What is your longest print in total hours?

836 views
Skip to first unread message

Jetguy

unread,
Oct 19, 2013, 6:22:15 PM10/19/13
to make...@googlegroups.com
Ok, so the longest print I ran previously was the giant 385mm grumpy pumpkin.
 
Now, I am running a scaled version of the Citadel by a local highschool student and it's at 74 hours and only 60% complete. So far, the machine has run nearly through 2 each 1kg rolls or roughly 4.4 lbs of PLA.
 
 

Steve Johnstone

unread,
Oct 19, 2013, 6:55:14 PM10/19/13
to make...@googlegroups.com
Hi Jetguy,

How do you manage when to load a new spool or do you just keep you eye on it?

Jetguy

unread,
Oct 19, 2013, 6:58:22 PM10/19/13
to make...@googlegroups.com
Just keeping an eye on it.  I do have P-stop enabled and using this just in case http://www.thingiverse.com/thing:83274
 

Darrell jan

unread,
Oct 19, 2013, 6:59:24 PM10/19/13
to make...@googlegroups.com
I don't think I could print that long and have it all fit. Unless it was something like a solid block done at 0.1 mm layer height, which wouldn't be very efficient. Hmm. Haven't done the calculation though.

Jetguy

unread,
Oct 19, 2013, 7:07:26 PM10/19/13
to make...@googlegroups.com
Of course I run Sailfish firmware, it's basically a Replicator 1 supersized to a 400x400xx400mm build area.
So you just use the normal manual pause from the menu to change filament and it resumes the print like nothing ever happened.

Wingcommander

unread,
Oct 20, 2013, 3:41:01 AM10/20/13
to make...@googlegroups.com
An early version of this pedal platform took 43 hours and about 0.9kg of plastic. Nylon prints are 0.5kg plastic and about 27 hours http://www.thingiverse.com/thing:161832/#instructions.

Like Jetguy, I have to keep an eye on the rolls to make sure they don't run out - I set a timer to keep reminding me to check it because its very easy to start reading posts on this group and loose track of time =) I tend to babysit the last meter or so, not much you can print with a small bit of plastic, so I tend to use it all up, before reloading.
nylon-printing-1.jpg

Wingcommander

unread,
Oct 20, 2013, 3:54:20 AM10/20/13
to make...@googlegroups.com
One extra detail, I use a set of kitchen scales to weigh the roll of filament and subtract the weight of an empty one so I know how much plastic is on the reel. If I already have a finished part, I weight the printed part, so I know how much filament I have left and whether I have enough for the print - which is very accurate. However if I don't have enough, I have enough info to estimate how long it will take for the reel to run out - so I set an alarm to remind to remind me to check and change the reel before it does - less accurate so you need to give yourself about 10% margin. I guess you could also set a pause@zpos to do a similar thing if printing unattended. But I am sure that you can guess that if the filament runs out, before the alarm reminds you, the print is somewhat buggered.

Jetguy

unread,
Oct 20, 2013, 9:22:46 PM10/20/13
to make...@googlegroups.com
Ok, so new discovery on Sailfish firmware. If you print for over 100 hours, it rolls over  to 0 on the statistics menu.
So technically, anyone with 99.9 hours beats me?
 
I'm still at 81% hoping this print is done by morning.

Dan Newman

unread,
Oct 20, 2013, 9:43:00 PM10/20/13
to make...@googlegroups.com

On 20 Oct 2013 , at 6:22 PM, Jetguy wrote:

> Ok, so new discovery on Sailfish firmware. If you print for over 100 hours,
> it rolls over to 0 on the statistics menu.
> So technically, anyone with 99.9 hours beats me?

Fortunately, that's just the ticker display. The values stored in EEPROM
for elapsed filament & print time go much, much higher. You need to print
past 65535 hours to wrap that field (7.47 years of continuous printing.)

Dan

P.S. Can likely make the ticker handle up to 999 hours (41.65 days).

Dan Newman

unread,
Oct 20, 2013, 10:02:59 PM10/20/13
to make...@googlegroups.com

On 20 Oct 2013 , at 6:22 PM, Jetguy wrote:

> Ok, so new discovery on Sailfish firmware. If you print for over 100 hours,
> it rolls over to 0 on the statistics menu.


