Rebuilding/customizing a Replicator 5gen

[Bjørn Borud]

I'm on my second extruder, the printer still misbehaves and I suspect this extruder isn't going to make it to the 200 hour mark either.  So this got me thinking: perhaps it would be worth it to just rebuild/customize the thing?  Has anyone tried this or know of someone who has done it?

As a starting point I was considering the following:

• Mount a proper extruder to it -- either by rebuilding  the XY-gantry or by just using what is there.  That is: re-use the magnetic mounts, make an adapter for the castle drive thing and possibly re-use the pogo pins for connections.
• Look at the electronics:  if it is some integrated mess, just rip it out and replace it with something that runs Marlin or similar and new stepper drivers.
• Make a dial indicator mount on the extruder for manual build plate leveling. (I think you can get adequate accuracy for not a lot of money, plus you can jog the thing around the build plate to find out if it is flat)

Other mods that might be interesting is to replace the build plate with a heated one.

Anyone tried?  Anyone interested in trying?

-Bjørn

[Jetguy]

Already brought all this up, already modded a smart extruder, this has been investigated.

 

The question is why?

H bot is never going to be right.

So for all the precision linear ways, it's big pile of junk parts in front of you, crap electronics etc. You end up throwing more away than what you use. Between a decent extruder and mount, between wiring, adapters, and cables, and finally even throwing say $200 worth of decent electronics (If you say ramps you deserve to be hit in the head) you are over budget.

Again, I'm talking about doing it RIGHT, not some half Jimmy rigged ductape J-head hotend contraption. Do it right or don't do it at all. A real extruder based on a Replicator 2 functional design with integrated aluminum mount that bolts directly to the linear bearing and properly captures the belts. A real upgrade to inlcude Core XY using additional idlers and proper mounts. A properly rated flex cable system to mimic the original. A proper X, Y and Z limit switch retrofit. It doesn't take much to figure $200 for electronics, $200 for the extruder and all machined mounting parts, leaving only another $100 for custom wiring harnesses and adapters to attempt to reuse existing wiring if possible.

 

I already figured such a kit for the Z18 that all it did was replace the entire extruder module and replace the mainboard and controller is $500 and not exactly a ton of profit margin to build that kit, let alone support it and deal with "average" clueless customers trying to install and use it.

 

$500 on top of a $2800 mid sized Replicator is a joke. Simply sell the thing to some other sucker and buy a decent printer from the start. Let them deal with the problem.

[Jetguy]

FWIW, I shipped my modded Replicator MIni to TobyCWood so he can independently test the mods I have made to the smart extruder.

 

I gave him 2 extruders clearly labeled. One is modded the other is untouched stock.

 

Here is a flicker album just for the Mini https://www.flickr.com/photos/90025904@N04/sets/72157645441994041/

The mods are as follows:

 

Drilled heater core block to reduce melt zone to match Replicator 2 basic design. AKA heated section is now 10mm thick

 

Installed Duplicator 4 PTFE lined thermal barrier and nozzle. Note, I attempted and the goal was to use a genuine MK8 all metal thermal barrier but could not get sufficient cooling with the stock heatsink. As such, I had to fall back to PTFE lined, but it's STILL better than stock.

 

Installed stiffer spring to return the nozzle down in hope of improving nozzle repeatabiltiy of positioning after retractions.

 

Drilled the plastic magnet holder/filament guide that rides on top of the hotened for Z homing. Stock molding leaves a tapered small diameter bore that is where most jams happen. Drilled out so that if hot molten filament is pulled from the hotend it won't jam here anymore.

[TobyCWood]

We intend to do a critical review of the Gen5 Mini on the podcast. Our previous comments were second and third hand. Next time they will be based on hands on experience with this bot. We will also explore what Jetguys's mods are to the extruder and describe any reliability improvements. If it works... stock or not... we will say so. If it sucks... we will say so with no worry about loss of sponsorship... since we have not as of yet excepted sponsorship.

