fan causing heated bed to warp prints

need some help guys

i have slowley been getting there with my prints and had the a major overhang issue were they were useless. added a fan as suggested by you guys and there much much better.

the fan has solved every problem i had (overhangs and curling up in corners)

problem is now that when i have an 80mm case fan blowing into the print after about 8mm in height the print will warp in the corners and come away from the bed. only comes away in the corners though the middle stays attached.

any ideas?

the fan is not blowing into the heated bed i have it positioned so its 1cm above the bed blowing across it. also the bed stops at 60c all the time never drops.

im printing with pla at 185C with heat bed on glass at 60C

Have you tried The Brim(TM)?

---

--
-Nudel
Blog with RepRap Comic

i was planning on trying this tomorrow but surely there's an answer to why or fan placement or something

I've never had good luck with PLA directly on glass, my experience is that large prints detatch and warp.
PLA on Kapton at 60 degrees seems to be much better. Other than the fact I hate applying Kapton.
You can also use Blue painters tape on the heated bed IME it sticks so well you have to reapply it before every print.

Enlightx Wrote:
-------------------------------------------------------
> i was planning on trying this tomorrow but surely
> there's an answer to why or fan placement or
> something

The why is simple. You are cooling the print too much, and with each successive layer shrinking, the stresses cause the part to warp. It's somewhat of a trade-off.

I use a fan on smaller prints where warping isn't as much of an issue, but cooling each layer is. I turn it off for larger prints where the print has time to cool, but a fan would cause warping.

---

Help improve the RepRap wiki! Just click "Edit" in the top-right corner of the page and start typing. Anyone can edit the wiki!

im going to try slowing the fan down (using software etc) but was also thinking as iv just seen a video of the up printer.

what do people think about printing a raft under the print?

I guess a raft gives about the same effect as a brim? It's pretty old school though, if you ask me, it makes artefacts on the print and shouldn't be needed with a heated bed.

I often angle my 80mm fan up towards the carriage when printing large things. The increased airflow helps the print quality even though it's not directly over the bed surface. I also up the heated bed from 55C to 65C if I'm having problems.

Also, remember to take off and clean the glass plate regularly with dish washing liquid, and avoid fingerprints. Some people report ABS juice works fine with PLA as well.

---

--
-Nudel
Blog with RepRap Comic

Enlightx Wrote:
-------------------------------------------------------
> problem is now that when i have an 80mm case fan
> blowing into the print after about 8mm in height
> the print will warp in the corners and come away
> from the bed. only comes away in the corners
> though the middle stays attached.
>

I had the same problem as you. With PLA there is a balance to be found: On one hand you need sufficient cooling to prevent warping caused by the proximity of the hot end to the print. On the other hand, too much cooling of the print can cause separation of print layers from each other and from the heated bed, resulting in corners warping up.

Here are the ways I used to solve the problem:

1. I used a smaller fan - an 80mm case fan is overkill and hard to duct to where it is needed. A 40x40x10mm fan should do the job. I eventually settled on a 40x40x20mm fan for a bit more fan power.

2. The cooling should be ducted to the area right near the nozzle. There are a number of fan ducts on thingiverse, or you could make your own. You will need to choose one which suits your x carriage but make sure it ducts the air at the print. Some ducts are designed for ducting air at the thermal barrier of the hot end - but I just use a 2nd fan pointed at the thermal barrier with no ducting.

3. PID temperature control of the hot end (found in marlin or repetier firmware) will keep your hot end closer to the target temperature when cooling is variable. Otherwise, you could find that extrusion temperature lowers when cooling increases.

4. A PWM controllable fan, so the firmware can control the fan speed. Some electronics/firmware combinations make this tricky. I use repetier (software PWM) + sanguinololu with success. I use a transistor to switch 12V from the 5V PWM output. I think I saw a simple mosfet fan driver board on thingiverse recently.

5. Software variable fan control - slic3r contains parameters to vary fan speed according to the layer time, so the level of cooling is automatically increased/decreased according to the layer time. Faster layer times need more cooling. This way you get cooling only when you need it. The slic3r cooling parameters are well thought-out.

With all these measures I was able to eliminate PLA warping problems through fine grained cooling control. I have never needed to use a brim with PLA but with ABS a brim is probably helpful if you haven't got a heated build chamber.

When I was printing PLA I had a120mm case fan blowing on it continuously. Same problem you had. Turned the bed temp down to 35C after the first layer or two and no more warping...

