Ebay SINTRON Kossel Mini full kit

Quote
Tha_Reaper

Quote
DRTak
DO NOT put the RAMPS below the heatbed. I did that and I fried my MOSFET. Why? Because you cannot see the fan moving and blowing air. Yes I installed a fan below the heatbed blowing onto my RAMPS. It looks amazing. All electronics hidden. But you cannot see or hear if the fan is on. If for some reason the fan is off. Your RAMPS will smoke and you will not know until it is too late. Also, I insulated the heatbed bottom. But you are still getting cool air below counteracting the heat above. So it takes longer for the heatbed to get hot.

Thats where this thing comes in handy: [www.thingiverse.com]
you can see when your fans are running. I use 2 of those, and 1 30mm fan inside of the frame. The 30mm fan is very, very noisy. as soon as its up to speed you can hear it above all the other printer sounds. Hard to miss when that thing is not on.
I know the risk of smoking the RAMPS is higher, but i still prefer this to external electronics (at least untill i tidied up the cable mess and i'm able to build a good case for the RAMPS)

Good idea. But I found that the one Mosfet gets way too hot. So I added another heatsink onto it. And make sure that the 50cm fan is blowing directly on it. I have a 40cm fan blowing the hot air out. My ramps box is outside of the kossel so everything is working great now,

I can suggest buying solid state relay to connect heating bed via it. It will eliminate all problems with mosfet and fuse overheating.

How long does it take to 'initialize' printing? I loaded the gcode from slic3r and something happens. Probe is up and everything.

Quote
Bigkahuna2187
How long does it take to 'initialize' printing?

AFAIR, the default Slic3r sequence is: printer waits for bed is heated, then homes, them moves the head closer to the bed, heats the head, and starts printing.
Heating the bed is the most time consuming procedure. Heating the head takes about 1.5 minutes.

both hotend and bed are heated. Nothing happens at all once I hit print.

The LCD screen usually shows current stage - "Heating the bed", "Heating...", "Heating done." And right after that actual printing is started.
You can look into g-code file and see what actual sequences are sent.

ok looks like the temp was off by just a tiny bit. Now I just need to sort out the actual printing. It just squirts in one place while moving up in the Z haha.

Thanks everyone for the help. Couldnt have gotten this far without this site.

Quote
GrAndAG
I can suggest buying solid state relay to connect heating bed via it. It will eliminate all problems with mosfet and fuse overheating.

Much cheaper (and easier) way is to just solder a wire to the Gate Signal of the internal MOSFET and place a IRF540Z external - connect Heatbed there....
Wire from RAMPS to GATE of MOSFET
"-" from Powersupply to SOURCE of MOSFET
"+" from Powersupply to Heatbed, other Wire from Heatbed to DRAIN of MOSFET
(Configuration is called "Low Side Switch")

IRF540Z does not need any cooling for up to 10Amps.. - I have no FAN blowing on it, and no Heatsink mounted.
Heatbed gets hot verry fast, IRF540 stays absolute cool, RAMPS stays cool too.

And with this config, you can easily change heatbed to 24V - just make sure "-" from both Powersupplies are connected together, and connect +24V direct to Heatbed like mentioned above...

Quote
Tha_Reaper

I also made a M666 button in pronterface. i use that button before i start every print. after i replaced the tape on the bed, or removed the bed, i totally recalculate the M666 settings again. takes me about 2 minutes to get it manually right if i use my last M666 settings as a starting point.

Thanks for the Idea with M666 Button - until now I did mechanical calibration - and was thinking of how to insert the M666..
As I'm often printng from SD, I will include this M666 to start g-code too.

Okay, so I almost finished height calibration yesterday. However, when I tried to do a 'hot calibration' (with hotend and heatbed set to PLA temperatures), the heatbed stopped warming after some seconds. From what I have read in this thread, it seems as if the stock PSU (rated at 10A) simply cannot provide enough current to heat the bed. Suggestions ranged from increasing voltage to something above 13V (as it won't affect linear regulators too much but might kill your fans) to adjusting PWM frequency by modifying MAX_BED_POWER in the Marlin 'configuration.h'. Further, I have an old ATX PSU lying around which is rated at 17A. I could modify this one without too much effort. What is the way to go here?