Ahhh, stats menu and not the scrolling ticker. I was confused there as I
just checked the scrolling ticker and it will show up to 999H59M. But
this is your dual extruder bot so you don't get the scrolling ticker.
The stats menu was MBI's and sure enough, it restricts the hours of
the last print to two digits. I'll change that in the 7.6 beta and
update the beta site. (You're safe with 7.6: nothing significant
right now: SD card fix; a deep heater related bug associated with
how a heater is paused while the HBP heats up -- you don't have that
issue ;)

Dan

Dan Newman

unread,
Oct 20, 2013, 10:27:32 PM10/20/13
to make...@googlegroups.com

On 20 Oct 2013 , at 7:02 PM, Dan Newman wrote:

>
> On 20 Oct 2013 , at 6:22 PM, Jetguy wrote:
>
>> Ok, so new discovery on Sailfish firmware. If you print for over 100 hours,
>> it rolls over to 0 on the statistics menu.
>
>
> Ahhh, stats menu and not the scrolling ticker. I was confused there as I
> just checked the scrolling ticker and it will show up to 999H59M. But
> this is your dual extruder bot so you don't get the scrolling ticker.
> The stats menu was MBI's and sure enough, it restricts the hours of
> the last print to two digits. I'll change that in the 7.6 beta and
> update the beta site.

Done: build 1151 at the beta site. I had never looked closely at that
particular code so hadn't noticed the 2 digit restriction until you
pointed it out. I had updated it to include the current Z position
and filament usage which are two items MBIs firmware doesn't display.
But, I must not have given any thought to the hour display on that
screen, assuming I looked closely at it to begin with.

Dan

Jetguy

unread,
Oct 20, 2013, 10:29:50 PM10/20/13
to make...@googlegroups.com
Thanks for looking at it. Honestly, I don't expect many people to ever reach 100 hours on a single print. I'm not saying it's impossible, certainly a low layer height, extremely dense print that fills the build volume is possible.
But then it comes down to affording it. This single print is now working on the 3rd 1kg of filament and I still have 18% left to complete. It's the kind of print that you overnight order filament rolls because you anticipate running out.
At $40-50/kg this is a $150 print. Failure is not an option at this level.

Dan Newman

unread,
Oct 20, 2013, 10:32:09 PM10/20/13
to make...@googlegroups.com

On 20 Oct 2013 , at 7:29 PM, Jetguy wrote:

> Thanks for looking at it. Honestly, I don't expect many people to ever
> reach 100 hours on a single print. I'm not saying it's impossible,
> certainly a low layer height, extremely dense print that fills the build
> volume is possible.
> But then it comes down to affording it. This single print is now working on
> the 3rd 1kg of filament and I still have 18% left to complete. It's the
> kind of print that you overnight order filament rolls because you
> anticipate running out.
> At $40-50/kg this is a $150 print. Failure is not an option at this level.

Ahhh, so I guess you won't be upgrading the firmware mid-print, eh? (Now
wait: someone will ask.)

Dan

macouno

unread,
Oct 21, 2013, 8:12:44 AM10/21/13
to make...@googlegroups.com
jetguy: you win! I got at most 14hrs I think... And that's quite a big print... I'm really curious to see your end result... you will share some pictures I hope? And you are printing at really high detail I guess? I'm curious to see what you're willing to spend $150 on plastic on as well... though I could easily spend that on an old fashioned 2D print as well..

Jetguy

unread,
Oct 21, 2013, 9:31:53 AM10/21/13
to make...@googlegroups.com
"And you are printing at really high detail I guess?"
No, actually not. I went with rather large  0.30mm layer height and max speed. Had I done anything lower it might take a month to print.
It's simply volume. The printer is 400x400x400 (well not true, it can go bigger but I am self limiting to a smaller volume just to be reasonable).
Everybody wants more build volume, more build space, etc. I don't think you realize how long it takes to fill that volume.
Switching to a large nozzle wouldn't help here because it shows folks don't understand how it all works. This is printed with 2 shells and only 4% infill.
Wider nozzle would only make the shells wider. The only place the wider nozzle helps at all is filling floors and roofs. Some might argue that going to a wider nozzle would let you go 1 shell, but that still results in a weak print. 2 shells @ 4% infill is about as low as you can go an have a reasonably durable print. One other area that large nozzle would help in is taller layer heights thus less layers, but just like you implied, the trend is to go even the other way and go lower layer heights and more layers.
 
Here is the file being printed but note I scaled it to fit in a 400x400x400mm build volume so this file is not that size because you cannot post files that large. https://groups.google.com/d/msg/makerbot/nmY0_GHrZNs/UobjApEfDiQJ
 
Consider this the architectural print test. You guys print the little model buildings all the time (someone is always asking in the forum). This is showing what it would take to print the size most of these folks really want to print and what it would cost.
I'm pushing the outer limits of the tech.
 
