My Ender 3 Pro is better on overhangs than my MK3S?
 

My Ender 3 Pro is better on overhangs than my MK3S?  

Page 1 / 2
  RSS
fernwehmind
(@fernwehmind)
Trusted Member

Hi!

I've realized that the overhang and bridging performance of my MK3S worse than my Ender 3 Pro. The left one print on the image was printed on MK3s whereas the right print was printed on Ender 3 Pro.

Why it's like this? Shouldn't Prusa suppose to be better than the Ender 3 pro?
Both prints were printed at the same temperatures. Anybody got an idea about why this happens?

This topic was modified 2 months ago by fernwehmind
Posted : 03/09/2020 2:08 pm
Neophyl
(@neophyl)
Noble Member

Its all down to cooling.  Why do you think there are so many different fan shroud designs for every printer around.  People are always trying to find better ways of cooling their prints.  
I'd put the Mk3s cooling in the middling to low range.  I think the MK3 was better personally.  Neither compare to the petsfang style cooling I added to my cr10s which which I can do 150mm+ bridges due to its bigger fan and better airflow that comes from both sides.  

Also just because you set the print temperatures to be the same does not mean the thermal characteristics of the hotends are the same.  Thats why temperature towers are done on a per filament / per printer basis.  You always have to work out suitable settings for each printer for each filament if you care about whats printed.

 

Posted : 03/09/2020 2:56 pm
bobstro liked
fernwehmind
(@fernwehmind)
Trusted Member

@neophyl

I get you but I still think stock mk3s fan should work more effectively than the ender 3 pro's given that the price of the mk3s..
If I need to change the mk3s cooling fan with another design that is the bad thing for mk3s and that doesn't make sense because I've seen so many people who print better supports.

Regarding the temperatures, I print always at the general temperatures. (210/60)

I printed the temperature tower in the past and i concluded that 190-200 celsius is the sweet spot for the better supports. But at some point, the hotend gotclogged at those temperatures. Therefore I had to up the temperature to 210 desperately.

Maybe this is the hardware problem. I don't know honestly. 

This post was modified 2 months ago by fernwehmind
Posted : 03/09/2020 3:52 pm
jsw
 jsw
(@jsw)
Prominent Member
Posted by: @fernwehmind

@neophyl

Maybe this is the hardware problem. I don't know honestly. 

Did you use the same slicer and the same speed parameters for both devices?

I do agree that, from the prints I've seen and done, one of the weaker points of the I3 series is the quality of bridges and the surfaces directly above supports.

Posted : 03/09/2020 3:59 pm
fernwehmind
(@fernwehmind)
Trusted Member

@jsw

I'm sorry I forgot to say that both prints were sliced on the same slicer and the default settings. (PrusaSlicer)

Speed parameters are little bit different. 60mm/s for prusa, 45mm/s ender 3 pro.

Oh, but bridges and the surfaces above the supports are the critical things. Aren't they? I have a hard time thinking prusa is worse than creality in these aspects.

This post was modified 2 months ago by fernwehmind
Posted : 03/09/2020 4:07 pm
jsw
 jsw
(@jsw)
Prominent Member
Posted by: @fernwehmind

@jsw

I'm sorry I forgot to say that both prints were sliced on the same slicer and the default settings. (PrusaSlicer)

Speed parameters are little bit different. 60mm/s for prusa, 45mm/s ender 3 pro.

Oh, but bridges and the surfaces above the supports are the critical things. Aren't they? I have a hard time thinking prusa is worse than creality in these aspects.

I usually end up using a file and then sandpaper for the visible undersides of supports and bridges.

I'm not familiar enough with the Enders to know what to expect on them.  My impression is that the MK3S, in general, produces equal or better prints.

Now the gold standard for bridges and supports has to be the Dimension at the local 'makerspace', but the cost of that was well into five figures.

Posted : 03/09/2020 4:18 pm
Neophyl
(@neophyl)
Noble Member

Try slowing the Prusa down to the same speed.  That will more time for cooling so it might make a large difference. 

Prusa Slicer doesn’t do a great job of bridges imo, you can get decent results using different slicers.

 The physical printer still isn’t anywhere near as good at raw cooling  than some of the user modified fan setups. 

Posted : 03/09/2020 4:25 pm
charles.h13
(@charles-h13)
Famed Member

My ender and ender Pro are our garbage.  The Ender took me 2 years to get reliable printing though $600 of mods.  The Ender Pro was slightly better at a year.  I think I got lemons, but both are garbage.  

Posted : 03/09/2020 6:41 pm
bobstro
(@bobstro)
Illustrious Member
Posted by: @fernwehmind

[...] I get you but I still think stock mk3s fan should work more effectively than the ender 3 pro's given that the price of the mk3s.

You're going about this the wrong way turning it into a debate about the Ender vs Prusa printers. Price has nothing to do with it. If you buy a $1 paint brush and a $20 paint brush, you're not automatically going to get better results with the more expensive tool. You still have to learn to use it to its fullest.

What you want to do is post your results with the Prusa and ask how to improve them. That way, people who don't have an Ender to compare to can help out, and you won't turn this into a stupid pissing contest. 

Having said that, undersides can often be improved by:

  • Slowing speeds.
  • Decreasing temps or increasing cooling (and making sure they're not fighting each other).
  • Lowering layer heights.

If you'd like to save your part & settings off in a 3MF project file, zip it and attach it to a response here, you'll likely get more suggestions. Just looking at pics is not the best way to analyze a print.

This post was modified 2 months ago by bobstro
My notes and disclaimers on 3D printing and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Posted : 03/09/2020 6:51 pm
fernwehmind
(@fernwehmind)
Trusted Member

@bobstro

I didn't mean turn this into a stupid pissing contest. I just sharing my experiences of both printers and I just wanted to know what cause of this issue.  I appreciate your help. And I'm sorry if I explained my thoughts in the wrong manner.

Here is the 3mf file;

https://drive.google.com/file/d/1Dpu2C6PQL2JEJ7_kJcXsMXSzmnbQSr5w/view?usp=sharing

Posted : 03/09/2020 7:41 pm
nikolai.r
(@nikolai-r)
Noble Member

As already mentioned here, it comes down to cooling performance. The ender 3 pro original fan duct is very beneficial for this model. Most of the air is coming from both sides against the overhang layer (+X, -X axis). I assume it's also pushing the bridges on both sides towards the model. A perfect condition in this case and the result speaks for itself.

The MK3s original fan duct is also letting a lot of air from the front (maybe the most?). So it's blowing more in the same direction as the bridging line for this model. Slowing down might help but don't expect wanders. You can also mount a different fan duct or close the front part of the fan duct, so more air is coming from the sides.

For other models, which have the overhangs 360 degrees around, you should see different result on both printers.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram...
Posted : 03/09/2020 10:08 pm
jsw
 jsw
(@jsw)
Prominent Member

That .3mf brought my slicer to its knees trying to load it.  I had to kill the process.

Posted : 03/09/2020 10:27 pm
fernwehmind
(@fernwehmind)
Trusted Member

@jsw

Oh, i'm sorry to hear that. That's because the model file probably. 

@nikolai-r

I see. Is there any fan duct that you could recommend?

 

Posted : 03/09/2020 10:37 pm
bobstro
(@bobstro)
Illustrious Member

One quick observation: You are printing severe overhangs without supports. It's a surprise that it prints usably on either printer. Slice and look at the undersides for blue areas indicating overhangs in excess of what the printer is expected to handle. For bad prints, you may have found a case where the Ender looks better (though I can't really tell what's going on from your pics and lighting) but that doesn't mean it will work out the same way in every case. With a good print (dealing with overhangs properly), the results might be different. 

I'm running a cut down and scaled down test and may have more suggestions later. 

My notes and disclaimers on 3D printing and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Posted : 03/09/2020 11:42 pm
fernwehmind
(@fernwehmind)
Trusted Member

@bobstro

Thanks for the help. Really appreciate it! 
I left those overhang areas deliberately actually. I thought the printer could handle those overhangs. Apparently I was wrong. I'm still learning 3d printing.
Can't wait to see your suggestions!

Posted : 03/09/2020 11:52 pm
bobstro
(@bobstro)
Illustrious Member
Posted by: @fernwehmind

[...] I left those overhang areas deliberately actually. I thought the printer could handle those overhangs. 

Overhangs are tricky for any consumer-grade FFF printer. Unless you use supports, extreme overhangs are literally trying to print in air. There is very little underlying layer to squish against, so whatever overlap there is with the underlying layer has to have sufficient adhesion (grip) to hold on until it cools.

