Jeff- fantastic review so far, I have really enjoyed reading it, and have shared with the other Makerheads at the shop here. I think we'll stick with the Rep 2's for the time being... :)
Printable View
Jeff- fantastic review so far, I have really enjoyed reading it, and have shared with the other Makerheads at the shop here. I think we'll stick with the Rep 2's for the time being... :)
Thanks Tim! How's your Form 1 doing?
When I next went to use the machine, I heard the familiar clicking sound of a Makerbot filament jam. Usually you can clear these out by unloading the filament, cutting a clean edge off and reloading. All of our printers have had filament jams at various points, and when the unloading, cutting and reloading routine doesn't clear the jam the extruder must be disassembled to get at the jam. That was the case this time. Even though Makerbot supposedly had a replacement smart extruder on the way, we're professionals (literally) and know what we're doing.
After removing the extruder and allowing it to cool, the bottom half of the duct that blows onto the build is unclipped from the top half of the ductwork.
http://american3dprinting.net/images/IMGP55231024.JPG
Next, slide the top half of the duct off.
http://american3dprinting.net/images/IMGP55241024.JPG
IMPORTANT and not shown here because I didn't realize it was important until I almost caused myself some grief. The next step is to remove the filament feed tube from the top of the extruder. This part goes over an angled feature at the top of the extruder and retains the two halves of the enclosure to each other. The enclosure halves cannot be separated until this part is removed!
http://american3dprinting.net/images/IMGP54731024A.JPG
Next, VERY CAREFULLY pry the 6 retaining clips from their mate and remove the cover to reveal the inside! You can see the 4 magnet on the periphery, a large plastic idler wheel, a bearing, a couple of circuit boards, the small filament idler and the plastic spring loaded lever which tensions the small idler up against the filament. On the left you can see the heat sink fins for the cool part of the extruder. Close examination of the circuit board at the lower right reveals an optocoupler to detect the presence of filament.
http://american3dprinting.net/images/IMGP55251024.JPG
Next, removing the bearing allows the hobbed shaft to be retrieved. You can see that coupling feature I mentioned at the beginning of this thread. The 5th Gen extruder design puts the extruder motor inside the X carriage, and the hobbed shaft is passively driven.
http://american3dprinting.net/images/IMGP55261024.JPG
The idler and lever assembly can now be lifted out, and the hot end set aside, revealing the main circuit board. Examination of the components reveals that it is not "smart" in the sense that it has a microcontroller of any kind, just some active and passive circuitry. Fairly unremarkable IMHO.
http://american3dprinting.net/images/IMGP55271024.JPG
Examining the inlet side of the hot end reveals the blob of stuck filament. What had happened, was when I went to unload the filament, obviously this end was hot enough that it separated and left that blob in there. There was enough friction between the blob of filament and the inside walls of the hot end tube that loading filament would get stuck when it got to this obstruction. The actual offending blob turned out to be remarkably small, and a little tug with some pliers got it out.
If you look carefully, you will notice a small, shiny metal cylinder in the black plastic by my index finger. That is a magnet, used with the SOT23 hall effect sensor on the circuit board to detect the extruder's position. I forgot to mention in the post immediatly preceding this, you will see a small gauge spring near the top of the photo. This pushes the hot end downward from the rest of the extruder assembly, When bed leveling or setting Z home before each build, that hall effect sensor can tell when the tip of the nozzle just touches the bed.
Fortunately, reassembling the extruder went smoothly and it worked right off the bat after returning it to the X carriage.
http://american3dprinting.net/images/IMGP55281024.JPG
http://american3dprinting.net/images/IMGP55291024.JPG
http://american3dprinting.net/images/IMGP55301024.JPG
Jeff,
Brilliant, detailled honest review. You have done well to maintain the matter-of-fact tone of your posts as you work through the challenges of a new printer.
It seems to me however that you are experiencing many of the types of issues that those of us who bought a printer at the cheaper (much, much cheaper) end of the market have to deal with.
