Correcting part-warp

Has anyone tried heating a warped part in the oven? 190? deg for a few mins, does it correct the warp or just make it into an even more useless blob?

Tried once, useless blob. The warping forces come from inside the part, and your are heating it from the outside.

---

repraplogphase.blogspot.com

What about a microwave? I know they'll heat some plastics up, CAPA certainly, and PLA may be polar enough as well. It would at least heat the inside of the part more than the outside.

Microwave works off water content, PLA SHOULD be uneffected by microwave because it is hyper dry(please don't test this).

Microwave JUST excites water, that's why it heats the middle more than the outside of food. ABS is pretty dry after extrusion, so again should not respond much (at least till it erupts in flames and kills your family).

---

repraplogphase.blogspot.com

Also remember that it is a myth that microwave ovens heat 'from the inside out'. They actually heat from the outside in, just like a regular oven. It is just the for them, the 'surface' is more than a centimeter thick! If you are cooking something that is less than 2 cm total thickness, the the bottom 'surface' and the top 'surface' overlap, and heat the most in the center. When you put a big roast in, it becomes clear that the center is still raw when the outer cm or two is cooked all the way around. Not that this tid bit of info has much bearing on this question. I just don't like to see this misconception repeated.

The salient points, as spacexula pointed out, is that microwave ovens are tuned to a particular water molecule energy transition, and has little effect on most other substances. If you heated it long enough to soften it, you would also probably find hot spots the meted through, and cold spots that weren't hot enough yet.

I would guess that the best way to unwarp a piece to clamp it in a form that matches what it SHOULD be, then heat to the bottom end of the plastic range, increase clamping pressure and let it settle for an hour or so before cooling. Or, just build another part!

Mike

I used a hot air gun to flatten the slightly domed bottom of an ABS pot before I had a heated bed.

---

[www.hydraraptor.blogspot.com]

Keep in mind that the main cause of warping and internal stresses seems to be an uneven rate of cooling at different areas of the object during the print. So if you put your part in the oven, heat it, and then take it straight out again you may still end up with the same uneven cooling and so lead to warping.

So maybe experiment with a controlled warm up to a temperature just below the plastic Tg, hold it for some time, and then a slow controlled cool down. Something along the lines of a 1 to 2 degree temperature raise/drop a minute should keep even the largest object at the same temperature all the way though. Just a thought.

I think if you heat it up and cool it while it is still attached to the bed it will be less warped. I think if you heat it up after it is removed from the bed it will warp more.

I think this is what must have happened to these parts [cgi.ebay.co.uk]. I believe they are parts I sold originally that were perfectly flat but were then used as patterns for moulds, before being re-sold. Somehow they must have been heated and it made them warp. The warping is not the same as that due to the extrusion process because the sides of the objects have shrunk inwards towards the top. E.g. the z-pulleys look half way to being bevel gears.

While the object is being built each layer cools from molten to room temp. When it passes the temperature where it becomes hard it starts to exert force on the layer below. The amount of warping is determined by how much the plastic shrinks from that temperature to room temp. PLA only becomes hard at its glass transition, which is only about 50C, which is why it warps so little. At the other extreme HDPE gets hard just below its melting point at 130C and it has a much higher thermal expansion, so the amount it contracts after it has grabbed the layer blow is far more, ~ 2%.

The bottom layers are flat when they first solidify but subsequent layers on top apply a curling force. If you soften the bottom afterwards it will curl more if it can. If it is still attached to the bed it will be held in place and if the upper layers are softened the stress on the bottom will be relieved. However, from the example above it looks like the top will contract and the sides bend inwards.

Better to use a heated bed / chamber and not print warped parts to start with.

---

[www.hydraraptor.blogspot.com]

Quite a few of the casting materials can heat up as they cure and a few other types are heated up before being poured, but I doubt they would have used these? - Vinamould

Thinking about it Vinamould may make a good support material... I have used it for casting, and it can be reused many times.

Plaster mold would have also heated them.

Makes me wonder if you alternated each layer temperature slightly, Would you counteract warp? Or alternated print speed fast/slow every other layer? would this put a balance of stress into the part and stop warp? - Another thing to try at some point.

Any idea how far up a part heat from the bed is transfered up into the part being printed, when you stop printing is all of the part at the same temperature on a heated bed?

Quote
Any idea how far up a part heat from the bed is transfered up into the part being printed, when you stop printing is all of the part at the same temperature on a heated bed?

I guess that all depends on the print bed temperature, the ambient air temperature, the speed of the print, the type of plastic and how open to convection (free or forced) the printed object is.

I did a little 'out loud thinking' on the issue of warping here which you may find interesting. However please note that I don't yet own a reprap and so whats mentioned on the blog is just my current (possibly flawed) understanding of why warping takes place inferred from reading other peoples blogs.

---

-------------------------------------------------------------------------------------------------
capolight.wordpress.com

Hi Capo

Nice blog about warping, it's an interesting area for development. I'm in the same position at the moment without a working machine, but I will experiment further when I have a repstrap up-and-running.

It would be good to investigate if you can counteract the warp actually in the design / build process but I'm not sure how much of a temperature range you get to play with (ABS) it looks like most people go from 240 - 255 deg C, so I'm not sure that small difference is going to help if trying to print different layer temperatures. Again same as you, just thinking out loud.

So who's going to break out a thermal imaging camera on a part as it is warping?

Seeing is believing

If only infrared cameras didnt start at the $5,000 mark.. Although I'm sure someone who works in an academic or heavy industry environment could get there hands on one.

What if you were to attach a non-contact IR thermopile sensor to the head of your Mendel and use it to take a 'scan' of an object? If you reduced the aperture of the sensor with some kind of pin hole cover you may be able to achieve a respectable resolution for less than $15USD. I suppose you could even use it to map the hot spots on a heated print bed.

---

-------------------------------------------------------------------------------------------------
capolight.wordpress.com

Capo, I think that it would be important to get thermal images as a print is in progress and don't see how that would be possible if you were to use the printer to create the image. I also feel that the z direction temperature gradient is much more important than the x or y and don't see that possible with your proposed solution.

What infill options are available and has anyone experimented with the infill pattern to reduce warping?

I will explain my thinking. If layer one infill and layer two infill are identical, laid over the top of one another then the entire length of layer two is pulling on layer one as the temperature cools. The contact between layer one and two is 100% and thus the forces that cause warping are at there "highest". Now let's look at a situation where layer one and layer two are at 90*. Because there is a gap between passes in layer one, when layer two is laid over the top, the percentage of contact is proportional to the gap in layer one- and thus the warping force is lower.

Let's extend this thinking a bit further. Let's say the infill is at 90* and the gap between passes is ten times the pass itself. Let's also assume that the part we are printing is a 3" cube for visualizing sake.
The top layer infill cools and tries to pull at the layer below, but our infill is sparse and that makes it so the cross sectional area of a top pass can change. It's easier for the plastic to change diameter slightly than to warp the entire model.