What you are probably looking for is the setting that controls how extreme of an overhang will trigger automatic support generation. You can find this setting under Print Settings->Support material->Overhang threshold. This setting is the angle off horizontal at which supports are triggered. If the layer being printed is at a lower angle from the underlying layer, supports will be generated is you've enabled Print Settings->Support material->Auto generated supports. This value defaults to 45 degrees. On the Mk3, you may be able to get to 35 degrees before print quality degrades. The exact value will vary by filament type, temperatures, layer heights, cooling -- and yes -- hardware. You may have just found a print at which the Ender defaults happen to work "better" than the Prusa. Thinner layers will result in extrusion that weigh less, so have a better chance of not sagging when printing overhangs.

Whether something printed atop supports is good enough is a personal preference. Since they're printed in air, there is no squish to give nice, tidy appearances, so they tend to look stringy. Some slicers handle supports a bit differently, so may be more pleasing, but I've yet to see supported surfaces that are as good as other layers.

 

My notes and disclaimers on 3D printing and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Posted : 04/09/2020 7:18 am
bobstro
(@bobstro)
Illustrious Member

Some final notes on this topic.

I sliced ol' Epictetus into quarters and shrank him to 50% for a more manageable print. This is sufficient to show the results without taking too long to print. I've attached the resulting test STL for anybody that wants to play along at home.

I tried 6 different variations on settings to deal with bridging. All are printed in PLA at 190C with a 0.4mm nozzle.

  1. As delivered in the 3MF project file.
  2. Variable layer heights with automatically generated supports.
  3. 0.15mm layer heights with no support.
  4. 0.15mm layer heights with automatically generated supports.
  5. 0.10mm layer heights with automatically generated supports.
  6. 0.2mm layer heights tilted at 45 degrees with automatically generated supports.

Here are the results with the base snapped off to view the overhangs.

As you can see, versions 1-5 vary in finish and may or may not be acceptable, depending on what you're trying to accomplish. Option 6 shows the best way to deal with overhangs and supports: Avoid them. By rotating the print 45 degrees (resting ol' Epictetus on his back), ALL of the underside prints well without supports.

If you can't avoid them, playing with print settings can help, but supported surfaces will always look worse than normal surfaces.

My notes and disclaimers on 3D printing and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Posted : 04/09/2020 5:51 pm
fernwehmind
(@fernwehmind)
Trusted Member

@bobstro

Thank you for the detailed experiment. I know that I can rotate the model 45 degrees to deal better with the overhangs actually, but I didn't. Because if I do that the face and the chest of the bust become perimeter. So they lose resolution/quality. I don't think I would want the most important parts (face, chest) to lose details in order to deal with the underside. I know I can use variable layer height but even if I do that still doesn't look good in resolution wise. I think I'm going to try another fan duct design because it makes me sad the fact that the overhang performance of my beloved Prusa is worse my ender 3 pro because of the stock fan design.

Posted : 04/09/2020 6:28 pm
fernwehmind
(@fernwehmind)
Trusted Member

@bobstro

Out of curiosity, do you use stock fan design or changed that with an alternative fan? 

Posted : 04/09/2020 6:29 pm
bobstro
(@bobstro)
Illustrious Member
Posted by: @fernwehmind

Out of curiosity, do you use stock fan design or changed that with an alternative fan? 

I've got some modifications from standard that might come into play:

  • Silicone sock on heatblock. Helps with cooling.
  • Modified fan duct. I've tried several, and none have made a big difference over the default, IME.

Before you go making too big of a deal about alternate fan duct designs, keep in mind that Prusa supplies STLs of these parts, and that fans are often damaged and need replacement over the lifetime of a printer. You'll no doubt wind up replacing ducts for both of your printers at some point. Printing and installing a replacement is a 1 screw operation.

I'm not sure I follow what you're saying regarding the front perimeters when angling the print. They are still vertical surfaces mostly affected by layer height. They are perimeters by definition. The "resolution" shouldn't be any lower, just at a different angle. This is a common trick with high detail miniatures and is quite capable of printing fine detail.

 

This post was modified 2 months ago by bobstro
My notes and disclaimers on 3D printing and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Posted : 04/09/2020 7:03 pm
Page 1 / 2
Share:

Please Login or Register