Fix up old machine, or make/buy a new printer?

RepRap Prusa/Mendel V2/V3

My 3D printer is a total basket case. My question is, can I rebuild it into a reasonably decent functional machine, or am I better off buying a whole new machine? Would this machine be on par with newer machines if fixed up properly? If I can fix it up, what parts do I need to replace, and with what should I replace them? My goal is that I want to be able to print random replacement parts for stuff (brackets, gears), mainly centering around my vintage computer collection and other hobby projects. Biggest "accuracy" issue for me is that I will need to create housings for small projects, and I'll need screw holes in the housing to line up with the mounting holes in the PCB. I'll need to make boxes and lids, and the lip of the lid will need to line up with the box so it can be screwed together nicely.

The printer is a RepRap Prusa/Mendel V2/V3. It has a horizontal carriage assembly. Originally designed for the filament feed motor to be mounted directly on the carriage, however has been modified with a bowden type remote extruder. Hot end is some no-name off of ebay, held in place by a piece of paint-stirring stick (because the bracket was made for a direct drive not bowden. It's got a genuine arduino mega 2560, and RAMSP 1.4A board. Yes, it says "RAMSP" not "RAMPS" Heated bed, all steppers are "Kysan 1124090". Frame is thick acrylic, Plexiglas, or similar. "Geeetech" LCD display with rotary encoder, speaker, and momentary switch.

What are the problems:

- Weak gear in servo motor occasionally causes arm to partially extend for z-axis auto level switch, causing hot end to ram into and shatter glass bed. (I think some sort of fixed-mount optical sensor would be better.)

- Z-axis motors make "jamming" sound, even with threaded rods disconnected from motor. I've tried adjusting the little driver boards on the RAMSP. It works more reliably at extremely slow speeds, although never 100% reliable.

- Intermittent yet frequent random heater error messages (thermal runaway, etc...) causing print job to halt.

- Filiment jams up inside of hot end causing print job to be ruined.

- After completing a layer, the printer does not correctly increment the carriage up "one notch" (z-axis) causes filament to be dispensed too far above the object being printed. By about a quarter inch worth of layers, the item being printed is a mess and ruined.


I would imagine the thing to do is replace the carriage assembly with one made for a bowden filament feed, and perhaps replace the arduino/ramsp board with something newer. Replacing the carriage could include an optical Z sensor, and proper cooling for the cold end.

So, is this machine worth putting a few $$ into? Do newer designs spin circles around it? Time to toss it?

Thanks!

-Stu

... depends mostly on your skills - some friends optimized/rebuilt some of their old printers and print now with more than exellent quality and speed

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Ok. I'm open to fixing it up. Any specific details you can share on how they accomplished that?

Thanks

-Stu

Genuine and it says RAMSP?

I have a deal you may be interested in... I'm partnering with a Prince from Nigeria to free up some funds in a Swiss bank account for his family that was wrongly exiled from his home country. All we need is...

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Cynic... But yes you're going to need to spend some money if you want a better machine at the end of it. But use the parts that work, motors, cables, heated bed, hotend, controller (ramps isn't great but it does enable you to print stuff, you can always transplant a smoothie/Duet into it if you come up with something decent).

I've no experience with mendels, bit before my time, but you can use most of the vitamins to make something more modern like an i3 mk2, or a corexy with addition of some new frame components. Anythings possible I'm designing a delta based on an cylindrical, acrylic fishtank!

Edited 1 time(s). Last edit at 12/23/2017 08:37AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
Stu
Ok. I'm open to fixing it up. Any specific details you can share on how they accomplished that?

Thanks

-Stu

From your questions and the fact you already somewhat failed, I would advise you look for some decent working out of the box 3D printer to achieve your goal, that is make these boxes for your vintage machines.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

You're experiencing multiple problems with the machine all at once and it looks like a huge undertaking to fix it. But think back to the first day of geometry in high school. They handed out a text book and you flipped through it and your first thought was panic: "I'll never learn all this stuff". But by the end of the year, you learned it and it turned out to be pretty easy because you learned it a little at a time. I suggest you approach the printer the same way. Bite off little pieces. Pick one problem, understand it thoroughly, and fix it. Then move on to the next problem. Before you know it, all the problems will be fixed and the machine will be printing. Then you can decide if it prints well enough, or is reliable enough, or should be modified or replaced.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

As another hobby, I restore vintage saxophones. The older, the better.
It is much more satisfying to play with those horns and give them a second ( or third ) live.
But it takes much more skills and stamina to get there.
Be prepared to spend way more time and survive many failures until you finally "play in tune".
As a collector of vintage computers, you seem to be the guy who knows, what I mean...

But you could as well look for a mech. kit only and borrow the rest from the existing printer, but it seems even your electronics are not 100% reliable.
So in the end you'll replace most of the parts anyway. Except the acrylic frame, which isn't exactly the part worth keeping...

This is going to depend on a number of factors.