[lassi kinnunen]

for most 3d printer guys 60-70 bucks ramps combo, sanguinololu or whatever(which one of those you choose hardly matters at all) is enough. quite frankly to say that it is unusable, can't work etc is rubbish, since many people are doing beautiful prints with them every day on all manner of printers from deltas to everything including hbots, corexy.

certainly he could do a rebuild for 200 bucks that would be better and more reliable than 5th gen as it is anyways.. maybe even smoothieboard or x5 mini it with that budget and a new extruder, including even new steppers if the extruder isn't too expensive(but I suppose, you're just going to say that all reprap used extruders are rubbish too).

and just one new smart extruder costs more than that. bear in mind to that smart extruders are significantly more expensive for europeans.

It seems. I'd switch it over to x5mini or smoothieboard. at least you get away from a lot of things that way and there's some possible firmware upgrade path in the future.

as to your kit, 

-lassi 

[David K.]

I don't think it is possible to do a rebuild for 200 bucks on a 5th gen.

There are no endstops on a 5th gen!

You would need to wire in endstops and then you would have to do a significant mod to the extruder.  It would be easier to cut your loses and buy a clone for $1000.

The stiffer spring mod for the smart extruder works, meaning it helps reduce the z-banding lines.  I have attached pictures of a before and after spring mod. 

There is still no fix for the poor thermal design, so expect the smart extruder to jam over time.  I have used three of them, some jam immediately, and others I have printed for about 200 hours then they begin to jam.  (Usually jam immediately with Makerbot default settings.)

The problem with Makerbot is that I would expect the printer to work with their default setting for at least 1000 hours.  But its like they didn't do extended QC testing.  If the settings worked for their sample prints, they declared it good and shipped.

[Ryan Carlyle]

They packed up the assembled 5th Gens before they had working firmware. As soon as working firmware existed, they shipped hardware to users and made them do an online firmware update. Any hardware testing that WAS done must have been with some kind of dev board or an older Mightyboard. 

[Dan Newman]

On 21/11/2014, 10:43 AM, Ryan Carlyle wrote:
> They packed up the assembled 5th Gens before they had working firmware. As
> soon as working firmware existed, they shipped hardware to users and made
> them do an online firmware update.

Actually, in all seriousness they started shipping them a day or two in
advance of working firmware. Recall that when they started shipping, MBI
also stated that when you received them you should first let the bot do a
firmware install/upgrade before attempting to use it for a print.

Dan

[Jetguy]

Missed the point there.

I didn't say RAMPS wouldn't work. It the hack of putting Rerap firmware into a MakerBot frame and more importantly- MakerBot user.

If you use a Mightyboard, Sailfish firmware that runs on a Mightyboard has CORE XY/H bot specific firmware today.

That means we have existing tutorial and settings that instantly maintain compatibility with Makerware since it is still a MakerBot.

So, you maintain all the rules, all the tutorials, all the defaults, and all the knowledge learned for Replicator 2 and 2X, suddenly transported into a 5th gen frame.

You use a Replicator 2 extruder so again maintains that KNOWN toolchain and settings.

You could use anything or you could use "the right thing".

[Bjørn Borud]

Did not know about your podcast. Subscribed now. Looking forward to watching!

[Bjørn Borud]

On Thursday, November 20, 2014 8:08:39 PM UTC+1, Jetguy wrote:

Already brought all this up, already modded a smart extruder, this has been investigated.

I was more interested in just replacing it with an extruder that works and which is serviceable.

 

The question is why?

H bot is never going to be right.

So for all the precision linear ways, it's big pile of junk parts in front of you, crap electronics etc. You end up throwing more away than what you use. Between a decent extruder and mount, between wiring, adapters, and cables, and finally even throwing say $200 worth of decent electronics (If you say ramps you deserve to be hit in the head) you are over budget.

The "why" is easy:  if I get stuck with the thing and it won't work I have a choice between throwing it away and modding it.  (Suggesting someone buy a Makerbot 5gen and then modify it is, as you correctly point out, not something any sane person would do).

I don't know enough about mechanical design to judge if the precision will always be crap on this thing.  I'll take your word for it.