I seriously considered a MakerBot before ultimately buying a 3DStuffmaker and did initially think that perhaps I had made a poor choice when it turned out that you really need to dedicate some time to learning the quirks and odd behaviors of the lower end printers (what a n00b I was back then), but it turns out that the 5th Gen Ferrari is slower and just as temperamental as any other printer.
I struggle to see how MakerBot can justify classing this printer as a generational change when very little seems to have changed, certainly very little has improved.
Many aspects are different but hardly improved. Smart nozzle? That's more effort to clear a simple nozzle blockage than I have ever dealt with.
At this rate of "improvement" the Gen 6 wiii take 4 hours to prepare to print and 4 days to print a rook.:D
And also how they can justify the price tag!!! I have to say, after Jeff's tear down I am even less impressed. I already had a bit of a gripe with this machine for the fact it looked so sexy and was well out of my price range for what it offered.
Now I see the actual components they are putting in, and a hot end that while has handy features like auto sensing loss of filament etc, it looks really cheap.
Thank you for all this. Very detailed and factual. The X&Y belt arrangement to remove the motors from the gantry are a nice idea but the rest kinda kills it for me.
I'm sorry to say, but our Form1 spends most of its time sitting in the corner. We had an encoding issue with X printing twice as big as Y, and my guys kind of lost interest in diagnosing it. The material price also has us a little guy shy about messing around too much, plus the fragile nature of SLA resin. Don't mean to derail this Maker thread (hey you started it! :rolleyes:) but will end by saying we've got over 1000 hours on one of our Rep2's, another with around 700 hours... we've had each apart fairly extensively a couple of times, replaced most of the wire harnesses, repaired some of the pulley housings, but at this point have a lot more confidence in our ability to run a good part on the Makers than the Form1, so they're the first choice.
I could see us adding a third Rep2- was considering the new Gen but after your review there's no way.
Quick update - the replacement extruder came in yesterday. The fins on the heatsink are thinner and there's more of them. The plastic idler has a rib on it, the old extruder doesn't. I took pictures but don't have time to post them up right now. Install went smooth, it didn't fix the bed leveling issue. I haven't had time to call Makerbot tech support, will get after that on Monday about the bed issue. I did do a couple of builds with the new extruder. They came out fine, no better or worse than the original extruder.
One other thing - I was looking into Makerbot for Desktop (the replacement for Makerware), and in the setting profile, they have dropdown selections for other filaments just like Makerware does, but it won't let you select anything but PLA. I had a customer in yesterday who owns a 5th Gen and was interested in buying a roll of Makerbot flexible filament. It doesn't look like you can use Makerbot flexible filament on the 5th Gen.
And get this - the tab for temperature is inactive. You CAN NOT CHANGE TEMPERATURE from Makerbot for Desktop, at least not the current revision. Maybe you can slice it offline and sideload the memory? There's no removable SD card on the 5th Gen like a Rep 2, so I will see if that is possible. But I can't see building with nylon or PET or HIPs at the default 215°C! Oh yeah, that's another difference from the Rep 2. The Rep 2 defaults to 230°C (240° for the stretchlet preloaded from the factory in the SD card), and the 5th Gen defaults to 215°C.
Just got off the phone with Makerbot tech support. They want to have the machine back. I don't know whether they plan to repair it and return it, or replace it. The tech (Alex) said that when they designed the 5th Gen, they wanted to make it more of a consumer oriented machine, rather than the Rep 2 which was based on the original Replicator, aimed at hobbyists and tinkerers. As such, it isn't intended to be repaired in the field, other than replacing the extruder. So if my car breaks down I have to send it back to the factory? Bad paradigm if you ask me.
Hold time before speaking with anyone was 28 minutes, 32 seconds. I also got put on hold for around another 15 minutes after the initial explanation of symptoms and what had been done. Total call time 55:05.
I did extract some (potentially) useful information from him. I asked about the difference in default temperature, 215° for the 5th Gen and 230° for the Rep 2. He said that the Rep 2 does not necessarily "see" the nozzle temperature at the thermocouple, but that it was a little low and that was to compensate for it, and that the 5th Gen had a redesign where the thermocouple sees the actual nozzle temperature more accurately. Maybe so?