Anyway, before going any further I will order some 40mm fans and put them outside of the empty space below the heatbed using some kind of adapter (which I still have to design and mill). Further, as Tha_Reaper suggested I will put a heatsink on the MOSFET. This might solve the heating problems mentionedabove until I can afford to buy a Duet board, which I will mount in a separate box anyway.

Quote
DRTak
Also HaGeEm what CAD software are you using. I am using TurboCAD pro.

At home I am using the developer version of FreeCAD, mainly because it is open source and runs on my linux box. I like it a lot! However, at work we are using SolidWorks altough I did not have too much time to play around with it.

Quote
HaGeEm
Okay, so I almost finished height calibration yesterday. However, when I tried to do a 'hot calibration' (with hotend and heatbed set to PLA temperatures), the heatbed stopped warming after some seconds. From what I have read in this thread, it seems as if the stock PSU (rated at 10A) simply cannot provide enough current to heat the bed. Suggestions ranged from increasing voltage to something above 13V (as it won't affect linear regulators too much but might kill your fans) to adjusting PWM frequency by modifying MAX_BED_POWER in the Marlin 'configuration.h'. Further, I have an old ATX PSU lying around which is rated at 17A. I could modify this one without too much effort. What is the way to go here?

Anyway, before going any further I will order some 40mm fans and put them outside of the empty space below the heatbed using some kind of adapter (which I still have to design and mill). Further, as Tha_Reaper suggested I will put a heatsink on the MOSFET. This might solve the heating problems mentionedabove until I can afford to buy a Duet board, which I will mount in a separate box anyway.

Quote
DRTak
Also HaGeEm what CAD software are you using. I am using TurboCAD pro.

At home I am using the developer version of FreeCAD, mainly because it is open source and runs on my linux box. I like it a lot! However, at work we are using SolidWorks altough I did not have too much time to play around with it.

You could use the Heatsink Method and connect the second Powersupply to the RAMPS Input marked with 11A (but you still might get Problems with the polyfuse), or you could do the method I mentioned 2 Posts before with IRF540 MOSFET, and connect the second Powersupply there...

Edited 2 time(s). Last edit at 11/13/2015 04:02AM by Sir_Death.

Quote
HaGeEm
Okay, so I almost finished height calibration yesterday. However, when I tried to do a 'hot calibration' (with hotend and heatbed set to PLA temperatures), the heatbed stopped warming after some seconds. From what I have read in this thread, it seems as if the stock PSU (rated at 10A) simply cannot provide enough current to heat the bed. Suggestions ranged from increasing voltage to something above 13V (as it won't affect linear regulators too much but might kill your fans) to adjusting PWM frequency by modifying MAX_BED_POWER in the Marlin 'configuration.h'. Further, I have an old ATX PSU lying around which is rated at 17A. I could modify this one without too much effort. What is the way to go here?

Anyway, before going any further I will order some 40mm fans and put them outside of the empty space below the heatbed using some kind of adapter (which I still have to design and mill). Further, as Tha_Reaper suggested I will put a heatsink on the MOSFET. This might solve the heating problems mentionedabove until I can afford to buy a Duet board, which I will mount in a separate box anyway.

Quote
DRTak
Also HaGeEm what CAD software are you using. I am using TurboCAD pro.

At home I am using the developer version of FreeCAD, mainly because it is open source and runs on my linux box. I like it a lot! However, at work we are using SolidWorks altough I did not have too much time to play around with it.

I haven't even tried the stock PSU, still lying around somewhere, can't say much about it, I have ATX PSU connected. Even with that I had to lower bed duty cycle and even then I had to replace bed fuse on ramps (alternatively you can fan it like crazy). Check if it's hot.

Duet is indeed somewhat expensive, so I think it makes perfect sense to get things working with ramps first before investing in duet.

