Rostock Knocking over parts.

[6bolt2g]

I struggled a decent amount to get things set up right at first, but I’m incredibly happy with how it’s been performing lately. I’ve printed damn near half a roll of calibration cubes and the end fill test part as recommended by mhackeny. I decided to try to print the owl on a log piece that I see everywhere and I couldn't be happier with how it turned out. There are a couple abnormalities, but overall I’m extremely satisfied.

[img]http://img.photobucket.com/albums/v737/ ... 402fd3.jpg[/img]

The main reason I got into this was I wanted to have the ability to make prototype boat hulls, so I decided to attempt one yesterday, it gave me an ETA of 18 hours! About 5 hours in, it did this…

[img]http://img.photobucket.com/albums/v737/ ... 6e9e0e.jpg[/img]

[img]http://img.photobucket.com/albums/v737/ ... 634896.jpg[/img]

I attempted another piece later and it did the same thing. If you look at the second picture you’ll see how bad that piece that got knocked off is warped, I don’t know if it peeled up and got knocked by the nozzle, or it bowed like that after it fell off.

All the settings are the same from when I printed the owl, I didn't change anything.

Unfortunately I just don’t have enough experience to effectively troubleshoot problems like this, I’m hoping someone has some insight for me.

Thanks a lot guys, this forum has proven to be priceless for me!

[cassetti]

Honestly, looks like a warping issue to me. Simply - your bed isn't perfectly level - for whatever reason, the distance between the print head and the nozzle vary. In the past 10 months I've been printing, it seems the larger your object, the more likely it is to warp. I typically try to start small and gradually work up to larger items once I'm comfortable with my techniques.

Few suggestions:
1) Use paper to calibrate your bed surface - ensure the Z0 coordinates along each rail are just enough to 'grip' the paper.
2) Add a 3 or 4+mm Brim to your print - this helps prevent warping by warping first (instead of the actual print) - I always add at least a 3mm brim to my prints (larger for longer prints)
3) If you REALLY cannot get your bed level, add a 5 or 6 layer raft to the object (New feature in 0.9.9) - Rafts are great for this sort of problem, but only if your bed is nearly level - which from the looks of the owl, yours should be sufficient
4) Increase temp of bed - with the onyx, I'm running around 77c (bad heatup times), but on my work's replicator, I keep the bed at 110c for ABS
5) Adjust your extrusion width for the first layer (in the advanced section) - at one point I had my extrusion width at 0.70mm for only the first layer and that worked good enough. (Really it should be the same as the others)

[6bolt2g]

cassetti,

Thanks for the reply, I’ve spent probably about 8 hours working on leveling the bed. I can command the z axis down to 0 and steer the nozzle around the entire bed (using the manual control arrows) and it will drag a piece of paper with it everywhere, never digging in, or letting go (while at 100*). I also added a dial indicator to the carriage and it keeps it within a couple thousandths across the entire bed.

The bed is at 100* (verified with thermocouple), my unit struggles to get to 100*, let alone 110*, I’ll have to address that issue. I have a replicator at work as well, and your right, it uses 110* on the bed, but it warps things much worse than my Rostock does.

I’m going to look into the brim, I’m not entirely sure what that is. I thought the brim was the “skirt” but I think I’m mistaken.

[Polygonhell]

This is not really a Rostock issue, it's a printing big things on a 3D printer in ABS issue. The warping is caused because the plastic shrinks as it cools pulling the cooled layer below it upwards as it does, the bigger the layer the more the warping force, acute angles result in more force than gentle curves.

The easiest fix is to not print in ABS. Printing big parts in ABS can be an exercise in frustration. ABS juice on glass instead of tape will help, printing on a raft or using a "brim" will also help but there are just limitations to how big you can practically print.
PLA warps considerably less, it's still a challenge to print anything in the 200+mm length range, I've printed parts fairly reliably in the 200mm range, but I still have failures on parts that large.

The first thing every newcomer to 3D printing seems to want is a bigger bed but they don't realize a bigger bed doesn't remove many of the issues with printing bigger parts.

The way commercial machines solve this problem is they heat the entire build envelope to ~60 degrees for the entire print then let the part cool as a whole. Of course Stratsys or 3D Systems has a patent on that.

One thing you can try is using a heat lamp over the print to keep the ambient temperature up.
I've also heard of people using oven safe "turkey" bags over the print area to hold in the heat from the bed and achieve much the same thing.

[mhackney]

Polygonhell is right on. Once you have the first layer or 2 down, any warping or non-planarity of the bed is effectively removed. This is exactly what the "raft" was designed for on early slicers.

ABS is horrific when it comes to warping and delaminating at large sizes - especially with thin walls and sparse infill. You will find even if the part sticks and stays stuck to the bed, it can and will warp and crack much higher up the print. I have a table full of this type of defect! All of them were well stuck to the build surface. Switch to PLA and you will get much better results when it comes to shrinkage and warping. The huge vases I posted in the what are you printing section are all PLA and quite nice. I tried replicating these in ABS, forget it!


