myx_2000
Posts: 5
Joined: Fri Dec 04, 2015 2:09 pm

Printing time significantly different than build statistics

Fri Dec 04, 2015 2:36 pm

Hey all,
This is my first post here. I just got my printer about a week and a half ago. Before posting, I decided to get my hands dirty and walk through this several times first. I have watched the videos, and most of this makes sense.
So... Having now tried many prints, the one thing I keep running up against is the build time listed in the "Build Statistics" screen is off by over half in many cases. A couple examples, I have a print that stated 14 hours which took 32, and a filament guide tube offset gizmo which stated 1 hour 40 took closer to 3.

Is this just something that is always off? On my CNC, there is a guestimated time, but it is usually within fifteen minutes to a half hour off on longer jobs.
I know that it can't be exact, but it should at least be in the ball park too.

I was wondering if it might be something related to accelerations and such. Currently I have not modified my printer settings (on the printer itself, not print settings from S3D)

JoeJ
Posts: 1436
Joined: Sun Feb 16, 2014 10:52 am

Re: Printing time significantly different than build statist

Sat Dec 05, 2015 5:51 am

That probably means you have some really slow acceleration settings. But more importantly, please search! This has been asked MANY times before ;)

myx_2000
Posts: 5
Joined: Fri Dec 04, 2015 2:09 pm

Re: Printing time significantly different than build statist

Sat Dec 05, 2015 7:06 pm

Joe, I can see that you have a lot of posts, and can see you have been here for some time, so hear that I will not just cork off. I do my homework ahead of asking.
That said... I did search. I didn't find an answer. I have been looking for some way to adjust my acceleration settings for the good part of 2 hours.

I am on a Flashforge Creator Pro. It says it is running Sailfish which I understand to be the firmware of choice. I have found sections of the sailfish manual that speak to max accelerations. I also understand (if I am reading correctly) that Flashforge has modified the code for the Creator Pro.
But none of this speaks to how I modify the settings. All the settings are stored in the EEPROM in GRBL (the CNC equivalent of Sailfish).

On the CNC, it is easy, you just connect, and send it a few commands, test, and repeat. I can not find any way to adjust the acceleration settings on my machine. In Simplify3D, I have a machine control panel, and I would expect that I could modify the communications tab as I would in the CNC world, but it is strictly binary, so I have no idea where to begin.

CompoundCarl
Posts: 2005
Joined: Wed Aug 05, 2015 6:23 am

Re: Printing time significantly different than build statist

Sun Dec 06, 2015 10:50 am

This may be helpful: http://www.makerbot.com/sailfish/setup#step5

You can use ReplicatorG to edit the EEPROM values following those instructions

scalci
Posts: 10
Joined: Sat Nov 21, 2015 12:14 am

Re: Printing time significantly different than build statist

Tue Dec 08, 2015 1:45 am

I have the same. My printer is a Marlin Rerap using thead rods in stead of ball screws. I am thinking to change this....I tried many configs I saw on this and other forums but my times are more than 3 times more.

I am starting to think it might be hardware related. I am still in search of a solution :(

Fred B
Posts: 30
Joined: Thu Jan 22, 2015 7:07 pm

Re: Printing time significantly different than build statist

Tue Dec 08, 2015 3:02 pm

We all have this problem to varying degrees and yes it's mostly related to acceleration for each individual printer.

Simply cranking up the acceleration may seem like a good idea but it does come with possible print quality issues. Corners can get messed up, any start-stop point may get messy, and worst case is shifted layers during print.

I do think Simplify could implement some type of user generated "correction" that would simulate acceleration but that's probably an old feature request.

For now, I simply double the time shown for each print. Worst case it's close and best case it finishes early.

printhatter
Posts: 11
Joined: Wed Feb 25, 2015 8:58 pm

Re: Printing time significantly different than build statist

Tue Dec 08, 2015 9:38 pm

I have a dual extruder (Powerspec 3d pro, standard 3DS profile) and occasionally get prints that take double the estimated time. Eventually found what causes it:
Prepare to Print! > [it selects all the profiles] > OK > start printing

The fix is to unselect all the profiles i do not want, select only the profile i do want and then it works. Otherwise it does a weird thing of air printing from the other extruder after filament printing from the active extruder. Hence the double print time. Eye-ball your printer as it prints to see if it is fila-air printing.

HTH

skelator
Posts: 7
Joined: Fri May 08, 2015 3:18 am

Re: Printing time significantly different than build statist

Wed Dec 09, 2015 1:04 am

JoeJ wrote:That probably means you have some really slow acceleration settings. But more importantly, please search! This has been asked MANY times before ;)
I know we can change our printer setting, but the main question is what baseline S3D use to calculating the time? as we know they are dozen of various printer available which diffrent kind of setting and capability, shouldnt the S3D have the ability to follow our current acceleration settings instead of the printer need to follow S3D? Just a thought as I dont see any place in S3D to define our acceleration setting nor any information what the baseline S3D used to calculate the build statistic

Fred B
Posts: 30
Joined: Thu Jan 22, 2015 7:07 pm

Re: Printing time significantly different than build statist

Wed Dec 09, 2015 9:25 am

Pulling the acceleration values from the machine can be problematic because different firmware packages calculate acceleration and jerk settings differently. This can have a pretty big effect on time estimates. Simplify could get burned by calculating acceleration differently than your machine (and have to listen to the complaints)

Just wondering if people are including warm up times in this because Simplify is not. Warm up can be pretty long on some of the under powered machines. This could be a huge part of the issue when print times are lower.

Don't get me wrong here. I think adding a user input for acceleration would be useful as long as it doesn't slow things down in the software. We also need to understand that the estimate will never be perfect.

myx_2000
Posts: 5
Joined: Fri Dec 04, 2015 2:09 pm

Re: Printing time significantly different than build statist

Wed Dec 09, 2015 7:41 pm

CompoundCarl, Thank you. I knew before I got my printer that I was going to purchase S3D. So, even though it came with the machine, I did not install ReplicatorG. I would like to think that Simplify3D could add the acceleration page to their software. It would be nice to only need dwell in one piece of software.

It took me some time to tune my Shapeoko CNC (my only reference in all of this). But I know that ever since I it tuned up, it has run amazingly well. So, I am now in tune up mode for my FF Creator Pro. I have been trying some faster settings and this has brought the times down significantly, and far closer to the predicted print time.

It is just a time consuming. On the CNC I can make the changes, then drive the head around to see how the changes effected everything. Here you kinda have to print something to see how it will behave, so the trial and error is a bit more tedious. It's also hard as you have no real time reference between changes as they happen over the period of say 30 minutes, not 30 seconds.

I started Max Speed Change (X&Y) at 15 (default on the machine) and 30 is probably too much. I don't think it will be a good idea to go any faster as the infill alone will probably rip my machine apart.

I am also toying around with the Max Accelerations. It is not really clear on what is doing what. There's both Max Acceleration (Magnitude of Acceleration Vector) and XY&Z Max Acceleration. If they are both max, which is more maximum'er 8-) . I am not too worried about Z as there is not much Z action here where there might be on the CNC. But the X & Y are to be investigated.

The pop up hints in the page suggest that my magnitude vector accelerations not be higher than 10,000. My machine was set at 65,535 for both that and extruder-only moves. I have not touched this setting (yet) If it's truly maxed out, then I don't have to worry about this causing a slow down.

I am unsure of, if the machine get's tweaked for speed/movement, does it know to compensate for filament?

As to warm up time effecting outcome times, I would agree if it were a difference of 10 or 15 min, but several hours is a different story all together.

Return to “General Discussion and Tips”