I asked about using Makerbot flexible filament, and he said that you can not use it in the 5th Gen as currently designed. Lovely.
I asked about the inoperable temperature tab in Makerbot for Desktop on the slice settings, and specifically said that we use nylon and PET in our Rep 2s and wanted to on the 5th Gen. He said that using any other filament than Makerbot brand would void the warrantee. He did mention that you can create a custome profile and save it to a *.json file and that is a way to specify a different temperature than default. I'll definitely have to try that.
I asked him how making the new machine build so much slower than the previous model was supposed to be an improvement? He said they continue to work on firmware updates. I personally speculate that there is some kind of issue with their acceleration profiles in firmware. I know for example, that building with the exact same GCode, our Z-Morphs are nearly twice as fast as our Lulzbots. But the Z-Morphs will definitely skip and slip on the X and Y when it tries to go too fast, and I have to slow things down sometimes to get a good build. The Lulzbots don't do that, and you can easily see that the acceleration is just smoother.
Here's a measurement I did this morning. I had written earlier that it takes "about" twice as long for the 5th Gen to heat up than the Rep 2, and had estimated the total heat time at around 5 minutes (preheat to 180° then final heat to 215°). I started both of them up at the same time, and watched the temperature indication. Bear in mind that the 5th Gen is only going for 180° and the Rep 2 is going for 230°, but I can still see how long it takes to get to 180.
At the 1:00 minute mark, the 5th Gen was 100° and the Rep2 was at 145°.
It took the 5th Gen 2:20 to reach 180° and the Rep 2 1:20. The Rep 2 Reached its 230° setpoint after 2:00 minutes.
I fiddled around with the *.JSON file today, it is just an ASCII text file you can edit in Notepad or equivalent.
Here is a snippet:
So basically it looks like there are a lot more things you can specify outside the Makerbot for Desktop environment.Quote:
{
"comment" : [
"This is a custom profile for MakerBot Slicer. Editing it will modify your",
"slice settings. For documentation on MakerBot Slicer parameters, see here:",
"http://www.makerbot.com/support/makerware/documentation/slicer"
],
"version":"3.0.0",
"layerHeightMinimum": 0.12,
"layerHeightMaximum": 0.27,
"layerHeight": 0.20,
"layerWidthMaximum": 0.4,
"layerWidthMinimum": 0.4,
"layerWidthRatio": 1.481481,
"bedZOffset": 0.0,
"startX": -110.4,
"startY": -74.0,
"startZ": 0.2,
"defaultExtruder": 0,
"rapidMoveFeedRateXY": 150,
"rapidMoveFeedRateZ": 23,
"minLayerDuration": 5.0,
"minSpeedMultiplier": 0.3,
"doDynamicSpeed": false,
"dynamicSpeedCurvatureThreshold": 15,
"dynamicSpeedDetectionWindow": 3.0,
"dynamicSpeedSlowdownRatio": 0.3,
"doDynamicSpeedGradually": true,
"dynamicSpeedTransitionWindow": 6.0,
"dynamicSpeedTransitionShape": 0.4,
"doDynamicSpeedOutermostShell" : true,
"doDynamicSpeedInteriorShells" : false,
"doSplitLongMoves": true,
"splitMinimumDistance": 0.4,
"extruderTemp0": 230,
"extruderTemp1": 230,
"platformTemp": 0,
"numberOfShells": 2,
"infillShellSpacingMultiplier": 0.7,
"insetDistanceMultiplier": 1.0,
"roofThickness": 0.8,
"roofLayerCount_disabled": 4,
"roofAnchorMargin": 0.4,
"floorThickness": 0.8,
"floorLayerCount_disabled": 4,
"coarseness": 0.0001,
"sparseInfillPattern": "hexagonal",
"infillDensity": 0.1,
"infillOrientationOffset": 0,
"infillOrientationInterval": 90,
"infillOrientationRange": 90,
"gridSpacingMultiplier": 0.99,
"solidFillOrientationOffset" : -45,
"solidFillOrientationInterval": 90,
"solidFillOrientationRange": 90,
i'm having multiple issues with our z-18. Support is somewhat helpful, depends on who you get. Its aggravating sitting on hold for 45min to get someone who cant help you.
