Welcome! Log In Create A New Profile


Porting Prusa's reliability to a custom made printer.

Posted by lightyears101 
Porting Prusa's reliability to a custom made printer.
August 10, 2021 06:13PM
When it comes to Prusa and their Einsy Rambo Board running Marlin. What % of Prusa safety is based strictly on the board design vs. stock Marlin safety features that control any board with maximum safety vs. Prusa's version of Marlin with added safety.

If I wanted to build a custom printer with max precision on the gantry so I never second guess my self if my prints are bad because of gantry or electronics. so lets say my budget is in the $3000 range. With built volume 350mm X x 250mm Y x 150mm Z.

So if it's belts ok, if it's high precision linear actuator - ok too.

Lets say I need to print only at 220C not faster then 60mm /sec, using only one or two filaments so the settings are always the same. Printing on cold bed with glue stick, so no need for heated bed at all.

But printing is almost 24/7 in some cases prints would last 36+ hours, so must print unattended.

If I take Prusa electronics including their steppers / PSU which are selected for their gantry size and torque requirements so might not be appropriate for my set up.

If that Rambo board run on stock Marlin 1.8 or 1.9.... without any bells or whistles, just enter the SD card and print off it, and no auto-leveling no, no pinda or bl touch ...just simple one trick pony approach.

I'm not an expert on electronics but it seems to me that Rambo board has 3 fuses, which is more then most boards I see.

Prusa is known for reliability. Where does it come from...the board? I see people praising Duet2 on custom 3D printer built, but it seems to me it requires the user to make all the decision on safety and other characteristics that reflect both safety and quality of prints, this knowledge is hard to find even if you paid for it. Since Prusa prints well on an 8 bit board and if I don't need 32 bit board or my main focus is reliability not fancy features. Is Rambo + stock Marlin 1.8 or 1.9 as reliable as Prusa's version of Marlin. If not, how easy is it to take their version, and remove all the extras and only keep the safety parts, or the other way around copy the safety parts into a stock Marlin 1.8 and off you go the same reliability.

I guess the main question here is on money & time of R&D, can this reliability be achieved without paying an expert on firmware/ electronics / software development to look into Duet2 or other boards to produce a very reliable 3D printer electronics that is a one trick pony set up.
Re: Porting Prusa's reliability to a custom made printer.
August 11, 2021 10:18AM
What is your impression of Prusa reliability based on?

We have three (or is it four?) of them at the makerspace and they break down much more often than my self-designed printers.

Reliability isn't just using a specific controller or power supply. These are complex machines that ultimately have a lot of user interaction. That user interaction is a huge factor in the reliability of the machine.

When I finished building my last printer, UMMD, I had no problems with it at all, but I wanted to test it so I left it at the makerspace for a few months for others to use. They quickly found the weak points in the design and I made a few corrections. Fortunately, there wasn't much to go wrong. Later I found a few other problems with the belt drive Z axis that were easily corrected and haven't been a problem for the last 2 years.

If you are going to be the only user your chances of high reliability are better than if 20 people will be using the machine. A few simple rules will make it easier to maximize reliability.
1) don't operate components at their limits. This includes mechanical, electrical, and thermal considerations.
2) don't cheap out on the power supply. If the machine needs 200W, spend $5-10 more and get a 300W power supply. Don't buy crap "LED" power supplies via ebay. Pay $5-10 more and get a MeanWell power supply.
3) don't use a controller that uses tiny plug-in motor driver modules. They are not reliable.
4) use metal parts instead of printed plastic wherever possible. If you're using plastic, design the parts as if they are made of plastic, not metal.
5) use big bearings instead of little bearings
6) don't use steel core belts

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Porting Prusa's reliability to a custom made printer.
August 11, 2021 11:36AM
The issue is 99% of al 3D FDM Printer companies with fDM technology that came to market failed bemuse of firmware/ electronics / slicer / software. We no longer have enough expertise in that field it's a dying skill set. No one is teaching complex electronics and firmware development in school, so if you posses that skill, you either get $200 an hour job with the big boys, or you are self taught which is what most of these start ups come from, people throw things together and hope it will work out. look at CraftBot, Afinia, Leapfrog......the list is long of failures that are firmware software related.