Again, I'm talking about doing it RIGHT, not some half Jimmy rigged ductape J-head hotend contraption. Do it right or don't do it at all. A real extruder based on a Replicator 2 functional design with integrated aluminum mount that bolts directly to the linear bearing and properly captures the belts.

which extruders and hotends would you recommend?

A real upgrade to inlcude Core XY using additional idlers and proper mounts. A properly rated flex cable system to mimic the original. A proper X, Y and Z limit switch retrofit. It doesn't take much to figure $200 for electronics, $200 for the extruder and all machined mounting parts, leaving only another $100 for custom wiring harnesses and adapters to attempt to reuse existing wiring if possible.

 

I already figured such a kit for the Z18 that all it did was replace the entire extruder module and replace the mainboard and controller is $500 and not exactly a ton of profit margin to build that kit, let alone support it and deal with "average" clueless customers trying to install and use it.

 

$500 on top of a $2800 mid sized Replicator is a joke. Simply sell the thing to some other sucker and buy a decent printer from the start. Let them deal with the problem.

I don't have the conscience to sell this thing unless I sell it to someone who knows what he/she is buying and still wants to buy it.  (And even if I wanted to I would probably run afoul with the ethical guidelines of the company I work for.  Which owns the thing).

Hey, you wouldn't happen to be looking for a 3D printer!?  I can give you are real good deal on one.  Almost mint condition!  Just used 175 hours! :)

-Bjørn

[Bjørn Borud]

For reference: a new Smart Extruder sourced from Makerbot's distributor costs $580 here in Norway.  We've already bought one.  And it is clicking its way into an early grave.

(Given the cost of an extruder you can try to calculate how many months of operation before we've spent the cost of the printer in spare parts.  And people complain about expensive filament. HA!  The extruder is the most expensive consumable this machine eats)

-Bjørn

[Jetguy]

Here is my basic thoughts for a "kit"

Make a simple L bracket to hold the extruder of choice from the gantry. There are tons of options but a simple NEMA17 motor bracket will work.

Something like this http://www.amazon.com/NEMA17-Stepper-Motor-Mounting-Bracket/dp/B00JID5O6U/ref=sr_1_1?ie=UTF8&qid=1416841861&sr=8-1&keywords=nema17+bracket

The idea here is that part replaces the current cast aluminum bracket that screws to the X axis linear slider block. You need to drill the linear block mounting pattern into the long leg of the L bracket. and come up with a way to attach the belt to that bracket at well (heck, you'd be surprised how well a couple of zip ties work!!!).

That gives you an extruder motor solidly mounted to the X carriage and in the correct orientation. This opens up huge possible directions in what extruder you can use.

Personally, I believe in keeping MakerBot with MakerBot (even though the new stuff sucks, there is something said for the older stuff).

So I would use any of the printable or even metal upgraded filament drives such as David's 3 in1 extruder upgrade for the filament drive section using an original MakerBot drive gear.

That said, a second choice would be any Wanhao Duplicator 4 V9 drive gear and plastic parts or Flash forge or whatever other brand.

I would use a MK7 style cooling block and heatsink, and a genuine all metal thermal barrier from Carl's upgrades. Alternatively, get a Wanhao or Flash forge extruder kit and cut the bar in half

Or don't and build the first dual extruder 5th gen.

Point is, the bracket gets sandwiched between the motor and the drive system and hotend parts. You might need slightly longer screws to thread into the motor but the sandwich is the best mounting system I can think of for adapting.

Alternatively, if you insist on going Reprap, you still use the bracket, you simply use a different filament drive.

For example, the SeeMe CNC EZ struder bracket system can hold any old J-head or other hotened

Not unlike this, except the bracket will be between the motor and the extruder, and the drive gear will be out a little further on the shaft

[Jeremy Rosser]

When and where will this podcast be, I would really like to watch.

Thank you,

On Thursday, November 20, 2014 1:35:37 PM UTC-5, Bjørn Borud wrote:

[TobyCWood]

It's audio only...

threedprintingtoday.libsyn.com

or

Search "3D Printing Today" in iTunes or Stitcher.

My apologies to the Admin for the blatant plug.

[Jeremy Rosser]

Great. Thank you