They told me to make my own custom profile to solve problem of extruder paths being to wide and not making a complete layer on 100% infill. Well now for some reason after a certain height, the extruder head is not high enough and is dragged across the previous layer. it pretty much just remelts that already deposited plastic in the direction its moving. Interior hole look like crap, now exterior surface of circles are getting worse. Support wont respond to any emails, so I cant send pictures of issues. :mad:
Thanks for sharing that, exit.
My customer brought the original shipping carton to me on Friday, so I am getting ready to send this 5th gen back to get the bed leveling issue resolved. In the mean time I am continuing side-by-side builds of the exact same models with the Rep 2 for speed and quality comparisons. Again, all settings exactly the same for both the 5th gen and the Rep 2. I'll try to get that stuff posted up later today or perhaps tomorrow, depending on how much time I have.
Just to be totally clear, I did not start this thread to bash Makerbot. I am trying to be completely fair and factual, with everything fully documented so that anyone can do the exact same tests and get the exact same results, just like any science. To that end, if anyone wants the STLs I'm using, I'll be happy to share them, along with any pertinent info on settings such as scale, rotation, raft, support, etc. I really wish the 5th Gen worked better. I can't get the Rep 2 any more from my distributor, just the three 5th Gen models and the Rep 2X.
I do want to try to build with some Taulman nylon and T-Glase now that I know I can bump up the temperature in the *JSON file. I had previously sold a roll of nylon to one of my customers who has a 5th gen, and it would be nice if I could help him use that material. It will be interesting to see how to get it to stick, I use PVA glue directly on the Rep 2 acrylic build plate on the Rep 2, and the same PVA glue on the PE tape on the glass bed of the Lulzbot TAZs.
Attachment 2533 Attachment 2535
Well, after spending at least 1 hr each day last week on the phone, I think I finally talked to someone who knew how to answer all my questions. When I called Friday, I told him he wasn't getting off the phone until he looked at the pictures I emailed and I had an answer for causes.
1. Z-calibration - The new firmware 1.3 has a Z calibration problem on the Z-18's. This was messing up everything, from the raft to the printed model itself. They said new firmware should be out this week, maybe even tomorrow. Hopefully will fix this. Also should fix the circle problem.
2. Paths not touching - Still not sure on this one, but I had to make a custom profile to correct for the issue.
3. Raft problems - Sometimes the base layer pattern of the raft was to large and would leave a corner off (see bottom left of blue picture). It is showing the second layer of the raft as overhanging, and we all know this does not work out well. An example is in the corner of raft in the printed pic. Also, that is supposed to be a circle with 3 circles inside of it. I had to decrease the "square wave" pattern size to reduce this effect, you would think makerware would do this automatically to compensate for this issue.
Hi jeff,
Congrats for finding the *.json file, it took our experience of printing to another level, there are some unfortunate design problems in the way it is designed but it trully enable you to get much better prints.
We are using the custom profiles to print dana bloom jewelry and as you can see her jewelry do not have the regular engineering design so are not so simple to 3D print.
The problems that I hope someone in Makerbot (hopefully this thread has drawn their attention) will fix someday and will sure make my life and anybodies else who uses Makerware are the following:
Speed,Support,temp and raft should not be controlled in the profile or more exact: those setting in the profile should be able to be overwriten as in the standards profiles.
Regards,
Assaf
3d Printing Center Israel
Um - so a near $3000 (2899) printer doesn't have a heated bed ?
Wow. And looking at all the other stuff - I'm glad I bought a clone of makerbot replicator and not an actual makerbot.
As far as I can see, everything on mine is better or easier to use and fix.
Love the review - totally even handed and just goes to show that spending a lot more money doesn't necessarily get a better printer.