I need this for clients who are not 3D printer tech savy, yet I need scalability. I had samples printed on many printers, the only one that impressed me was Markforged, the gantry is properly engineered, it's simple bowden set up with a one trick pony material. I even considered looking at taking Markforged OnyxOne $4500 machine and retrofitting it to print my filaments, since the gantry is solid design. It's not a coreXY , more of a hybrid core XY kinematics.

But that retrofit comes back to electronics / firmware software again.

Prusa fails on mechanical design not electronics, that is why I focused on it, they run 24/7 the electronics are already proven, the gantry is an easy solution, since I can go up to $3000 - $5000 as far as clients would pay for a machine in that range as long as it's reliable with prediction and plug-play. By prediction I mean the Strstasys approach that after X hours they will be forced to change certain components to maintain precision. So in theory I can copy Markforged Onyx printer design via local CNC shop , but still need expertise on electronics. The fact that no one can really answer my question is the exact point I'm making, we don't really have any experts that can offer a good perspective.

Edited 2 time(s). Last edit at 08/11/2021 11:46AM by lightyears101.
Re: Porting Prusa's reliability to a custom made printer.
August 11, 2021 12:45PM
So what is the question you want answered?

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Porting Prusa's reliability to a custom made printer.
August 11, 2021 01:42PM
... it's not this easy eye rolling smiley

In one of my previous day-jobs I've developed a "reliable" system of 3-axis robot and 24V-electronics (based on ArduinoDue) and adapted/modified Marlin4Due and Pronterface for our needs - this was mainly focussed on paste-dispensing, but could "still" 3D-print and was too used for laser-engraving/marking and -cutting on industrial level (search here in the forums for "CNCsimply" and "gieCAPS").

The company couldn't start a self sustaining market on its own (only relied on the singe counts of "industrial" customers) and the "product" wasn't meant for DIY (too pricey) ... so it "dimmed" away ...

Edited 1 time(s). Last edit at 08/11/2021 01:43PM by VDX.

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

Call for the project "garbage-free seas" - [reprap.org]
Re: Porting Prusa's reliability to a custom made printer.
August 12, 2021 01:55AM
My home built and designed delta has been running for about four years now. If the filament makers could actually produce filaments that met their specs, I wouldn't have any trouble.winking smiley
Re: Porting Prusa's reliability to a custom made printer.
August 12, 2021 08:39AM
Where are you getting filament that is so far out of spec it causes printing problems? I used to measure every spool in 30 places and calculate the average diameter of the filament and used that value when I sliced. I stopped doing that about 3 years ago because I found that even the cheapest no-name brands of filament I routinely buy via amazon.com have been very good at controlling the filament diameter.

Back to the OP: I disagree with almost everything you've said about schools not teaching anyone about mechanical and firmware design/interaction/programming. Engineering schools are full of people learning about such things. Even hobbyists with no formal engineering training are learning about it and making great machines to do all sorts of tasks, including 3D printing. I don't know how you can base a business on "I'm going to sell slow printing, mid-sized 3D printers for $3-5k that only print PLA because they have no heated bed, and they will have none of the bells and whistles like autoleveling and bed flatness compensation that people have come to expect in $300 printers. I'm going to make them safe/reliable simply by copying the Mark Forged mechanism and using Prusa electronics/firmware, but I'll be charging 2-3x what Prusa charges." The only customer I can think of for a printer that requires absolute reliability over capability, even at very high cost, is NASA. Good luck selling to them.

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Porting Prusa's reliability to a custom made printer.
August 12, 2021 11:43AM
I wasn't seeking a business advice, clearly you have issues reading and understanding the actual request. There was no mentioning here what am I printing with as far as filament, who are my clients, what specific designs or industry am I entering or any business plan. The questions were very specific strictly on technical side. Your arrogant statement disqualifies you from credibility point of view in participating in this thread.
Re: Porting Prusa's reliability to a custom made printer.
August 12, 2021 02:10PM
I asked what the actual request was because it wasn't clear. There was a lot of rambling and complaining about things that don't seem to be true...
You're right, I am disqualified... best of luck to you.

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Sorry, only registered users may post in this forum.

Click here to login