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

Computer case and heatsink fans now come with four wires: two for power, one for speed reporting (closed-loop anyone?), and one for PWM control. According to the electrical spec, anything above 0.8V (and below 5.25V) is acceptable for the PWM signal. Freq. is 25kHz

I know the Sanguinololu has an extra PWM pin on the expansion header, so that might be golden. I think the Azteeg X1 derivative already breaks it out. Might have trouble finding a 40mm^2 fan with that connector, though. I know Delta makes one, but they aren't exactly known for being quiet.

Edited 1 time(s). Last edit at 06/18/2012 03:25PM by Mazaw.

If a layer of plastic is cool it is unlikely to warp.

Things that delay cooling.
Too high a bed temperature especially on the first layers.
Fast printing
Objects or vertical surfaces too close to each other.
Lack of airflow
High ambient temperature.

So aim to get the bed to the glass transition as soon as possible after the first layer.
Also note that if the bed is cooled too much by a fan then the first layer will not stick.
Enable the minimum print time in the slicing program.
Try to get decent gaps between items.
Think about fans if the last two apply.

Been messing all night with heated bed different temps ect all the same

When I'm printing the first layer how "stuck" should it be ?

Currently the first layer if I tap it with tweezers it comes loose even at 90c

Sometimes the pla also is a pain and just skids over the glass

I've tried cleaning ith everything from glass cleaner to isopropanol and even glass cutting solution all the same

When I fist had machine it used to be solid I couldn't get print to release if I tried until glass cooled

You can easily have the heated bed too hot for PLA, nothing will stick at 90 degrees C

I usually find that some colours may not stick at 60 - 65 degrees, but if I lower the temp to 55 or even 45 degrees they stick just fine.

I usually start with 60 and work my way down, not up.

Also try glass cleaner with Vinegar, seems to always work for me, If I use 'Astonish'Glass and tile cleaner, nothing sticks at all, and it needs to be cleaned with nail polish remover, so now I always use 'Mr Muscle Window and Glass with Vinegar' - for a streak free shine - it's very effective!

Edited 1 time(s). Last edit at 06/21/2012 04:55PM by richrap.

---

[richrap.blogspot.com]

Nice one richrap never thought of going lower

I'll have a look tommrow for mr muscle

Fingers crossed for tommrow

At 90C PLA should be a sticky gum like material. It shouldn't come off easily but it is too soft to hold the rest of the object down. I drop the temperature after the first layer so it hardens.

Having said that, PLA from Fabberdashery doesn't seem to need the high first layer temperature. It is very different from natural PLA I have had from the US and NZ.

---

[www.hydraraptor.blogspot.com]

well still struggling lol near just throwing out of the window now

cleaned with nail polish remover and also Mr Muscle.

checked hotend to glass height with a feeler gauge (0.2mm same as layer height)

its just swings and round abouts either it will not stick and if i get it to stick it just comes loose during the print.

should i be coverign the glass in tape? would this help ?

tommrow im gonna get my brother cannon SLR and do some close up HD recordering of it printing ill post it up

nophead Wrote:
-------------------------------------------------------
> Having said that, PLA from Fabberdashery doesn't
> seem to need the high first layer temperature. It
> is very different from natural PLA I have had from
> the US and NZ.

That does seem to be the case, I don't ever use a higher first temperature, same for the whole print - normally just 55 degrees, some are ok at 45, the Red PLA needs 62 degrees C for some reason.

---

[richrap.blogspot.com]

NOW!

im using Red faberdashery lol i was thinking ealrier well i didnt have this problem last week but i was mainly printing in clear and pink.

iv been trying to reprint my parts in the fab red as its a very cool colour and also i think it feels a bit stronger then the other colours.

ill give it a go at 62c tommrow

since putting some kapton tape on the glass i did get about 70% heigh before the part come off the bed so hopefully the small hipe in bed temp will get it to stick

i can then finally have some decent z motor mounts!

Quote

Red PLA needs 62 degrees C for some reason

I am printing red with a bed at 55C. Nothing very big so far though. There seems to be very little correlation between posts in this forum when it comes to what sticks to what at what temperature.

---

[www.hydraraptor.blogspot.com]

nophead Wrote:
-------------------------------------------------------
> Red PLA needs 62 degrees C for some reason
>
> I am printing red with a bed at 55C. Nothing very
> big so far though. There seems to be very little
> correlation between posts in this forum when it
> comes to what sticks to what at what temperature.

Agreed, maybe if we all printed the same model with calibrated settings we would start to be able to define temperatures and differences a little better? maybe not.

When I have needed very large prints in Red I have used a layer of clear/natural PLA for the first one/two layers and Red on top.

I also used clear/natural at the start for the large evil face planter pot to help stick and lower warp then switched to Glow in the dark - this is about ~160mm x 160mm. - Not a single bit of warp - that was printed at 52 degrees C with a small fan blowing across the print.



It's all a bit voodoo science when it comes to prints sticking it seems.

First layer speed is usually quite a critical factor for sticking - I always printed at 30% of the speed for the first layer but in recent months and with bigger prints I run at 26% for first layer speed and that's usually with a 80mm/sec print speed.

Ambient temperature also plays a part, I could get noting at all to stick at temperatures below 12 degrees C in the winter months.

Edited 1 time(s). Last edit at 06/23/2012 03:33AM by richrap.

---

[richrap.blogspot.com]

Just thought i would tell yo guys about the experiments iv been doing today with my heated glass.

after using kapton tape i found that i didnt have priting lifting in corners as bad as just with glass alone.

the problem with using the kaptop is 1 the cost 2 you have to have joins were you put the tape.

so i was thinking surley theres some sort of liquid i can spray/brush on the bed (like iv saw with i think the UP printer)

so 1st i tried some standard model primer on the glass. No good at all.

scraped it all off the glass

2nd i tried some Acrylic Polyurethane Primer. I have high hopes for this as its very tacky.
it was great water based as well so i can apply with a brush and easily clean the brush later.
with the bed at 62C PLA was sticking like nothing iv seen ! i did a couple of prints and nothing budged at all!
problem was getting it off the bed later even lettting it cool it wasnt buding and i ended up having to nock 1 print off and another cracked it was on there so hard. also parts had the primer stuck to the plastic which looks nasty

next i tried some Brusable matt varnish only a very light covering. again pla stuck like crazy and couldnt get anything to release easy.


then i deiced to go with a lighter varnish so to speak. got some PVA Glue and mixed it with water untill thin.
put a thin layer down again dried much the same as the primers.
it stuck nice not as good but i didnt want it as good.

so far iv done a few small prints and PVA glue is working a treat. best thing is when you let the bed cool and pull the model off the pva remains on the bed! so im not even having to reapply it like the primers!

will do some more prints tommrow but hopefully this will save me having to put expensive kapton tape

I have been having significant problems with print warping of late. I have been printing PLA on heated glass- small objects no problem. They stick to the glass until the glass cools down. However, when trying to print a disc about 78 mm in diameter by 6.5 mm high, the print would warp from the outer edges at about layer 15 (0.2 mm layer height- so roughly half way through the print). Warping was consistent no matter what I tried- different temperatures, different printing speeds, different layer heights, cooling fan, no cooling fan. Always roughly the same height, the disc would warp sufficiently to break free from the heated bed.

After perusing the various and sundry recommendations here, I switched from cleaning the glass with acetone (which worked OK with smaller objects) to cleaning with pure vinegar. Then I added the "BRIM"- six skirt loops at 0 mm from the object. Lowered the print speed significantly. I turned off the cooling fan, since the problem seemed to be the edges cooling faster than the center (I do not have a ducted fan set up, so the cooling fan works on the edges more than on the center, where the heat is retained). This seems to have solved the warping issue.

Based on measurements on my smaller printed parts, it seems PLA has about 2% shrinkage as it cools. The edges for my particular disc are much thinner (less thermal mass) than the center, and, being exposed to the cool air, the edges will cool and shrink faster than the central portion. Ergo, warping.

I suspect the "BRIM" solution referenced above was the more critical factor in solving this problem (on this print, at least).

2% seems a lot of shrinkage for PLA. I think I get about 0.5%.

---

[www.hydraraptor.blogspot.com]

My sample base for determining the 2% was actually quite small, but consistent over different geometries. I have not yet compared differently-source filament, or different colors.

richrap Wrote:
-------------------------------------------------------
> That does seem to be the case, I don't ever use a
> higher first temperature, same for the whole print
> - normally just 55 degrees, some are ok at 45, the
> Red PLA needs 62 degrees C for some reason.

I can confirm this with my printers.
With a freshly cleaned glass plate I can run at 45C, but I usually run at 55C. For larger parts that tend to warp, I use 60-65.

For tricky overhangs I use a 80mm fan across the bed.

This is with PLA from Ultimachine, Supply3DPLA and random ebay seller #7 (chinese filament at least).

There is minor differences with various colours, but these temperatures work most of the time.

---

--
-Nudel
Blog with RepRap Comic

2% is more shrinkage than PLA can give you no matter how you process it. Only Polypropylene, TPO, TPU, and some Polyethylene plastics can even approach that large of a shrinkage. Normally they are around 1.2% which is why RepRap printers don't use them.

ABS, PC, HIPS, PS, PC/ABS, and Acrylic are normally around 0.7 to 0.8%.

That is what maks PLA great for FFF in 3D Printers, it is only around 0.5% which is extremely low for any polymer.