Increasing voltage to feed a fixed resistor ( heatbed ) raises current instead of reducing it. ( At least until PID kicks in )

The only chance to reduce output power is PWM-reduction for the heatbed MOSFET, but then you might not be able to reach desired temp...( above 70-80°C )
-Olaf

Quote
Sir_Death

Quote
Tha_Reaper

I also made a M666 button in pronterface. i use that button before i start every print. after i replaced the tape on the bed, or removed the bed, i totally recalculate the M666 settings again. takes me about 2 minutes to get it manually right if i use my last M666 settings as a starting point.

Thanks for the Idea with M666 Button - until now I did mechanical calibration - and was thinking of how to insert the M666..
As I'm often printng from SD, I will include this M666 to start g-code too.

I used to do that too, but i gave up on that since everytime you tinker with the heatbed (even replacing tape matters), the calibration will be thrown off and you have to edit your gcodes. Pretty annoying when you want to reprint an old model and you have t edit your gcodes first to get it to print right. Knowing myself i will forget to do that 90% of the time.

Quote
Sir_Death
You could use the Heatsink Method and connect the second Powersupply to the RAMPS Input marked with 11A (but you still might get Problems with the polyfuse), ...

Just to make sure: from the schematics of the RAMPS board Q3 is the MOSFET we are talking about (provided the heatbed is connected to D8), right?

Quote
HaGeEm

Quote
Sir_Death
You could use the Heatsink Method and connect the second Powersupply to the RAMPS Input marked with 11A (but you still might get Problems with the polyfuse), ...

Just to make sure: from the schematics of the RAMPS board Q3 is the MOSFET we are talking about (provided the heatbed is connected to D8), right?

Yes Q3 - the one orientated 90° to the two others...
I unsoldered the Power-Jack from Arduino Board, so there is enough space to solder a wire on Bottom of the RAMPS-Board - there you have easy acces to the GATE-Signal (It's the Pin with the very small PCB-Wire - the 2 broad ones are for Heatbedpower...

Quote
Tha_Reaper
I used to do that too, but i gave up on that since everytime you tinker with the heatbed (even replacing tape matters), the calibration will be thrown off and you have to edit your gcodes. Pretty annoying when you want to reprint an old model and you have t edit your gcodes first to get it to print right. Knowing myself i will forget to do that 90% of the time.

I do not save my old gcode files - only the stl, as a new filamet spool from a new supplier can have great diffence in config-values of slic3r - Hotend temperature, extrusion Multiplier, printable mm³/sec, retraction,....
So without 100% knowing it is the same Filament from the same supplier, manufactured the same way, with the same Color and the same ingredients.... - I do not even try to Reprint an old gcode (If there would have been changes in Hotend, it would even get worse....)

Quote
o_lampe

Quote
nebbian
Alternatively if you're feeling brave you could disassemble the push fit connector and drill it out the same as the one on the hotend, which would allow you to push the bowden cable all the way through the hole in the extruder until it's almost touching the extruder gear and idler bearing.

Pretty risky with a M5 or M6 thread. On my e3d clones heatsink it was a 10mm thread with plenty of "meat".

If you still want to try, be aware of brass needs a different drill-angle than the average "all purpose" metal drill has.
-Olaf

Well I felt brave tonight and it all seemed to work fine, despite using an 'all purpose drill'.

Any ideas on why it keep printing like this? lasts about secs then homes. Only prints vertically.



Thanks

Finally got my test print completed! 10x10x10mm cube came out great. Only issue is that the nozzle is too close to the bed. It ended up scraping off some of the kapton I had placed. Any way to adjust this nozzle height after its gone through G28 and G29 prior to printing?

Thank!!

Quote
Bigkahuna2187
Any way to adjust this nozzle height after its gone through G28 and G29 prior to printing?

Thank!!

You have to properly tune the offset of z-probe if you use G29.

@BigKahuna
Your Kapton tape must not overlap. You better accept a small gap between the strips or be more accurate and place them side_by_side.
-Olaf

Quote
o_lampe
@BigKahuna
Your Kapton tape must not overlap. You better accept a small gap between the strips or be more accurate and place them side_by_side.
-Olaf

Thanks. I was trying to prevent gaps but I will replace it once the nozzle doesnt scrape anymore.