One thing - I've tried adjusting the json files for the default templates in makerware 2.4. I simply wanted the default print head to be lefty, not righty.
Didn't work - I only changed a 0 to a 1. Just didn't effect the software at all. Which .json file should I be changing to make a change in the default templates ?
I think I tried every file I could find.
Hey Assaf, glad it was helpful. I checked out your web site, but I can't read Hebrew :(
I honestly don't know how much can be "fixed" with this 5th Gen design.
Certainly my bed leveling issue is likely simply a defect in this particular machine, rather than a design flaw. I do see a philosophical problem with their approach to these 5th Gen models, where the user has less and less control over the build parameters, and worse, the philosophy of "no user serviceable parts". If they are going to go that direction, they better have a nationwide network of repair centers in place, ready to turn around the machines very quickly.
Another disturbing bit of news is an email I received from my Makerbot reseller team. I don't have it in front of me for the actual verbiage, but they are REALLY pushing Makercare, which is basically like those pricey extended warranties they try to sell you whenever you buy an appliance or TV and so forth. The latest paradigm is that you HAVE to buy Makerware to get phone support, and that if you don't you can only get email support, and furthermore if you don't buy Makercare, you only get any support at all for 60 days after purchase.
With those kinds of strategic decisions, it bodes well for Makerbot's competitors.
Hi curious aardvark, I am sorry but I do not know the answer to that question. It seems possible that the only *.JSON file that you can edit might be custom ones rather than the defaults? Plus it isn't clear that the slicing engine is capable of making a mirror image, you'd probably be better off doing that in whatever software created your 3D model.
As for the price point, I have a suspicion as to why it went higher, when on the outside it looks like their gross COGS are pretty much the same as the Rep 2.
Having worked in electronics product manufacturing for many decades, the new product team would sit around in meetings and make decisions such as price points. One of the tidbits I picked up from the marketing folks is as long as your product was somehow perceived as newer, better, unique from the competition or what have you, you could introduce it at a higher price point and that would set the perceived value of the item. It was mostly psychology, and had little to do with actual costs. A practical aspect to this was setting more margin so as to allow for discounts for distribution and commissions for salespeople and manufacturer's reps.
Here are a couple pictures showing the new extruder side by side with the old one. The only outward differences are that the new one has more fins on the heatsink and they are thinner, and the plastic idler has a slightly different profile.
http://american3dprinting.net/images/IMGP55471024.jpg
http://american3dprinting.net/images/IMGP55481024.jpg
Next are some more side-by-side builds. Below is a scanned customer. Her mother in law was going to give that to her husband for their first wedding anniversary. When you drop the full size scan into Makerware, it is well over 1,600mm tall, so you click on "maximum size" in from the scale function and it goes to 150mm tall. You can see the build time at 1 hour 53 minutes.
http://american3dprinting.net/images/IMGP55361024.jpg
Oddly, in Makerbot for Desktop, when you click "Maximum Size" for scale, it scaled it to 145mm tall, even though the published specs say Z can go to 150mm. Huh? Anyway you can see it took 2 hours and 46 minutes.
http://american3dprinting.net/images/IMGP55381024.jpg
Next, side by side comparison of build quality. Note that I am now using genuine Makerbot filament on the 5th Gen, and aftermarket Filament Central Chinese filament on the Rep 2. As always, the 5th Gen is on the left and the Rep 2 is on the right.
http://american3dprinting.net/images/IMGP55391024.jpg
The artifacts on the inside of her boots on the Rep 2 model seems likely caused by a lack of extruder retraction when going from one leg to the other. The 5th Gen, by virtue of the extruder design always does a Z-lift whenever it does a retraction, but can do the Z-lift much more quickly because it is using the extruder motor rather than the Z-axis motor. All our other printers have enough of the extruder and hobbed wheel (or bolt) visible enough that it is obvious when the slice has called for a retraction, however you can never see that on the Rep 2 because the hobbed wheel is hidden behind the heatsink. On the 5th Gen it's easy enough to see because of the Z-lift.
