Anyone working on any i3 revisions/upgrades to reduce maintenance?

Don't get me wrong, I love that I've got an i3 and it works for the most part. However I'm getting really tired of all the constant maintenance needed just to use it. Silly little things like having to check nuts are tight enough, rebalance the z-axis motors every now and then, etc.

It seems that the machine has an inherent flaw in that it needs constant work to keep it running in tip-top condition, and seems a lot of things were just completely forgotten about during it's design, such as the endstop locations, meaning we're all having to use little 'clamps' on the smooth rods, or glue magnets to bits of the frame for hall-stops.

It got me thinking about either moving to something better designed such as a rostock, or working on some sort of upgrade to increase the, quite frankly terrible frame design on the i3 (specifically the connection between the frame and y-axis), as well as lightening, and stabilising the x-axis by possibly using aluminium box sections instead of the vertical 2-bar system it uses.

I was curious to see if anyone else had come up with any decent frame, or axis upgrades to improve reliability and reduce maintenance (i.e reduce the number of nuts and bolts). I've seen a few fairly minor upgrades on thingiverse, but nothing on the scale I'm thinking.

Edited 1 time(s). Last edit at 12/18/2013 04:25AM by RickM.

This highly decreases vibration and sound [www.thingiverse.com] Without a vibration I do not see how nuts can get loose. I have printed around 80 hours now and have not adjusted anything except for leveling a bed.. for 2 times I think and it did not even need it.
I print at slow speed of 15-45mm/s so that might be a reason for reliability.
Thats on single aluminium plate prusa i3.

There are hundreds of different upgrades for the aluminum frame i3 out there. I can't even count how many different printable end stop mount variations I've seen. The main plate bracing fix has been out for almost a year now and it shows up in a number of kits. Nylock nuts and / or Locktite are your friend for nuts that come loose. Auto bed leveling takes care of a lot of minor shifts, even more so if you carry your printer around a lot.

Indeed if you are trying to print for hours at 500 mm/sec you will have problems. If you have not cleaned and lubed the machine in months, you will have problems. I believe that's going to be true of any of these printers. I believe you sacrifice print quality when you speed things up much past 100 mm/sec. You can indeed print much faster, I don't like the result when you do.....

I've been printing with my Prusa i3 for a while and it works fine most of the time. The only maintenance I do is to lubricate the rods and retighten the xy-z connection about once a week. What exactly are the problems you run into?

I think I'd toss in a quick wipe down of the rods once a month to the standard maintenance routine. The M-5 thread rods / nuts and smooth rods / linear bearings will all last a lot longer if you keep dust and dirt off of them.

Quote
myohmy11
I've been printing with my Prusa i3 for a while and it works fine most of the time. The only maintenance I do is to lubricate the rods and retighten the xy-z connection about once a week. What exactly are the problems you run into?

So far I've had a few, although some of them can be put down to user error:

- Y-Axis coming loose from the frame during a long print if the surface isn't perfect
- Head crash due to endstop slipping down the z-axis smoothrod (has this happen twice, it's a bad part that needs replacing so not really a machine issue)
- Pretty unreliable endstops overall, especially the x-axis one which seems to have no decent place to mount it whatsoever. I've resorted to using cableties and since doing that, it's been way better than the handful of printed endstop mounts I've used, including the ones that came with my first set of printed parts from 3Distributed.
- Two sides of the z-axis very slowly uncalibratng themselves. I did however know that this was a minor issue with many of the Mendel's prior to starting and its not a major issue, but a small annoyance.

I've not had so many issues lately. I'm trying out a few new parts right now and so far the results are good. I've also just finished moving over to a budaschnozzle by gadgets3d, replacing a rather crappy cheap J-Head.

I guess I'd just like it to work flawlessly, which I know I shouldn't. It just seems that things go wrong with it far too often. When it works, it works fantastically well, and even before it was calibrated was giving pretty decent prints. My new parts for the budaschnozzle were all printed by the i3, and they are very good quality - way higher than I was expecting.

One thing I have noticed is that the M5 threaded rods on the Z-axis do tend to wobble a bit. I was able to limit this slightly by tightening up the connection between them and the motors, but it seems as though they may be ever so slightly bent. Would it be worth trying out some new rods?

Have you looked at the Mendel90 from nophead? The nylon lock nuts do away with repeated re-tightenings. In the last 6 months of heavy use I have not spent a lot of time on maintenance. I rebuilt the y-idler and recently changed to an E3D hotend which is wonderful. The machine now sits on a 5cm thick block of styrofoam which cuts down noise and vibration considerably. When the weather got cold I made an enclosure to deal with warping issues.