For me, my i3 was never good. It was always uinaccurate, but I thought that I could make it better. The problem was with the tools that I had to make it better, which was largely the printer itself. I couldn't make the parts that I needed in order to make it better. I ran into a limitation on what I could do, and what I was willing to spend in terms of time and effort to do it. I packed up the printer, and stored it for a year and a half.

This fall, I pulled it out again. I looked at what it could do, and what I could improve without new parts. I fired up my CAD program and started looking at new parts. Something that I felt would give me a higher accuracy of motion, so that I could perhaps make the whole thing better. More rigid, more stable, more repeatable. It turns out that it's a daunting task, and eventually I found that I was either going to have to pay someone for a bunch of custom parts (3D printing/CNC milling service) or else I was going to have to buy another printer, and hope that it was better than the one that I got 3 years ago.

I bought another printer, and as I type this, it's printing parts for my old printer.

The new printer is nice to work with, but I'm stubborn, and since it's a hobby, I'm still fixing the old old one. Besides, I didn't read the fine print when I ordered this one, and it's missing a heated bed. I thought that it had one, but it doesn't. It will as soon as I set up parts to do it though.

A lot of great responses! Thank you all very much.
So, I'm thinking at this point, because of the random intermittent heating errors, the motor drive quirkyness, etc...
I think the best first step is to get new electronics. I doubt anything is wrong with the motors themselves, and that the problem is probably in that lovely "RAMSP" board, but thats just an assumption.

@O_lampe As a tuba player, I know exactly what you mean. Twenty years later, practice still starts with the same long tone exercises that practice started with twenty years ago. Only difference is now I play some of my etudes from memory, so I can't use the "I forgot my book" excuse that worked when I was ten years old! I've probably watched 40+ hours of youtube videos on 3d printing, and definitely plan to watch many more. No problem with patience here

I'd bet if I hooked my oscilloscope up to each of those Z axis motors and looked at the pulses, I'd find that they're either not in phase, or the rise/fall is weak or not reaching the correct peak. And that would be the controller. My guess is those little removable amplifier boards for each motor.

So, my question now is, what is the best thing to replace the electronics with? Keep the arduino and get a different shield? Replace the arduino with one of these new all-in-one boards? Other suggestions? Don't worry if your suggestion costs a bit, I'm not looking to be super cheap, I'm looking for quality and reliability.

Thank you ,

-Stu

Edited 1 time(s). Last edit at 01/16/2018 01:27PM by Stu.

Some things in your description give me some concern about the mechanics of your printer:

Quote

- Filiment jams up inside of hot end causing print job to be ruined.

It sounds like your hot end needs to be replaced. If you only print PLA then a PTFE-lined hot end is more forgiving than an all-metal one. If you want to print at high temperatures, then you need an all-metal hot and and the E3Dv6 is the standard choice.

Quote

- Intermittent yet frequent random heater error messages (thermal runaway, etc...) causing print job to halt.

The cause could be as simple as a bad thermistor connection.

Quote

- Z-axis motors make "jamming" sound, even with threaded rods disconnected from motor. I've tried adjusting the little driver boards on the RAMSP. It works more reliably at extremely slow speeds, although never 100% reliable.

and

Quote

- After completing a layer, the printer does not correctly increment the carriage up "one notch" (z-axis) causes filament to be dispensed too far above the object being printed. By about a quarter inch worth of layers, the item being printed is a mess and ruined.

Your Z drivers may be overloaded (RAMPS drives them in parallel, which is usually a bad choice); or it could be as simple as your Z speed and/or acceleration settings being too high. Can you tell us the make and model of the Z stepper motors?

If you want to change to better electronics, I am biased (see my signature); but if you think the mechanics of your printer are basically sound then I suggest you get a modern 32-bit controller such as the Duet WiFi or Duet Ethernet. There will be a learning curve to get used to the new firmware, but you will be able to change the configuration and tune the settings very easily, and even update the firmware when you want to, all through the web interface without having to edit the firmware source code, recompile it and re-upload it. Plus a pile of other features (depending on our machine properties) such as motor stall detection, resume after power fail, bed height map visualisation in the web interface, super-quiet TMC drivers, automatic sync of multiple Z motors, and a lot of expandability.

For the bed levelling, you could use one of my IR sensors if your bed surface is compatible. See section "Bed surface" at [miscsolutions.wordpress.com]. The Precision Piezo sensors are also very accurate and less dependent on bed surface.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Sounds like a lot of the issues are electrical/controler related. You can buy a complete Ramps kit for £30. Set of motors another £30. Then you need to do some fabrication and maybe get another hot end.
Noticed on Ebay you can get a Prusa clone for £109 now. Thats with heated bed but only a ply frame.
That seems good value to me since you get everything that should work. And its all new (belts, bearings etc)
In your case you can use it to print improved parts for your old machine. You can use your acrylice frame if you don't like the ply version.
I know some one will say you can buy the parts cheaper or you don't need all the parts but there is something to be said for a kit where the parts are more likely correct and all arrive together (preprogramed with instuctions - maybe).