http://american3dprinting.net/images/IMGP55401024.jpg
Newest problem is the auto Z-home. As previously mentioned, before each and every build it goes through this time consuming bit of finding Z-home, and now it is starting to get intermittently wrong. In other words, some builds it gets it right, other builds you get this from too much Z, and in other cases you get too little Z and the extruder can't extrude because it is pressing too hard on the build surface. You get that familiar "click click click" of not being able to pull the filament through.
http://american3dprinting.net/images/IMGP55451024.jpg
Well, I packed up that 5th Gen on Tuesday Sept 9th and dropped it off for shipping yesterday the 10th. I sent along this letter:
I will certainly share whatever we learn. In the mean time, I still have a huge backlog of posts and images to add to this thread. The scanned customer above was actually done on 8/18, so we've got several more weeks of builds and data to share.Quote:
September 9, 2014
Ref: case #00222903
Dear Makerbot Service,
This 5th Gen has a couple of issues. The main and most obvious one is that the build plate does not have sufficient mechanical adjustment to level the Y axis. The front is too low and the back is too high. You can cheat and get it to the second step by pressing upward on the front of the build plate until the light comes on. Then it has an odd issue with the X axis leveling. It wants it to be crooked – with the right side too low and the left side too high. At first I thought that the X axis gantry might also be crooked and it might be compensating for that, but this isn’t the case. There is only one sweet spot you can build on, in the center of the X axis and a little toward the rear on the Y axis. To make anything bigger than 4x4 cm or so footprint and you have to use a raft.
A second issue that has become increasingly frequent occurs intermittently. At the beginning of each build when it tries to find Z home, sometimes it gets it right, and other times Z is too great and it will end up half or even a mm or so above the bed. Obviously there’s no way those builds stick. Alternatively it will be too close, and smash the nozzle down onto the build surface preventing any extrusion. This is characterized by the familiar clicking sound of a filament jam or tangled spool.
It is puzzling why the designers chose to recalibrate Z home each and every build, and also puzzling why it takes so long. All our other machines, including our Rep 2s, Z-Morphs, Type A Machines and Lulzbots only have to be set once unless you fiddle with the plate or change extruders. It is also puzzling why bed leveling requires the filament to be removed, yet it can somehow sense Z home with filament loaded, at the beginning of each build.
Note that this machine has a new extruder. When I first called on case #00215194, I explained the bed leveling problem and it was somehow diagnosed as a faulty extruder. This extruder functions identically to the original one.
Please call at the number below if you have any further questions.
Sincerely,
Jeff Zepp
Here are some more head to head comparisons of the exact same build. This is a 100mm bust of a scan of one of our customers. In the 5Th Gen it takes 9:06.
http://american3dprinting.net/images/IMGP55571024.jpg
Same exact build, all parameters identical, it takes 5:47 in the Rep 2. BTW, if anyone's wondering why I'm building with him tilted up like that, it's so I don't have a bunch of support to have to clean up under the brim of his hat, all the support is on the back where it is less noticeable and easier to remove.
http://american3dprinting.net/images/IMGP55591024.jpg
Here they are side by side. Notice the voids in the 5th Gen build on the left! And again, this is with genuine Makerbot filament and a brand new smart extruder.
http://american3dprinting.net/images/IMGP55821024.jpg
Next we have a scan of a carved wooden bear, to which I added a little sign that my wife thought was funny. The builds are scaled to 60mm tall. You can see the 5th Gen took 1:13.
http://american3dprinting.net/images/IMGP55661024.jpg
Same build, same parameters in the Rep 2 took 50 minutes:
http://american3dprinting.net/images/IMGP55651024.jpg
Again, side by side, the 5th Gen build seems to have some voids. Also, it isn't bridging as well under the sign, although it is hard to see from this angle. Note that this particular build is NOT Makerbot filament, it is from Filament Central, however I am using the exact same spool of filament on each of these two builds.
http://american3dprinting.net/images/IMGP55681024.jpg
I have lots more pictures, with build times, etc., however after a while it kind of comes from the department of redundancy department. All of them show pretty much the same thing - the 5th Gen is slower and seems to have a lower build quality than the Rep 2, all when using the exact same settings on both machines.
One other important difference is that I am using Makerware for Desktop (version 3 point something) on the 5th Gen, because that's all that will run it, and I am using Makerware 2.2 point something on the older 32 bit laptop that's running the Rep 2.
When the 5th Gen is returned, I will certain repeat a couple of these builds to see if there is any difference whatsoever with the initial results. For the record, the firmware in the 5th Gen is 1.3.1.143, and was the 3rd update we had in the month or so we had the machine.
The Makerbot web site insists that the 5th gen has wireless capability, however I never discovered how to use it. I didn't try the ethernet, the machine was located too far from my router. I never figured out how to use the camera, and never figured out how to store slices into the 5th Gen internal memory (which presumably replaces the good old SD card on the front of the Rep 2). Perhaps after the machine is returned, I'll ping tech support again and see what they can share with me.
I got the following email today:
Quote:
Hello Jeff,
The repairs to your Replicator 5th Gen. are complete, and is currently being prepared to be shipped back to you. With your bot, you will receive a detailed service report along with test print(s) made on your machine. If you have any other questions or concerns feel free, to contact us.
Best,
MakerBot Support
ref:_00D30W6XV._50013l6IFg:ref
Hi Curious Aardvark (I am learning English all the time, thanks for this cute animal name ;) )
From our experience, you do not need to play with the default head. just set your settings for the head you want and choose using the UI of the makerware what toolhead to use to print your objects.
I think this is the right way to handle this issue and not choosing in the profile the toolhead to be used.
Assaf
Hi Jeff,
My site is in both languages: Hebrew and English, so you are welcome to visit us again :)
I think this link will get you directly to the English version.
About Makerbot strategy, I am not sure it would be applied to all countries, I believe that at least in Israel the regulation authorities will not allow such behavior but I defiantly agree that Makerbot Gen 5 3D printers do not look as promising as at least I expected from them to be after they were bought by Stratasys.
Really great thorough review. Thank you very much!
You're welcome, and thank you.
BTW, we got the 5th Gen back from Makerbot, but they asked me to have the printer's owner be the one to unpack it, so I'm waiting for him.
I was down at MCAD for a 3D printing conference last week. MCAD is the Denver area authorized Stratasys reseller, they sell all the enterprise class machines like the Fortus and Objet and the low end ones like the Uprint and Mojo. I got chatting with the sales guy, and he (always trying to make a sale) mentioned that Stratasys has lowered the price of the Mojo to $6k. It used to be $10k. One of the things I have heard "on the street" is that Makerbot Rep2s were KILLING Mojo sales, so you can see where the price increase on the 5th Gen combined with the drop on the Mojo were probably tangentially related.
So the 5th Gen owner came in yesterday and I had him do the honors of unpacking it.
See the linked PDF of the service ticket, but I will also summarize:
HUH???? First of all i KNOW for a fact that the only tape that was on that machine when I packed it up was the blue painter's tape on the build plate. What the heck is a "timing belt" and what would that have to do with leveling the build plate? What is a Z-stage assembly? Where is this lower left pulley? The only pulleys that are visible are around the TOP of the machine!Quote:
Reported problem: Unable to Level the Plate
Observed problem: Tape on lower left pulley
Action Taken: Replaced idler pulley, timing belt and belt clip; replaced xy gantry bracket; replaced z-stage assembly
Time to repair: 5
Anyhow, we set the machine up and proceeded to attempt to level the plate. The first thing the machine did was slam the extruderall the way to the left and then all the way to the back of the machine with this horrible sound of stepper motors attempting to move an immovable object. I quickly reached around the back and cut the power.
Try again, this time it did what it was supposed to do and went to the middle of the plate and found home. This time there was sufficient adjustment to get the LED to illuminate and say that the plate was level.
So to test it, I dropped a small object into Makerbot for Desktop, duplicated it 4x and place one copy on each of the four corners of the build plate. I loaded some genuine makerbot filament into the extruder and off we went (after the usual 15 minutes to heat up, find z home and then finish heating).
The plate was not even close to level, or properly homed. Two of the objects didn't come close to sticking, and the other two were jammed against the plate so no filament could come out.
So we went through the leveling process again (which takes a good 15 minutes in its own right, and of course you have to unload the filament for it, then reload when done).
Tried the 4 object build again, with the exact same result.
Next I simply observed where it was too high, too low etc. and after about 5 iterations and about an hour and a half of fiddling around, got it to where all four corners were about right. There is an additional problem though, in that whenever you go to start a build and it finds a new z home, it isn't very consistent, so sometimes the builds will stick and other times they won't, and yet other times the nozzle will be smashed up against the tape and nothing can come out and you hear that familiar clicking sound inside the extruder.
I had written about that issue (see prior post's copy of the letter I sent), but obviously either this wasn't addressed, or it is an inherent design flaw. I would speculate that this is why all the builds that come prloaded into internal memory are done on the very-forgiving Makerbot raft.
Jeff,
Thanks you for the excellent review - it's been in depth, illuminating, and more than a little disturbing. My company purchased a Replicator 2X about one year ago, but we passed on adding a Z18 because Makerbot refused to confirm (or deny) that it would run ABS. Your review is confirming that as a good decision.
I feel like Makerbot has lost its way. Our Replicator 2X remains a fine machine, but it certainly seems that the 5th generation machines are a step backward. I've also been quite uncomfortable with the changes to Makerbot Desktop. The slice function seems to be getting slower, the support for alternate materials seems worse, and the screen realestate given up to branding and on-line functions is disappointing.
Along with quite a few others who are reading this review, we look forward to any other information you care to share about the experience. Thanks again!
Wow, thanks again Jeff..
"Observed problem, tape on pulley"
Wow, what a load of BS, this is becoming very reminiscent of a fancy car that has too many automated features being rushed out to market and causes customers a load of headaches. Like a car, they can cost a small fortune.
The only difference here, car manufacturers have to recall the cars mainly for safety reasons, where as Makerbot can give you the run-around, literally for as long as they like and there isn't anything anyone can really do about it.
A warranty means very little when the company both denies the problem and also doesn't even look into fixing it, which seems like what's happening here.
With all that money, all that backing, all those people... How can they get this so wrong?
Hello Jeff
Thank you for the excellent review.
I am a new makerbot 5th gen printer owner. We bought it just this month, so maybe the company looked at your review and others on the web and decided to do some modifications. I dont know. I do know that the replicator we bought has not given us any more trouble than anything else new would give to a newbie. Other than the completely non-existent customer support.
We unpacked the printer, and started fiddling around it. The first step was leveling of the bed. Went through without a hitch.
Second was loading the filament. Now we got one from makerbot with the package, and it went into the attached spool holder without any difficulty.
For different color, we went and bought us some PLA filament from amazon. We tried that, and the spool did not fit in the holder. Does that mean that makerbot absolutely will not allow to use non-makerbot accessories? Totally. This voids the warranty too. For this, I tried to get some support both from makerbot and the spool manufacturer. Zilch. No support from the manufacturer. And makerbot, you get pay per ticket support, unless you buy makercare. There was this excellent google group for makerbot, which they closed off. I did sign up for the new makerbot users group on google. This will be my support I guess.
We have been printing with this almost every day since it has come, and there have been some minor issues.
After 3 or 4 prints, the filament refuses to attach itself, and we have to change the blue tape we use on the bed.
Sometimes, changing filament, the filament refuses to load. Went back to google, found that one has to push the filament really hard, and keep pushing until it starts extruding. Not what it says on the LCD - to just push the filament until you feel the extruder pulling it in.
The spool holder does not accommodate any spools other than makerbot.
I have bought a ninjaflex filament, but havent yet figured out if the makerbot will print that without problem or not. I read that Rep2x, there is an extruder accessory which helps in printing with flexible filament. Any help or experience would be appreciated.
This has been my experience till now, since we received the machine around the 24 or 25th of september.
thanks