@Anthony_White I have not tried to fix it in repetier yet, but i don’t think it will be that hard. The 0,0 middle is a better solution for some things.
@Nigel_Dickinson I didn’t redesign the simple because it worked out of the box. No issues. The frame is not what started me off on this redesign, it was the Z axis. But if you are going to redo something, may as well redo it right. I cant tell if you are against the slotted interlock design or for it. before it seemed that you thought that i should stick with the same design, no you are saying printed parts would be less accurate, my argument was to remove the printed parts and add laser cut interlocks. This would remove 12 printed parts and about 24 screws.
@Anthony_White @Nigel_Dickinson I am not against putting the bed center as 0,0. That is not really a printer design problem
@Camerin_hahn so your redesigning something that works to making something less workable. Its a proven design. Why didnt you buy one of the wooden box frame machines as then you’d have a point. But your just barking up the wrong tree. And does your new machine work out of the box…yes. Your finding fault where there is non. Just using more accurate parts doesn’t mean the machine becomes more accurate. As anybody with an ounce of sense knows tighten tolerances can cause more problems than having some play.
@Nigel_Dickinson The fault i found is that my Z-Axis mount was damaged by the limit switch at the top failing to engage (something was not properly inserted in the screw box, yes this was my fault). After getting that fix the nozzle still hits the bed. I have had to do more fiddling with the Mendel90 then i have had to do with the simple. I would argue that the simple worked much better out of the box. Aside from the pre-sliced file that came with the box, i have yet to get a good print out of the mendel90. So i would argue that no it didn’t work out of the box.
Lol so youve got lucky with the Simple mine needed all the E steps sorting and a lot of firmware messing to print and home right. Get that right first then do the upgrading. And your mistake may of damaged more than that part.
The old saying is a poor workman blames his tools. Seems your that poor work man.
@Nigel_Dickinson you make little to no sense, first you agree that it should home at the bottom, then you say i shouldn’t redesign it to home at the bottom.
No I didnt you changed the subject I said each to their own. I’m saying get your software right first then alter if needed. Seems to me you making an issue out of your lack of understanding and trying to defend your position when this machine has been proven and you seem to be the only one with a problem. I’m sure Nopheads customers are more than happy by the posts ive seen. Perhaps you actual need to listen and learn before making out its the machine and not your lack of knowledge.
I think we can all agree the Mendel90, being a self-build machine, can be subject to construction errors.
Also, I think it unfair to compare the Mendel90 first experience to an off-the-shelf printer. Aside from building the Mendel90 you are also required to test and calibrate it. The amount of care taken in all stages of getting a Mendel90 working is reflected in the end result.
I spent about 20 hours building my Mendel90 over 10 days and another 5 hours testing and calibrating before first use. Subsequently I have levelled the bed a further 3 times (improving the accuracy each time) and reset Z height many times as a result of measuring the first-layer height of several prints.
The end result of this effort is a machine that prints really well and my results have received an envious reception at my local RepRap User Group.
I have made one design change to my Mendel90 and that is the dust filter/feed tube anchor. I had an issue with the dust filter sponge breaking up and being transported along the PTFE tube after the tube had been pushed out of its anchor. After discussing the problem with Nophead, I redesigned the tube anchor to accept the tube from the outside and removed the need to apply tape to the tube.
Aside from this one improvement, which fixed a specific issue, I have found the Mendel90 design to be entirely adequate.
@Neil_Darlow why is 5 hours of calibration acceptable?
@Nigel_Dickinson I really don’t like homing away from the work piece, the reason given make no sense to me, it is more likely to damage the machine, adds print time, calibration time and difficulty. The only up side is the ability to resume prints that failed in the middle, the rest of the points can be replicated in a home down system.
@Camerin_hahn
I put testing and initial calibration together. I did not rush the testing to ensure there were no wiring faults and the bulk of the calibration time was in levelling the bed. This can be time consuming when using a dial gauge because of the iterative nature of the process. 5 hours is not a long time in that context.
@Neil_Darlow The simple was assembled and printing in 5 hours. From box to first print.
Edit: note my simple was a kit as well. This is why i think the kit needs to be updated, not because it is poor quality, but because the build time/calibration time could be dramatically reduced without sacrificing print quality.
@Camerin_hahn your comparing a simple to 90 is like comparing a 90 to a 125k printer there worlds apart. I built my simple in 5 hrs and spent 2 hrs getting it just right. And spend 15 to 20 minutes every time I want to print to set 0 as I want. Once you set your 90 up right you won’t need to do this.
And as for added time to zero away from the work area is seconds. And doesn’t effect calibration in the slightest. And at the end of the day I think this subject is beyond your comprehension.
@Nigel_Dickinson have you used a 90, time to home is 20-40 seconds. Hackers motto, shape your tools or they will shape you. Your elitist and complacent attitudes toward everyone has always been a sore spot with me. You don’t know me, stop pretending you understand what things I know and what I don’t. If you have never built or calibrated a 90…
Youve never calibrated a 90 either you post says as much. Others on here tell you the problem is you. …seems you cannot except that. And setting up any printer involves time logical thinking and above all advice from those who have…seems your ignoring the last one. And here’s a really Hackers motto only improve on things that can be improved upon.
I’m not elitist and if I was complacent I to would drive my print head where its not supposed to go.
@Nigel_Dickinson I did, i actually did get 1 acceptable print out of it, unfortunately i have not been able to get a second good print out yet, most of them are blobs. I have spent a few hour trying to get the calibration to work consistently and it doesn’t seem to work well i will get it working, home it a few times to check, then when i start a print it drives the head into the bed. Understand i do want this printer to work, and it isn’t currently.
@Camerin_hahn Is your Z-Drive train slipping? If repeated Z up-down motion causes your Z zero to shift then that might indicate this is happening. Did you degrease the threaded rods before putting them in the couplers? Is your top Z-limit switch working effectively i.e. does a Z-homing operation result in upward motion, hitting the switch then backing-off a little?
@Neil_Darlow I did not degrease my threaded rods thoroughly, i cam strip it down tonight and try that(to me burring into the bed doesn’t sound like slippage, if it was floating i would agree, i can sharpy the rod and mount to see if it spins), the limit switch is working fine(it does move down a little bit after it clicks), as part of my debugging i did run it up and down several times measuring the distance traveled, it always moved the exact distance i asked for in the host software, i tried jumping from the top to the bottom using the host software and it seems to be good, but if i run the same G-code file some times it hits the bed, not every time however.
@Camerin_hahn
OK, if there is no mechanical slippage then maybe missed steps. Is the GCode driving the Z-axis at too high a speed? That might account for this also. I am curious as to what values you put into the firmware for E_STEPS_PER_MM multiplier and Z height - just to see how they compare to mine - they might offer additional clues to what is happening with your prints.
@Neil_Darlow I will check when i get home, but currently the E-step per mm is the default value, it shouldn’t change, as it is calculated from thread pitch on the z axis.