Close



Page 2 of 2 FirstFirst 12
Results 11 to 18 of 18
  1. #11
    Staff Engineer
    Join Date
    Jul 2016
    Location
    South Florida, USA
    Posts
    950
    Add AutoWiz on Facebook
    Quote Originally Posted by engpro View Post
    That makes sense.
    So for high quality resulution what are the most important factor. Where is the bottleneck? I understand after reading a little after that 32bit processor is useful to drive the motors much more smoothly and quieter.
    I'm trying to understand what makes a good quality printer
    More than that. Marlin 2.0 was designed for the 32 bit boards. We are many versions of marlin into 2.x now. We are at 2.0.7.2 today. And marlin 1.1.9 was made a long time ago and does not come with support for the silent drivers and other features that have become common place since. And while we do see people running the newer marlin on the older boards they aren't doing it well because the firmware has grown and it is bigger with more lines of code and more libraries. And the older boards have absolutely no onboard memory and they struggle to run marlin 2.x and then there is the clock cycles, right. I mean how many decisions can you make in a second? the 8 bit boards are lucky to run at 16mhz. The SKR 1.3 and 1.4 run at 100mhz, the SKR 1.4 TURBO runs at 120mhz, and the SKR PRO runs at 168mhz. And I believe the newest addition to the Duet lineup is actually running at 300mhz. These boards power through the more feature packed marlin versions. Once again the best of the 8 bit boards run at a whopping 16mhz. Just saying. Running the silent drivers is good. Being able to get the mainboard to communicate with them while printing is priceless. Because in UART and SPI we can change the forward current in the firmware and we can have the stepper dynamically switch on its own between stealthchop and spreadcycle as the demands of the print change.

  2. #12
    I apprichiate the great information I am extremely happy with the resolution of my printrbot simple metal, so if i can achive that i will be happy with this build.But i wonder on this wanhao d9 400 mark 1 auction what would be a good price, right now. I beleive nothing is broken, just the z axis refuse to calibrate proper he said.I learn allot about the potential, you can even print bearings and use soft airgun pellets inside, amazing!wanhao d9 400 mark 1 does seem to have a really rigid frame and a big bedsize, and the steppers look solid, thats why this auction was appealing to me, but i feel i should limit to not buy over 150usd, what would be a reasonable price for one that requires some mantance?

  3. #13
    Super Moderator curious aardvark's Avatar
    Join Date
    Jul 2014
    Posts
    8,193
    wanhao d9 400 mark 1

    Looks like a decent machine and wanhao are known for quality machines.

    If it doesn;t work then it's most likely the board or screen that's gone - both easy and cheap to replace.

    I wouldn't bid more than $200.


    As far as components for building an i3 of your own.
    I'm mostly with autowiz (but he has much deeper pockets than I do). So always go for a new board. At the moment you can pick up the skr 1.4 turbo with touch screen and silent steppers for a small amount of money.

    I've seen no obvious benefit from 1.8 steppers over the usual 0.9 cousins.
    Given that my sapphire pro still prints faster and cleaner than autowiz's mega beast, that might as well be made from gold bricks. I'm still waiting to see where he's gained any benefit at all over my $300 stock machine.

    I mean there should be benefits from all that money/hardware - but so far, none have appeared to be obvious.

    Thomas sanlederer has a couple videos where he builds an improved prusa mk2 - possibly also a mk3 as well - so I'd start there.

    But as far as boards go - 32bit definitely.
    Sockets for replaceable stepper drvers - definitely.
    A rock solid frame - absolutely. You could drive a tank over my sapphire pro and it would just laugh at you. The base is 2.25 steel and the top is a solid sheet of 7mm aluminium. joined by 20x20 aluminium extrusion. Very simple design but absolutely rock solid.

    The rest is kinda optional.
    Although anytime you can replace roller guides with linear rails, I'd do so. You can get decent rails at a reasonable price.
    Okay it's harder to .line the buggers up - but the benefits once done are well worth the hassle. Both for print speed and precision.

    For an I3, having side support rails is essential and liner rails are better than the rods and linear bearings.

    I can't really see what the wanhao machine has. But should be side support rods as they've built a lot of i3 machines over the years and tend to stick to the original prusa design fairly faithfully.

    As far a smatching the printrbot metal - you might be surprised how hard that might prove to be.
    Printrbot made really good machines and ALWAYS had more orders than stock.
    I still don't know why the company folded, as they could sell anything and everything they could make.

    That one was a total surprise.

    Some of the startup companies you can see right from the start are going nowhere, but printrbot just looked like going nowhere but up.

    Kinda like Mcor - now there was a company that just seemed to lack marketing. Fantastic macgine with sme of the best full colour prints I've ever seen and handled. But just never seemed to find the right market.
    But printrbot had none of those issues. Total mystery why they went bust.

    Given that prusa sells more printers to the us than any us based manufacturer and the machines are still as much printed as possible. I just don't see how printrbot failed with an all metal frame that still came in cheaper than the prusa equivalent.