A fully enclosed heated build chamber is the only "simple" fix for big ABS parts and even then when you cool the parts slowly they can stress and delam or warp.

One word of caution - PLA is quite easy to print IF you have a small fan to cool the PEEK section of the extruder. Don't have a fan and I guarantee you will have many plugging problems and much frustration.

[Eaglezsoar]

Polygonhell is right on. Once you have the first layer or 2 down, any warping or non-planarity of the bed is effectively removed. This is exactly what the "raft" was designed for on early slicers.

ABS is horrific when it comes to warping and delaminating at large sizes - especially with thin walls and sparse infill. You will find even if the part sticks and stays stuck to the bed, it can and will warp and crack much higher up the print. I have a table full of this type of defect! All of them were well stuck to the build surface. Switch to PLA and you will get much better results when it comes to shrinkage and warping. The huge vases I posted in the what are you printing section are all PLA and quite nice. I tried replicating these in ABS, forget it!


A fully enclosed heated build chamber is the only "simple" fix for big ABS parts and even then when you cool the parts slowly they can stress and delam or warp.

One word of caution - PLA is quite easy to print IF you have a small fan to cool the PEEK section of the extruder. Don't have a fan and I guarantee you will have many plugging problems and much frustration.

Are you still using that small 25mm fan to cool your PEEK?

[dsnettleton]

Unfortunately, nobody told me I couldn't print large parts with ABS, so I've gone ahead and done it anyways. PolygonHell is right that your upper layers are shrinking, curling up the layers below them. This is especially difficult on perimeter-only prints.
I had the hardest time printing this sushi cup I made http://www.thingiverse.com/thing:62185 for that very reason. With some patience and tweaking, however, I was able to get some really good prints out if it. The trick is to make sure the previous layer is cooled as completely as possible before moving to the next layer. That way it's not soft enough to be stretched upwards. Prints like this benefit from speeds of 30-40 mm/s, and a hot end temperature of ~220C. Also, I don't have a fan, but I do have cooling enabled so that any layers which would have taken fewer than 60 seconds are slowed down. In the case of my sushi cups, I've also considered printing several at a time to allow longer cooldown times per layer.

As for the huge vases mhackney was talking about, I was inspired by his blog to make one myself. It's been printing for about five hours with no apparent warping.
[img]http://ubuntuone.com/7D9R5LEo1aRP3xbWnyM1mz[/img]
This is the item http://www.thingiverse.com/thing:37117 (twisty smooth variety), scaled by 1.25 in the X and Y directions and (I think) 1.8 in the Z direction.
Apologies for the bad image, by the way. The lighting is pretty terrible.

[mhackney]

Are you still using that small 25mm fan to cool your PEEK?

Absolutely! I wouldn't print PLA (or Nylon) without it! I actually use it for everything. A cool, cool zone is a good thing.

[mhackney]

@dsnettleton - very cool piece!

My vases in ABS printed for 6+ hours with no problems. They ran overnight and that is when my issues with warping/delaminating occurred. I suspect that my lower night time temperatures created the problem on this large vase.

[dsnettleton]

It appears I posted too soon.
[img]http://ubuntuone.com/51CzMA3NO1GhawKpV8vGDx[/img]
Definite delamination, as well as a strange shift on the Y axis, just over halfway up. I've had that problem on a print before. Any idea what could cause it? I checked the tension on my Z-tower belt, and it seems to be acceptable.

[cassetti]

It appears I posted too soon.
Definite delamination, as well as a strange shift on the Y axis, just over halfway up. I've had that problem on a print before. Any idea what could cause it? I checked the tension on my Z-tower belt, and it seems to be acceptable.

Delamination can be frusterating at times. I usually prefer to closely watch as the first 1-2 layers are applied before I even think about walking away from the printer. I don't know how many times I killed a print because I wasn't happy with the first layer.

Check your delta arms - is there any excess side-to-side "Play" in an arm or two? I had a similar issue halfway up one of my single wall vases, which I suspect was caused by 2 delta arms with too much play. I added some shimming washers to take up the slack, going to try some long prints and see how that helps

[dsnettleton]

Delamination can be frusterating at times. I usually prefer to closely watch as the first 1-2 layers are applied before I even think about walking away from the printer. I don't know how many times I killed a print because I wasn't happy with the first layer.

Check your delta arms - is there any excess side-to-side "Play" in an arm or two? I had a similar issue halfway up one of my single wall vases, which I suspect was caused by 2 delta arms with too much play. I added some shimming washers to take up the slack, going to try some long prints and see how that helps

That was the problem. The first few layers were perfect! The delamination didn't start until well over a quarter of the way up. As for my delta arms, there is some trouble there. I've known they were too loose, but haven't managed to replace them yet. I hadn't thought about shimming with washers, though.