The coolest part though folks should take away from this is reliability. Some folks, you might even argue many folks, have problems they can't pull off a print that takes 20 minutes to 4 hours without have an extruder failure, skipped steps and shifted prints, air prints, SD card read problems, etc. The electronics are rebuilt Replicator 1 parts that no lie, that been blown up once and repaired. It runs the same Sailfish firmware you folks do. I ran the same Makerware you folks do. The print head is Replicator 1 parts with David's 3 in 1 aluminum extruder upgrades. I'm printing on 1/4 inch glass plate with Aquanet hairspray heated to 50C by a custom silicone heater pad.
 
Basically, it's just a Replicator 1, scaled to extreme size and using premium servo drives with feedback encoders to never, ever skip a step.

big_red_frog

unread,
Oct 21, 2013, 12:12:39 PM10/21/13
to make...@googlegroups.com
Hi Jetguy, 

I assume ( dangerous, I know ) a lot of the print time is support in the cloisters and towers?

A little vaulting in those would of removed the vast majority of support...

Kurt @ Gmail

unread,
Oct 22, 2013, 11:09:53 AM10/22/13
to make...@googlegroups.com
Holy Crow JetGuy - that sounds like an Incredibly MASSIVE Print Job!
And, OF course, am assuming you are running it on your Monster Printer
you created. Too bad you didn't give us a pic of the print job so far. I
suspect, however, once I read more messages of this thread - that you
probably posted a pic as others probably already asked for it. Myself -
My Cable TV/Internet just got repaired this morning (Tues.) - and so I
am a bit behind in this group & reading of the posts!

-K-

Kurt @ Gmail

unread,
Oct 22, 2013, 11:46:58 AM10/22/13
to make...@googlegroups.com
JeyGuy - My Hat off to you - you Always tend to push the Envelope on
what is possible - and this proves it - and its Amazing to see you do
this stuff and you being a part of this Forum/Community. I like the way
you push the envelope w/BIG Arse Prints. I have tried to push the limits
- although - when I did it with the CupCake - it was about trying to
make something REALLY BIG - but, with parts limited in size to what the
CupCake could actually print - and therefore making LARGE things by
printing MANY Parts and Assembling them.

I hope to God you do NOT Have a Failure - as that would be Devastating
at this late stage in the whole print job!!!

L8r,
-K-

Jetguy

unread,
Oct 22, 2013, 12:14:17 PM10/22/13
to make...@googlegroups.com
125 hours and I was sweating bullets on the last bit of the print based on the fact I had only inches of filament left. I really cut it that close.
3Kg of filament.
Pics to be uploaded shortly.

Jetguy

unread,
Oct 22, 2013, 12:15:51 PM10/22/13
to make...@googlegroups.com
125 hours and 5 mintues to be exact. Or 5 full 24 hour days of printing, only pausing twice to change filament spools.

Darrell jan

unread,
Oct 22, 2013, 12:20:52 PM10/22/13
to make...@googlegroups.com
Do you think it would be possible to come up with, say, a 4 extruder design that could have printed your object 4x faster? 
The easy solution is that you could have 4 printers making 4 objects. Harder to have 4 extruders making one object. But maybe it can be done.

happyman

unread,
Oct 22, 2013, 12:34:15 PM10/22/13
to make...@googlegroups.com
The longest I did was 18 hours print, with 0.05mm layer height.

I wonder how long it took for slicer to generate this print file? I believe it probably added few hours to the total print job time.

Also, do you have a list or BOM part detail on your upgraded bot anywhere that I can check it out, I am looking into expand my Rep1 as well, wanting to go 50cm in Z direction.

Lastly, big salute to your bravery! :)

Jetguy

unread,
Oct 22, 2013, 12:35:49 PM10/22/13
to make...@googlegroups.com
Everyone loves to toss around the ideas for making this faster and it is good discussion and certainly promotes a better understanding of 3D printing and limitations in general.
Moving to a larger nozzle helps, but not in ways you want.
For example, moving to a larger nozzle, then changes the "ideal" layer height to a larger value. On something like this where it's 2 shells for strength, even if you go to a larger nozzle that say is as wide as 2 smaller nozzle paths, it's still stacking pencils 1 row high if you know what I mean. So you really always want 2 shells no matter what nozzle size. I already cranked down infill to 4% and that leaves some pretty huge gaps for surface layers to bridge.
I found the extruder was my real limit to speed. It would skip long before X-Y motion with the big servos would even begin to show weakness.
I went to the strongest springs I could on te filament drive, used higher temp than normal, and even turned up the current to the motor considering how well the entire system is heatsinked.
 