    I guess it goes to show that engineers don't generally make good business men. Josef prusa being an obvious exception.
    Anyway your printrbot is a great little machine and if it's the model with linear rails - you'll be pushed to make something without them, that comes close.
    Last edited by curious aardvark; 12-31-2020 at 11:56 AM.

  4. #14
    Thank you for the great information, it is good to know where i should put my priorites, good 32 bit board but the motors can be cheaper nema even 1.8, thats great to know.
    Also having a reasonable upper bound on the bid like 200 sounds very good, i do hope i win.
    I have learned allot when i am planning to build my own, but i canceled my orders of the cheap drivers thankfully.
    I been experimenting with my simple metal, and i can have good quality even at speeds like 120 mm/s, but it felt wrong to have so fast so i put it at 70 for now, i dont want to damage the hardware.
    In order for the print to stick to bed i need first layer to be 220 celsious, 230 makes it stick so hard it takes 10 minutes to pry off, 210 makes it fail the print.

  5. #15
    Staff Engineer
    Join Date
    Jul 2016
    Location
    South Florida, USA
    Posts
    950
    Add AutoWiz on Facebook
    Quote Originally Posted by engpro View Post
    good quality even at speeds like 120 mm/s
    Is there any way you can post up your acceleration and jerk settings so we can know how fast you actually mean. for example, 120mm/s doesn't mean much if your bed size is 120x120 and your acceleration is so low that it takes 200mm of travel to actually hit that speed. ya know? 120mm/s is an incredibly blank statement without acceleration at least. It's like asking someone to solve a math problem and then just saying "12". I mean it is very important to also know about jerk or junction deviation because these are powerful modifiers to this speed or max feedrate you are claiming. if we are trying to actually visualize a print speed in our minds. A better way if we do not know what these settings are would be to just make a simple video with your cell phone of your machine printing and then post it up in high quality to youtube and then provide a link here so we can see how fast your machine moves. You know what I mean? Was it this fast: https://www.youtube.com/watch?v=DXpZJzbA5TU . Or was it this fast: https://www.youtube.com/watch?v=uCg4gqGRbmY . Or was it faster than that?

  6. #16
    Well there is still allot i dont know, when i say 120 i meant that was the setting i placed in cura, i have not touched the acceleration settings, so those things are the defualt.
    I would say the speed was like your last video example, but it did not feel good. Right now the speed is about the one of your first video. Do i have to worry about if i go to fast the hardware will break? Or is it more i can go as fast i want so long build quality is good?

  7. #17
    Staff Engineer
    Join Date
    Jul 2016
    Location
    South Florida, USA
    Posts
    950
    Add AutoWiz on Facebook
    I am sure or I would hope you would just have to deal with degradation of the print quality as the speed overwhelmed the mechanical limits of the printer and things got to shaking. Maybe even start skipping steps. But I would hope not to see any breakage of the hardware. I have another thread where I am trying to compare print speeds with other people and am learning that slicer settings have a big impact on how long a print takes regardless of print speed settings in the firmware. And so a video uploaded to youtube will always be the best way we have to show each other exactly how fast our printers are moving and laying down the filament. I made those videos with my cell phone and uploaded each to youtube absolutely free in the span of a few minutes.

  8. #18
    Super Moderator curious aardvark's Avatar
    Join Date
    Jul 2014
    Posts
    8,193
    A far as sticking to the bed goes - have youir bed temp at 65c - I find that's optimum for pla.

    Change from cura to prusaslicer - it's just Soooooooo much easier to mess about with settings to tune printers in.

    To adjust 'stickability' rather thn changing the print trmp. You need to play with the speed of the first layer and how close to the build surface the print nozzle is.
    This is easier in prusaslicer than cura.

    Different machines need different settings.
    My sapphire pro with PEI and some build plate adhesive, can print a good first layer at 75mm/s - I usually have it nearer 20 or 30 - depends on how large an area the print has. Larger surface areas mean you can generally print the first layer faster.

    Now my delta - with a similiar buildplate setup, I generally run at 10% of the actual print speed. so usually somewhere between 10-15mm/s

    My replictor pro - which ahs a mystery print surface I picked up for free at a tct show - runs at about 10mm/s, but has a slightly larger z-axis gap as whatever it is - pla really sticks well to it.

    All the printers run pla at 200-210c (depending on speed) and 65c printbed.

    For really fiddly prints, I've had the delta as low as 3mms with cheap nylon filament.

    Speedwise - lol, unless it's rattling around and making horrible noises - run it at the fastest speed that produces the print quality you want.

    For big prints or prints that don't need to look smooth I usually run the delta and sapphire at 150mm/s and 0.3 or 0.4 mm layer height.
    The rep clone maxes out at 75mm/s for any prints as it's got an extruder carriage with two nema 17's and two hotends on it - and you just can't shift that kind of weight around too fast without some kind of counter balance. So that runs at 65mm/s for pla and about 20mm/s for flexible filaments.
    Flexible filaments are sticky so you can run the first layer the same speed as the main speed.

    But basically the print temperatures are determined largely by the material and the print settings: speed, zgap etc are determined by the printer and the build plate material/coating.

Page 2 of 2 FirstFirst 12

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •