagerly waiting on Carl to release his direct-drive Y upgrade to get the last plastic structural parts out of the chamber.

[David Kessner]

Of course it is better to only enclose the print area, but if you are retrofitting an existing printer then it's much easier to put the whole printer inside the chamber.  If you're just doing PLA at 40C then this is fine.  But if you're doing ABS at 90C then you have problems.  Unfortunately Joe90 didn't elaborate on what he did, or how.  

I've been running my Rep2 at 40C ambient for more than 6 months and have not noticed any warping.  But, I should also say that I keep my printer and chamber at 40C 24/7/365.  I do this for three reasons:  1. It takes a long time (hours) to bring it up to a stable temp, and I don't want to wait that long to start a print.  2. I don't want the thermal cycling to make me have to re-level the build plate all the time.  3. My chamber is insulated well enough that it doesn't take much electricity to keep it hot.  Maybe a side effect of this is that the plastic X-Arms haven't sagged, or maybe they just don't sag at 40C.

Correction:  The only time I had warping was with my PLA printed Rep2 Extruder Upgrade.  I changed that to Aluminum, but that's all.  There are other PLA parts in my chamber that have not warped, but these are not under the stress/heat that the Extruder Upgrade is on.  

-David K

[joe90]

I have built a unit which replaces one of the sides of the 2X, it consists of a large heatsink and 4x 200W PTC heaters, controlled by a PID temperature controller. The inlet to the fan is at the top of the side of the printer, and the outlet at the bottom just above the floor. The fan is running constantly, circulating the heat. The heatsink is on the large size and so effectively acts to stabilize the temperature. I can run at any temp upto around 80C. Even having the temp at 50C makes a difference, It seems to even out the chamber temp.

Let me know if you are interested in more design info.

I'm just running my first large and long print.

I have everything as stock apart from the extruder which is here http://www.thingiverse.com/thing:222667

I accept that I may need to change other things such as metal arms, but one step at a time.

[Dan Newman]

On 13/04/2014, 4:59 AM, joe90 wrote:
> I have built a unit which replaces one of the sides of the 2X, it consists
> of a large heatsink and 4x 200W PTC heaters, controlled by a PID
> temperature controller. The inlet to the fan is at the top of the side of
> the printer, and the outlet at the bottom just above the floor. The fan is
> running constantly, circulating the heat. The heatsink is on the large size
> and so effectively acts to stabilize the temperature. I can run at any temp
> upto around 80C. Even having the temp at 50C makes a difference, It seems
> to even out the chamber temp.

Keep in mind that some SD cards are only rated to 75C. (Back when I checked
a year ago or so the various manufacturers were listing numbers ranging from
75 - 85C, if they provided the info at all.) One issue with MBI's Rep 2 and
2X design is they put the SD card inside the chamber and at an orientation
which welcomes debris falling into the slot. The Rep 1 has it outside
the chamber and at an orientation for which catching debris is much less
of an issue.

Dan

[David Kessner]

800 watts?  That's crazy!  :)  My chamber uses a single 72 watt heater made from power resistors mounted to a heatsink+fan, also controlled by a PID controller.  I general you want to use the smallest heater you can, while still reaching the temps you want.  While this does increase warm-up time, it also reduces temperature differences within the chamber.  But the big thing it does is reduce the damage that can happen to your chamber, printer, or yourself should something go catastrophically wrong.  

Most PID controllers have an LED that will tell you if your heater is on or not.  If you watch that LED you can get a rough idea for the power that it is using to maintain temp.  For example, if it is on 25% of the time then you are using roughly 200 of your 800 watts.  This in turn will tell you if you really need 800 watts.  My guess is that you'll need 200 or 400 watts max, It's more than my 72 watts because you don't have as much insulation.

At temps above 40C you should consider putting heat sinks on the motors.  The extruder and Y-Axis motor (the one that goes front to back) are the most important, since these do the most work and are the easiest to put a heat sink on (on the Rep2). I used 40x40mm heat sinks attached with thermally conductive epoxy.   Alternatively, you can add more fans and have them blow on the motors.  

Anyway, it all sounds good.  Show us a pic!

-David K

[Ryan Carlyle]

My experience with a sealed-up R2x is that you need around 6-10 watts per degree temp rise.

I don't think plain heat sinks on steppers accomplish much. If you actually look at the heat dissipation specs for 40mm heat sinks, you're probably not even getting a single watt of cooling by gluing a heat sink to the back of the stepper. Natural convection doesn't do much -- forced air is important to getting an appreciable amount of cooling.

The steppers are rated (I believe) at 50C ambient, which means they can handle the internal heat from running at max torque/power in a 50C environment. So running them at lighter duty should allow higher ambient temps.

I have a 50w peltier cooling my Y stepper right now, which probably gives around 5W of net cooling. That drops the stepper running temp by about 20 degrees. It also provides 50w of heat to the enclosure which is a good thing in this situation. 55C requires no extra heating -- just the Y cooler, HBP, and two hot ends put off that much heat.

I plan to do the same peltier cooling with my X stepper and extruder stepper but there's more mechanical design to mount those... so I haven't bothered yet. Too many other projects to work on. Upgrading to Carl's Alu parts helped dissipate a little heat from those too.

[joe90]

I did try with lower powers, but I wanted to use PTC heaters as they are 'self limiting'. I built and tried lower powers, but was limited in the build chamber temp achievable. With the set-up I have now, the chamber heats up fast, in a couple of minutes, and then pretty well stays off for the remaining build. The power light flashes every few seconds...In reality once heated, the power will be low.

The main thing was to use the PTC heaters, with these they are limited in surface temp. The heat sink really holds a steady temp. I have another temp sensor in the chamber, and it holds +-1C....

A couple of photos are attached...

[David Kessner]

My measurements of the steppers before and after attaching the heat sinks showed a 10-15C drop in temp, in my already 40C chamber.  You're assuming that there is no air moving, but there is.  Not only is the extruder moving left/right, which produces it's own air movement, but I also have many fans inside the chamber to circulate air and even out the temp-- four fans to be exact.

At one point I did build a shroud/holder that would attach a 40mm fan to the Y motor heat sink and blow air directly on it.  While that dropped the Y temp by more than 20C, I decided that it wasn't required and just cluttered up the wiring.  If I ran my chamber higher than 40C I would probably reconsider that.

-David K

[Ryan Carlyle]

Yeah, heat sinks work well if you have good air circulation. I have so little air motion that the bottom of my chamber is as much as 15C cooler than the top. Been too lazy to install my chamber fans.

I've also been considering making a "Bowden blower" with some air tubing, to inject cooler outside air directly onto the extruders/steppers. Only issue is finding fans that have high enough static pressure without being unacceptably loud outside the chamber.