What I'm saying is, solutions like 4 heads only make the machine more complicated and 400 times more prone to failure.
I cannot imagine how may lines of code ran over 5 days continuously and didn't have one single glitch. Not an external power glitch and this was not on a UPS. It's only after the print started that i really realized how much work I need to do to develop support systems to make this thing bulletproof. It's easy to armchair the ideas, much harder to implement them one by one. Then there is testing.

David Kessner

unread,
Oct 22, 2013, 12:54:47 PM10/22/13
to make...@googlegroups.com
The main benefit of the larger nozzle, IMHO, is that it allows you to more easily go to thicker layer heights-- and thus reducing the total number of layers for a given print.  The downsides, of course, are that you now are printing with thicker layers!  But I completely agree with you with regard to larger nozzles on the same layer height-- not much benefit at all.

-David K

Jetguy

unread,
Oct 22, 2013, 12:56:14 PM10/22/13
to make...@googlegroups.com
FWIW, here is a rough BOM.
Working on getting the open source files up shortly.
 
There likely are parts missing from that and a whole lot of details obviously. If one had to do it from scratch, estimate about $3k.
Also, you don't have to go servo. It is better IMO, and folks who have gone servo understand, but during initial testing I did some back to back tests running it with Ultimachine NEMA17 motors for X and Y.
After all, it is just a Replicator 1 extruder head system with a lot more mass and span. Stepper works and it a lot louder and doesn't give the same print quality IMO, but is nearly $1K cheaper all said and done.
The 8 frame corners are what makes this thing and I'll have those cade files up shortly. Also, this is scalable to some extent, smaller or bigger. I wouldn't double size it as the span then becomes a problem for rigidity but this system with the corners means that you can changes the aspect ratio and general size up or down to fit your needs and cost.
One tip, this is about 28 inches wide, and the max allowable to fit through most doorways. Keep that in mind while building.
 
FWIW, I am investigating going Core X-Y or H-bot which would reduce parts count possibly drive the cost down. But, that requires an extra special version of Sailfish yet to be developed.
Also, a personal goal is to keep this in MakerBot compatible mode meaning things like Makerware, Sailfish, Rep-G, GPX, and others all work.
 
Sure, I could go Linux CNC, RAMPS, RAMBO, whatever, but that walks it straight out of being MakerBot compatible.
Utili-Rep.xlsx

David Clunie

unread,
Oct 22, 2013, 2:14:38 PM10/22/13
to make...@googlegroups.com
My Ant Farm (that eventually killed one colony) which you can read about and check out the pics here! http://www.dbclunie.com/2013/07/3d-printing-ant-farm-ant-queens-and.html

-david c.


On Saturday, October 19, 2013 3:22:15 PM UTC-7, Jetguy wrote:

Steve Johnstone

unread,
Oct 22, 2013, 3:17:16 PM10/22/13
to make...@googlegroups.com
Congratulations on your HUGE achievement Jetguy... this has to be some sort of record!

Thanks for sharing your BOM and like many cant't wait to see the photos.

Dan Newman

unread,
Oct 22, 2013, 5:07:51 PM10/22/13
to make...@googlegroups.com
>
> I cannot imagine how may lines of code ran over 5 days continuously and
> didn't have one single glitch.

For grins, let's say the average speed was 60 mm/s and the resolution was
50 steps/mm. So, right there is the stepper interrupt code at 3600 times/s.
In 4 days, that's 20.75 million calls to the stepper interrupt which is
just one routine which was being called repeatedly (3600 times) every second.

Now, since the bot doesn't have an idle loop doing no ops, and the AVR is
clocked at 16 million cycles per second, and the typical instruction is either
1 or 2 cycles, the bot conservatively did 46 billion instructions in those 4 days.

No matter how you cut it, it was busy.

Dan

big_red_frog

unread,
Oct 22, 2013, 5:16:46 PM10/22/13
to make...@googlegroups.com
I take the integrity and repeatability in the firmware domain for granted, the phone in your pockets is surpassing this by several orders of magnitude.

It's the integrity and repeatability in the physical domain I find hypnotic!

Steve Johnstone

unread,
Oct 22, 2013, 8:05:05 PM10/22/13
to make...@googlegroups.com
Mind blowing when you put it like that Dan... Reminds me just how much we just take for granted.

Dan Newman

unread,
Oct 22, 2013, 8:15:36 PM10/22/13
to make...@googlegroups.com

On 22 Oct 2013 , at 5:05 PM, Steve Johnstone wrote:

> Mind blowing when you put it like that Dan... Reminds me just how much we just take for granted.

Well, I need to drop a note to MBI letting them know that someone did a 125:05 print.
Mind you, it wasn't on one of their printers. And it was derivative firmware. But
I think they'll be pleased nonetheless.