Quote
RickM

Quote
myohmy11
I've been printing with my Prusa i3 for a while and it works fine most of the time. The only maintenance I do is to lubricate the rods and retighten the xy-z connection about once a week. What exactly are the problems you run into?

So far I've had a few, although some of them can be put down to user error:

- Y-Axis coming loose from the frame during a long print if the surface isn't perfect
- Head crash due to endstop slipping down the z-axis smoothrod (has this happen twice, it's a bad part that needs replacing so not really a machine issue)
- Pretty unreliable endstops overall, especially the x-axis one which seems to have no decent place to mount it whatsoever. I've resorted to using cableties and since doing that, it's been way better than the handful of printed endstop mounts I've used, including the ones that came with my first set of printed parts from 3Distributed.
- Two sides of the z-axis very slowly uncalibratng themselves. I did however know that this was a minor issue with many of the Mendel's prior to starting and its not a major issue, but a small annoyance.

I've not had so many issues lately. I'm trying out a few new parts right now and so far the results are good. I've also just finished moving over to a budaschnozzle by gadgets3d, replacing a rather crappy cheap J-Head.

I guess I'd just like it to work flawlessly, which I know I shouldn't. It just seems that things go wrong with it far too often. When it works, it works fantastically well, and even before it was calibrated was giving pretty decent prints. My new parts for the budaschnozzle were all printed by the i3, and they are very good quality - way higher than I was expecting.

One thing I have noticed is that the M5 threaded rods on the Z-axis do tend to wobble a bit. I was able to limit this slightly by tightening up the connection between them and the motors, but it seems as though they may be ever so slightly bent. Would it be worth trying out some new rods?

1) The end stops can be mounted to printed parts that screw down solidly to the frame / carriage / end block / thread rod (depending on which parts you decide to print). You don't need to hold them in place with tape or bubblegum.

2) The Z axis rods *must* move at their tops (unlies you really like z axis banding) Mounting them at the bottom with cheap vinyl tube works amazingly well if you are careful setting it up.

3) Nylock nuts won't come un-screwed. McMaster Carr will sell you a boatload of them for very little money.

4) Auto bed leveling will take care of any slow creep issues.

There are existing solutions to every issue you have brought up. You simply need to implement them. You bought an i3 early in it's design life and got the early adopter version. With the fun of having it when it's just come out also comes the need to tweak things later on.

Many of those issues are a tradeoff of the low cost, minimal design for DIY builders. For a more robust machine braced or extrusion based designs work quite well. I've got a Mendelmax 2 now in addition to the i3s and while it's more robust mechanically in every way, it's also considerably more expensive to build. At wholesale pricing the parts for a complete MM2 are more than a retail i3 kit. I consider these as machine tools and as such give them an inspection periodically as I would any of the other tools in my shop.

I was hoping for a tradeoff when I bought an aluminium plate Prusa i3 but here is what I have to say... My well calibrated i3 prints in quality as good as Stratasys uPrint SE. Price difference between these printers is 20 times. Stratasys makes you buy material in their spools and that costs 10 times more than normal 3D printing ABS (atleast that is a price my university pays and maybe thats for 2kg total- 1kg of ABS and 1kg of support). Uprint SE even makes same wavy looking surfaces after change of direction along X and Y axis perimeters as Prusa does.

Edited 1 time(s). Last edit at 12/22/2013 03:36PM by Edvardas.

The tradeoff in using a Reprap vs. Uprint is time to time to build, implement and train. In a larger commerical environment it's quicker to get up and going by buying a complete package backed by a support contract. The Uprint is also a couple years old at this point and the disruption in this segment could mean that whatever markets were using the lower end (US$10-20k) machines will start using something like a Makerbot and increasingly Repraps implemented by specialty contractors. It's just been in the last year or so that the acceraltion of the low end machines has hit a price/performacne ratio that rivals entry level commercial machines. Many businesses also lease equipment so a commercial machine is attractive to them even with a premium price. There is a buzz in that smaller shops that could afford the lower end commercial machines are looking to less expensive Reprap models but I don't see them doing it with inhouse staff at least at first. Over the last few months I'm seeing more interest and some customers that will pay a premium for to have a Reprap built then have someone like me train them and help maintain the machine at least in the near term. I expect the trend will accelerate when the sintering patents expire and cause the same downward price disruption now being seen in the FFF/FDM segment.