Bad prints out of brand New Form 3

[lmlloyd]

Can you go into explicit detail (or show pictures) of the symptoms? I have an early rev Form 3 and I experienced a lot of issues that might be similar.

I was having a lot of failures (around a 72% success rate), and on most of those failures the raft would either incompletely adhere to the platform, or the first several layers of the raft would be on the platform, but then about where the lip of the raft would start, it would just cut off, and all subsequent layers would be stuck to the bottom of the tank.I eventually had a set of parts that just wouldn’t print at all, no matter what I did. I was talking about it on the forum, and someone from Formlabs urged me to call support. When I did, they swapped out my machine.

My original machine was from the first batch of orders, pretty much ordered as soon as they announced the machine. The new machine is not having the same problem, and is so far running at an 85% success rate, with most of those failures coming from one difficult part I finally got printing correctly with some work.

There’s not much to see, but this is what one of my failures would look like in clear resin.

[roypnyc]

I just finished a test of rings with new firmware and update form software- grey V4 looks MUCH MUCH better then any other result I got in the past.

[monk2002uk]

Just printed off a WW2 Russian howitzer with the latest PreForm software upgrade. Printer is the Form 3, with Gray v4 resin at 25um layers setting. Here is the user-visible surface of the model. The base is 30mm x 30mm, just to give you a sense of scale for this 1/285 model. The details are excellent, bearing in mind that a standing figure is only 6-7mm tall:

Here is the top surface, nearest the build platform. You can see the scalloping effect of extra curing between the support connection points. Not an issue for my models but would still be an issue for others:

Robert

[roypnyc]

I might be wrong - but I think what you see on the bottom is not over curing- was your model parallel to the build platform? Is so- you should angle it slightly and that will prevent this situation along with adding more support

[monk2002uk]

‘Over curing’ was just my interpretation of previous reviews of the potential cause for this phenomenon. Happy to be corrected on the terminology as this is a specialised area. The top surface issues are not a problem from my perspective but thank you for the reminder about angling. I think the bigger issue is that, for some prints at least - for example where there is no surface that is out of view to users, any problems of this nature with the top surface rendering will continue to affect the quality of the prints.

Robert

[Olivier_Nicolas]

Here is the top surface, nearest the build platform. You can see the scalloping effect of extra curing between the support connection points.

This is due to the peel forces that “pull” on the thin first layer when it is parallel to the build surface. The support touch points do not pull because the supports are much more resistant to the peeling forces. These slowly equalize from layer to layer and the surface eventually becomes flat. Hence the “pillow button” effect.
Note that this effect is a lot more pronounced on the form 2 ( higher peel forces ) and is one of the main reasons why the Form 3 can get along with smaller touch points.
The solution is to tilt the part.

[larsenstephen]

@lmlloyd - very interesting post. Similar to your experience, late last year, I had an early Form 3 that did not fully print models. It would only print the rafts. The rest of the model was just a thin wafer stuck to the bottom of the resin tank with other debris swimming about the resin. This required wafer removal and resin filtering after each of the first three prints I attempted and a call for help from Formlabs. Removing the wafers stuck to the tank damaged the tank. Formlabs quickly swapped that machine out and replaced the tank, too.

The replacement machine does not have that problem. It sounds like Formlabs was able to correct whatever was going wrong with our early machines.

New tech is high adventure.

The problem I am presently experiencing with the Form 3 printing in black V4 is that it takes a considerably greater force to remove black models from the build platform than gray V4 models. So much more force is required to remove a model from the platform that I am breaking 1 out of every 3 models trying to free them.

This has been talked about in other threads.

I’ll be contacting FL support. Hopefully there is a fix.

[leonhart88]

See my post in the other thread - but i don’t have the same adhesion issues if you sand and clean the build plate well between builds and also specifically use side cutters & the metal spatula to get parts off.

[David_Miller]

See my post in the other thread - but i don’t have the same adhesion issues if you sand and clean the build plate well between builds and also specifically use side cutters & the metal spatula to get parts off.

What grit?

[leonhart88]

400 or 600 grit works well for me without taking forever. Just make sure you clean it really well afterwards to remove aluminum particulates.

Even if I don’t do this - I always clean the build plate with generous amounts of IPA and several paper towels to remove residual resin. I have a pretty OCD process, but it seems to be working really well.

[monk2002uk]

Very helpful information about adjusting wash times and about prepping the build plate, thank you.

Some more pictures of the latest build - Grey v4 at 25 microns:

Some tidy up still needed, as you can see with the remnants of supports around the base edges, but I am very pleased with the print.

Robert

[Boris_Krzic]

Hi Alexander. Your orientation is wrong. I’m also making model railways (with form2) and I’ve experimented with orientation a lot to get best results. My first attempts were as your here. You should tilt locomotive to 30-45 degrees to the platform. That will result in supports on the front (or back) of the locomotive but it just cant be avoided because otherwise you get really big surfaces parallel to build plate and they get all messed up while peeling off. This way your print surface will be minimized and with that chances of layers peeling off or getting garbled are getting lower.

Grills like the ones you’ve circled won’t print good ever if every single one of them isn’t supported. Even if they print they’ll get messed up either by IPA or agitation of IPA so I advise you to make them from etched metal (I use service from https://www.etchworks.eu/). If that is not an option print them as relief. I know that doesn’t look the best but it is better than this surely.

Here is how I orient my models:

[monk2002uk]

Here is further proof of the improvement with Grey resin from the last PreForm software update. @larsenstephen provided a sample file that I printed before and after the update. The first photo shows the two prints, both at 25 microns:

There has been a very significant reduction in layer shifts. Closer inspection shows that the issue with the port holes is still not quite right - there is still a flattening of the top of the circular holes. The following photo is a close up of a second print at 50 microns:

Still some tweaks needed but kudos to the FormLabs team for making such a big improvement.

Robert

[leonhart88]

Whoa - that’s a huge improvement!

[Cesar_Rullier]

Closer inspection shows that the issue with the port holes is still not quite righ

you should support those window headers and the top of the round opening. I Am sure those can be fixed that way. Otherwise the print looks good. It hard to tell how big it is in real life, that would be helpful.

[fantasy2]

Looks like a totally different resin. Wow, well done!

[Alexander22]

Hi Alexander. Your orientation is wrong. I’m also making model railways (with form2) and I’ve experimented with orientation a lot to get best results. My first attempts were as your here. You should tilt locomotive to 30-45 degrees to the platform. That will result in supports on the front (or back) of the locomotive but it just cant be avoided because otherwise you get really big surfaces parallel to build plate and they get all messed up while peeling off. This way your print surface will be minimized and with that chances of layers peeling off or getting garbled are getting lower.

Grills like the ones you’ve circled won’t print good ever if every single one of them isn’t supported. Even if they print they’ll get messed up either by IPA or agitation of IPA so I advise you to make them from etched metal (I use service from https://www.etchworks.eu/ ). If that is not an option print them as relief. I know that doesn’t look the best but it is better than this surely.

Hi Boris. Your model looks nice! You are making definitely H0… My models are smaller - 1:120. I have been making on Wanhao D7 in the past and now I am lucky starting it with F3. I oriented models on Wanhao so as you wrote - 30-35° because otherwise I nave had no chance to have a little bit acceptable quality. And now according my little experience with F3 I see an absolut hight quality nice and no roughness surfaces so that now I would let it so… An another point: If I place it under 30 -35° I have to print about 30 hours and if so(parallel to a platform) - only 18…

Thank you for the link about etched metal. I will try.

It is interesting, the printing quality can not be compared, but… the same grills printed good my old Wanhao. I will more “close” the side grills in CAD, use more supports inside and test… The grill on the roof can be printed separately.

[larsenstephen]

Thanks, Robert!

Hi Cesar,

Regarding supporting the openings, the openings are very small. The model is a ship’s superstructure in 1/350 scale. The round openings represent portholes. They are just 1 mm in diameter, too small for supports.

It should be noted that misshapen openings are occurring mostly in the vertical plane, and most noticeable in gray V4, not black V4. The model does print perfectly on a Form 2 in gray V4 with very precisely shaped openings all around. You can see that Robert’s test print compares very favorably with the Form 2-printed gray V4 model shown below.

Since the Form 3 prints this model very well in black V4 resin with excellent opening shapes, it can be sold in that material. Customers, however, prefer gray for easier painting.

Robert’s test print indicates that the Form 3 update has achieved a much improved print quality in gray V4. This is extremely welcome and encouraging. Once FL is able to improve the shape of openings in the vertical plane, this model will be sellable in gray V4 resin, too, not just black, which is what my customers want.

Very happy with this new update.

[monk2002uk]

It hard to tell how big it is in real life, that would be helpful.

@Cesar_Rullier, the highest point of the model (ie the point that is closest to the ‘e’ in the word ‘release’ on the left of my first photo), is 5cm above the base of the raft:

The widest dimension of the raft is 5.5cm.

Robert

[larsenstephen]

Last night’s Form 3 test prints in black V4 indicate that Form 3 print quality in black V4 is approaching Form 2 print quality in gray V4 for small models with detailed surfaces (see photos below). The Form 2 prints are a bit better, having sharper detail and smoother surfaces.

Hopefully, future Form 3 firmware and PreForm updates will bring further improvements.

Test models are two different kinds of cannons in 1/100 scale, a pair of 1790 Royal Navy 12-pounder Blomefield guns and a set of seven 1790 Royal Navy 18-pounder Blomefield guns, each about 4 cm long.

On the top of the barrel is a decorative King George monogram (a stylized “GR” below a crown), a very challenging detail for 3D printers to reproduce accurately. A rendering of the actual 12-pounder model from which the prints were made is shown first so you can see what the model and the monogram are intended to look like. The 18-pounder is similar.

Form 3 prints in black V4 at 50 microns. Form 2 prints in gray V4 at 50 microns.