Dan

Jetguy

unread,
Oct 22, 2013, 9:39:55 PM10/22/13
to make...@googlegroups.com
To be fair, it was a previously blown up Mightyboard  Rev E. and matching LCD/SD panel.
Sliced with Makerware.
 
So for all the crap I give them, not everything is bad. It just takes a few mods here and there....................
 
I mean the fact it was done on a repaired board...........  at least I might now how to repair them correctly?

Jamesarm97

unread,
Oct 23, 2013, 12:09:07 AM10/23/13
to make...@googlegroups.com
Jetguy, wasn't the tardis body you printed for me clocking in at 23 hours? That was a pretty long print.

happyman

unread,
Oct 23, 2013, 11:52:33 AM10/23/13
to make...@googlegroups.com
Thanks for BOM and part# list with sources. Await to see the video of this bot printing. With many of industrial grade parts, do you think the print quality is much better compare to the original bot?

Jetguy

unread,
Oct 23, 2013, 1:14:08 PM10/23/13
to make...@googlegroups.com
Define much better?
IMO, the Replicator 1 prints pretty good stock. This is about scaling it up with print larger and NOT impact quality in a negative way.
To improve quality from an already high standard is a challenge.
That said, I do feel servo does improve the quality and removed any and all ringing.
But that is one aspect of print quality. The Z stage is one of the most accurate and stable designs I've ever used, and that too imporves print quality.
 
But, I wouldn't go trying to print the smallest thing on the biggest printer I own. This is why I'm not going to give up my Replicator 1 XL, for small stuff it has it's place in the fleet. I printed 2 copies of the http://www.thingiverse.com/thing:136235 PHIL TV enclosure (specifically the pciture tube housing) and clearly prefer the look of the print off the big machine. But, sometimes it's nice to run the smaller machine.

Jetguy

unread,
Oct 23, 2013, 2:36:29 PM10/23/13
to make...@googlegroups.com
Sorry if my answer wasn't clear on quality.
I consider that subjective. The reason is the number of variables. Hardware, firmware, software, the plastic, enviromental settings, surface prep, general setup, even the model combine to determine what the final output looks like.
 
I consider a good printer one that faithfully repoduces what the gcode told it to do. This big bot does that, and even maybe tighter and reduced vibration than compared to my original Replicator. All that also leads to a very nearly silent machine. The power supply fans make more noise than anything else.
It's now proven itself as ultra reliable as well.
 
The print:

Jetguy

unread,
Oct 23, 2013, 2:45:35 PM10/23/13
to make...@googlegroups.com
What is sick to me is the other print I finished last night beside it kinda throws off the scale http://www.thingiverse.com/make:51229
The Citadel building is 385mm square. It's 6 lbs of filament after the support was removed. Unfortunately, my rolls of silver metalic PLA did not match, not even close!! Oh well, it needs a paint job anyway.

Obfuscated

unread,
Oct 24, 2013, 11:37:16 AM10/24/13
to make...@googlegroups.com
This is something I had been thinking about previously.. one of the problems with p-stop is that it just stops the printer whever it is which means that if the hot end is over the print it will very likely melt that area as it just sits there.. but what if your setup this filament run out detector to press the button on the front of the machine. You'd have to manually navigate the menu so that one click would trigger the reload function before leaving it unattended but then if the switch was triggered the reload function would lower the plate away from the hotend and wait for you to change out the filament. The only other problem would be to make sure the plate did not cooldown if you are printing ABS but I think is just as simple as changing the settings in the prefs.

On Saturday, October 19, 2013 3:58:22 PM UTC-7, Jetguy wrote:
Just keeping an eye on it.  I do have P-stop enabled and using this just in case http://www.thingiverse.com/thing:83274
 
 

On Saturday, October 19, 2013 6:55:14 PM UTC-4, Steve Johnstone wrote:
Hi Jetguy,

How do you manage when to load a new spool or do you just keep you eye on it?

Dan Newman

unread,
Oct 24, 2013, 11:47:31 AM10/24/13
to make...@googlegroups.com

On 24 Oct 2013 , at 8:37 AM, Obfuscated wrote:

> This is something I had been thinking about previously.. one of the
> problems with p-stop is that it just stops the printer whever it is which
> means that if the hot end is over the print it will very likely melt that
> area as it just sits there..

Actually, P-Stop triggers a normal Sailfish pause. That means that

Replicator 1, 2, 2X: Build platform is lowered away from the extruder
Cupcake, Thing-o-Matic: Z carriage with extruder is raised away from the print

It's the E-Stop which stops everything leaving the extruder over the print.
But that's the protocol for an Emergency Stop: after all, the emergency
may be that someone's hand is caught in the machine.

> but what if your setup this filament run out
> detector to press the button on the front of the machine. You'd have to
> manually navigate the menu so that one click would trigger the reload
> function before leaving it unattended but then if the switch was triggered
> the reload function would lower the plate away from the hotend and wait for
> you to change out the filament. The only other problem would be to make
> sure the plate did not cooldown if you are printing ABS but I think is just
> as simple as changing the settings in the prefs.

If you set the bot for a Hot Pause (Sailfish feature), then nothing cools
down for 30 minutes. After 30 minutes of sitting idle, then the heaters -- all of
them -- cool down.

Dan

Jetguy

unread,
Oct 24, 2013, 12:02:45 PM10/24/13
to make...@googlegroups.com
Anyway, office fun with the other large print 30 hours on the grumpy pumkin. I felt like folks nover got a true idea of the size of the print from previous pics.
This was printed as one piece.
Try that on an stock MakerBot!
 

. .

Obfuscated Records

unread,
Oct 24, 2013, 1:03:38 PM10/24/13
to make...@googlegroups.com
Dan,
 Ah.. well that's good to know.. Somebody else on here gave me the impression it didn't lower  the platform automatically.. guess they were thinking of e-stop.

But what about the issue of the platform cooling.. in my case it could very well be hours before I can reload and in my current setup the print will release itself from the glass plate and hairspray if it cools off? I still haven't installed Sailfish.. maybe this weekend.. maybe they'd be open to changing the idle time for the platform?

Thanks

Adrian



--
You received this message because you are subscribed to a topic in the Google Groups "MakerBot Operators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/makerbot/SEKe_NCTNhU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to makerbot+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



--

Dan Newman

unread,
Oct 24, 2013, 1:08:26 PM10/24/13
to make...@googlegroups.com

On 24 Oct 2013 , at 10:03 AM, Obfuscated Records wrote:

> Dan,
> Ah.. well that's good to know.. Somebody else on here gave me the
> impression it didn't lower the platform automatically.. guess they were
> thinking of e-stop.
>
> But what about the issue of the platform cooling.. in my case it could very
> well be hours before I can reload and in my current setup the print will
> release itself from the glass plate and hairspray if it cools off?

Yes, it may well. It's all tradeoffs.

> I still
> haven't installed Sailfish.. maybe this weekend.. maybe they'd be open to
> changing the idle time for the platform?

That's a safety issue and we -- Jetty and I -- follow MBI's lead on that.
MBI has decided that 30 minutes is how long the bot should be allowed to
sit idle with the heaters on. Unless they think it's okay to make that
configurable and let people set it longer, then it's not wise to make
Sailfish allow that.

Dan

Obfuscated Records

unread,
Oct 24, 2013, 1:58:04 PM10/24/13
to make...@googlegroups.com
Ugh.. well that does seem kinda counter intuitive since obviously it can be left printing for as long as it takes.



Dan

--
You received this message because you are subscribed to a topic in the Google Groups "MakerBot Operators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/makerbot/SEKe_NCTNhU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to makerbot+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Dan Newman

unread,
Oct 24, 2013, 2:04:23 PM10/24/13
to make...@googlegroups.com

On 24 Oct 2013 , at 10:58 AM, Obfuscated Records wrote:

> Ugh.. well that does seem kinda counter intuitive since obviously it can be
> left printing for as long as it takes.

The filament does cook sitting in a hot extruder. And if the P-Stop has triggered,
then theory says that something unexpected may have happened. So, even letting
things sit hot for 30 minutes could be bad news. Of course, things only sit hot
during a pause if you have enabled "Heat during Pause". The default is to let
the heaters turn off during a pause.

Dan

Obfuscated Records

unread,
Oct 24, 2013, 2:38:12 PM10/24/13
to make...@googlegroups.com
Sure well I could see turning off the extruders but not the bed. They should have separate settings.



Dan

--
You received this message because you are subscribed to a topic in the Google Groups "MakerBot Operators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/makerbot/SEKe_NCTNhU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to makerbot+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Dan Newman

unread,
Oct 24, 2013, 11:37:45 PM10/24/13
to make...@googlegroups.com

On 24 Oct 2013 , at 11:38 AM, Obfuscated Records wrote:

> Sure well I could see turning off the extruders but not the bed. They
> should have separate settings.

We have it for the Thing-o-Matic -- it has an ATmega 2560 with
sufficient program space. There literally isn't room for it in
the Replicators which have half the program space. The real cost
is the UI elements in the LCD menu to support the setting.

Dan

Obfuscated Records

unread,
Oct 24, 2013, 11:44:06 PM10/24/13
to make...@googlegroups.com
Well I suppose the other option is bigger rolls of filament lol..

Although use for a similar switch I was considering was to detect if the filament has stopped feeding and stop or pause the printer to prevent an airprint and scorching filament in the hotend.



Dan

--
You received this message because you are subscribed to a topic in the Google Groups "MakerBot Operators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/makerbot/SEKe_NCTNhU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to makerbot+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Elbot

unread,
Oct 26, 2013, 3:18:18 AM10/26/13
to make...@googlegroups.com
 I am working on a bicycle right now. As soon as it is ready and thoroughly tested, I will post on Thingiverse. I have been working on it on and off for about a year now, ever since I saw those Brits 3D print that nylon bike on Youtube. The shape of the bike is very weird but that is necessary since the ABS bicycle would have to hold up at least my weight (190 lbs). Actually, it is a trike because I need to distribute the weight over more wheels or the wheels will collapse. The hub weighs 1 kg and will take about 40 hours to print. The entire bike will take about 200 hours to print with standard firmware. I am dragging my rear end on updating to Jetty firmware, but I know I'll have to do it in order to print out the bike in a reasonable amount of time.....

Elbot

unread,
Oct 26, 2013, 3:24:23 AM10/26/13
to make...@googlegroups.com

OK, you have me beat. The bike is not a single print but many different prints glued together with acetone/ ABS. The hub takes the longest to print. It is 1kg of plastic at 80% fill. I wanted 100% fill but that was impossible because a 1kg roll will only allow you to do the print at 80% fill. I don't know how to change the filament in mid-print. Besides, the print might stop while I am not near the printer, so it might be impossible to do anyways.

I may have to cheat and just use PVC pipes for the bikes tubes.

big_red_frog

unread,
Oct 30, 2013, 12:55:08 AM10/30/13
to make...@googlegroups.com
Just for the record my longest print is now 16 hours for a single build plate, printing multiple parts that is around 60 hours overall for the final design.

I have printed 3 sets now with no failures.

So my reliability on the rep2 is way beyond my original experience.

This is nothing compared to jetguys monster, but still impressive to me.

My only mod is the filament extruder and I am expecting something to go bang soon.

The front fan is definately on the way out, I sometimes have to tap it at startup to get it to stop whining :-(

Jay

unread,
Oct 30, 2013, 8:25:17 AM10/30/13
to make...@googlegroups.com
Printing my Supersized Bathtub Submarine (http://www.thingiverse.com/thing:151435) I have well over 240 hours of just print time. I have almost half that much just cutting, scaling, and fixing the files...15 files that took @ 12 hours each....

The longest single print I've done is the Pen holder that I printed max size...came in the next morning after 8 hours of printing and it was like 21%.....If I remember it was right at 19 hours to print......


Cheers 

Jay

Jetguy

unread,
Oct 30, 2013, 10:13:14 AM10/30/13
to make...@googlegroups.com
Jay, was that mostly sliced using Rep-G with Pypy?
Just curious. I've had to switch to Makerware simply due to the failures of Rep-G. It's done me well for a long time but I had to move on.

Jay

unread,
Oct 30, 2013, 12:25:38 PM10/30/13
to make...@googlegroups.com
Jetguy....

Nope that gave me errors too...

Sliced with Skeinforge...only slicer that (consistently though SLOWLY) gave me good files...

Jay

Steve Johnstone

unread,
Oct 31, 2013, 8:32:03 AM10/31/13
to make...@googlegroups.com
Hi all,

A couple of 1st for me yesterday..

Shipped my 1st in house designed and printed prototype parts. The parts will be used in a working prototype so needed to be right. I wasn't total happy with the black part but had to let it go. Eventually the black part will be manufactured of cast / machined aluminium and the yellow part will be injection moulded.

These have been the longest and largest prints to date on the 2X. The longest single piece taking 25h 39m :)

Thanks to all of you guys on the forum for helping me get here.

I have posted some video of the parts being printed - http://youtu.be/0JCkhygZlbM



On Saturday, 19 October 2013 23:22:15 UTC+1, Jetguy wrote:
Ok, so the longest print I ran previously was the giant 385mm grumpy pumpkin.
 
Now, I am running a scaled version of the Citadel by a local highschool student and it's at 74 hours and only 60% complete. So far, the machine has run nearly through 2 each 1kg rolls or roughly 4.4 lbs of PLA.
 
 

Kurt @ Gmail

unread,
Nov 6, 2013, 10:09:55 PM11/6/13
to make...@googlegroups.com
JetGuy - I just have to say.

In the world of 3D Printing - you are like a GOD in my Eyes. I know - that has Gotta sound Freakin' Weird - but, you just KNOW Your Sh-t, and it blows me away.

Yeah - like Darrell - I also contemplate printing at a more Extravagant level.  Like - what can we do to make things Faster. But, honestly - in MY Case - its all about making things BIGGER! Hell - those that KNOW me - Kinda KNOW that is what I am about - if they know about my Dragon!

I look at experimenting with the Wood type Filament - and the concept of Bigger Extruders - what you already mentioned, along with the pitfalls - and I consider production of BIGGER Things. Come on - Extrude Wood - Big printer - do in Parts - and you have HIGHLY Customized and Really Big Tables - or even BIG Wooden DRAGONS!

What would people PAY for stuff like that???    Ah - in our world - there ARE those that Would pay Big - just to inflate their Egos - to say they have a One-of-A-KIND Big Object!   Its true - there ARE those that will pay - and, if one can make really BIG Printers - I gotta say - if done right - prints OFF those BIG Printers will sell BIG Time!!!

OK - I digress...

Back 2 U JetGuy!

:-)
-K-
--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To unsubscribe from this group and stop receiving emails from it, send an email to makerbot+u...@googlegroups.com.

Kurt @ Gmail

unread,
Nov 6, 2013, 10:17:52 PM11/6/13
to make...@googlegroups.com
David - that's a WILD Blog post - didn't read it all - but, did read the Text at the top (agreed - VERY Pricey) - and look at the pictures (Dang - now I gotta read to see WHY that BIG Queen Ant died - sorry to hear of her passing).

L8r,
-K-


On 10/22/2013 2:14 PM, David Clunie wrote:
My Ant Farm (that eventually killed one colony) which you can read about and check out the pics here! http://www.dbclunie.com/2013/07/3d-printing-ant-farm-ant-queens-and.html

-david c.


On Saturday, October 19, 2013 3:22:15 PM UTC-7, Jetguy wrote:
Ok, so the longest print I ran previously was the giant 385mm grumpy pumpkin.
 
Now, I am running a scaled version of the Citadel by a local highschool student and it's at 74 hours and only 60% complete. So far, the machine has run nearly through 2 each 1kg rolls or roughly 4.4 lbs of PLA.
 
 
--

Kurt @ Gmail

unread,
Nov 6, 2013, 10:21:00 PM11/6/13
to make...@googlegroups.com
OMG - Dan - R Those Calc's of your's correct? Hell - they GOTTA Be -
cause I know that YOU Know Your Sh-t!!!

And - FYI - as a programmer - I can REALLY Respect that - 46 Billion
Lines of code - which is NUTS. Good thing a Programmer did NOT have to
produce all those lines of Code - it would take a Millennium!

:-)
-K-


On 10/22/2013 5:07 PM, Dan Newman wrote:
>> I cannot imagine how may lines of code ran over 5 days continuously and
>> didn't have one single glitch.
> For grins, let's say the average speed was 60 mm/s and the resolution was
> 50 steps/mm. So, right there is the stepper interrupt code at 3600 times/s.
> In 4 days, that's 20.75 million calls to the stepper interrupt which is
> just one routine which was being called repeatedly (3600 times) every second.
>
> Now, since the bot doesn't have an idle loop doing no ops, and the AVR is
> clocked at 16 million cycles per second, and the typical instruction is either
> 1 or 2 cycles, the bot conservatively did 46 billion instructions in those 4 days.
>
> No matter how you cut it, it was busy.
>
> Dan
>

Kurt @ Gmail

unread,
Nov 6, 2013, 10:28:10 PM11/6/13
to make...@googlegroups.com
JetGuy - That PIC is F-ing AWESOME! Now I Truly can appreciate the SIZE of that print - and the Awesomeness of the print (oops - is that a REAL Word? WTF - Who cares!)...           :-)

-K-
--

Kurt @ Gmail

unread,
Nov 6, 2013, 10:32:09 PM11/6/13
to make...@googlegroups.com
Dude - you just GOTTA Show us some PICS of this Bike - hell - email me
directly. I'm totally into the idea of LARGE items based upon parts -
and ONLY Assemble via Acetone (all I EVER Print is ABS). - so - your
bike sounds REALLY Fun!

Do keep us posted!

-K-

Jetguy

unread,
Nov 6, 2013, 10:36:51 PM11/6/13
to make...@googlegroups.com
I'm just an operator like the rest of you folks.
In fact, I may be mental for spending way too much time and money on this hobby.
 

Jetguy

unread,
Nov 6, 2013, 10:40:03 PM11/6/13
to make...@googlegroups.com
Also, that "Unknown" guy apparently copied my design.
Reply all
Reply to author
Forward
0 new messages