I need some help...

as i started printing more complicated objects, i noticed a problem starting to appear in my prints: really bad overhangs.
upon further inspection, those bad overhangs only appeared on the side of objects facing the Z-tower.
I noticed that my hotend is not facing downwards. its at an angle towards the Z-tower, and i guess thats where my overhang problems come from. today i tried replacing the effector and hotend bracket, which failed miserably and only gave me more problems.
Here is a picture of my problem: [www.dropbox.com]

you will notice a thread hanging from the top of the hotend heatsink. That thread has a weight at the end, so it is perfectly vertical. you can see that the hotend is not parallel to the thread. the tower that you see in the picture is the X tower.

Anyone else faced this problem? and how were you able to solve it?
So far i tried: replacing the effector and bracket, putting washes on the Z tower side between the bracket and effector (which led to a bend bracket)

Quote
Tha_Reaper
I need some help...

as i started printing more complicated objects, i noticed a problem starting to appear in my prints: really bad overhangs.
upon further inspection, those bad overhangs only appeared on the side of objects facing the Z-tower.
I noticed that my hotend is not facing downwards. its at an angle towards the Z-tower, and i guess thats where my overhang problems come from. today i tried replacing the effector and hotend bracket, which failed miserably and only gave me more problems.
Here is a picture of my problem: [www.dropbox.com]

you will notice a thread hanging from the top of the hotend heatsink. That thread has a weight at the end, so it is perfectly vertical. you can see that the hotend is not parallel to the thread. the tower that you see in the picture is the X tower.

Anyone else faced this problem? and how were you able to solve it?
So far i tried: replacing the effector and bracket, putting washes on the Z tower side between the bracket and effector (which led to a bend bracket)

I can't help you, but I also have hotend a bit tilted when it's closer to the edge of printable area. I do not yet have serious problems associated with that. I suspect my pushrods are of different length, but haven't yet tried to deal with it. At any rate, I'd love to listen to the advice you get

Quote
hacker

Quote
Tha_Reaper
I need some help...

as i started printing more complicated objects, i noticed a problem starting to appear in my prints: really bad overhangs.
upon further inspection, those bad overhangs only appeared on the side of objects facing the Z-tower.
I noticed that my hotend is not facing downwards. its at an angle towards the Z-tower, and i guess thats where my overhang problems come from. today i tried replacing the effector and hotend bracket, which failed miserably and only gave me more problems.
Here is a picture of my problem: [www.dropbox.com]

you will notice a thread hanging from the top of the hotend heatsink. That thread has a weight at the end, so it is perfectly vertical. you can see that the hotend is not parallel to the thread. the tower that you see in the picture is the X tower.

Anyone else faced this problem? and how were you able to solve it?
So far i tried: replacing the effector and bracket, putting washes on the Z tower side between the bracket and effector (which led to a bend bracket)

I can't help you, but I also have hotend a bit tilted when it's closer to the edge of printable area. I do not yet have serious problems associated with that. I suspect my pushrods are of different length, but haven't yet tried to deal with it. At any rate, I'd love to listen to the advice you get

In my case i'm pretty sure its not a pushrod problem. the degrees of tils stay the same wherever i move the printhead.

Quote
Tha_Reaper
In my case i'm pretty sure its not a pushrod problem. the degrees of tils stay the same wherever i move the printhead.

If the tilt is constant then check that each of the carriages is truly upright, so that both rod bearings on it are at the same height.

---

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].

that would make the effector not level. as you can see in the picture, the hotend is tilted while the effector is level. It looks like the problem is in the bracket holding the hotend in place, not in the effector, rods, or carriagesits just very anoying that i dont know how to fix it.

If the effector moves without any tilt, you should have absolutely no problem even with non-vertical hot-end. Because this defect affects [X:Y] offset only, and it's constant. Which can be adjusted in Slic3r to make [0:0] point of the nozzle perfectly in center of the table if you wish. But it's absolutely not needed.

Edited 2 time(s). Last edit at 11/14/2015 03:33PM by GrAndAG.

Quote
GrAndAG
If the effector moves without any tilt, you should have absolutely no problem even with non-vertical hot-end. Because this defect affects [X:Y] offset only, and it's constant. Which can be adjusted in Slic3r to make [0:0] point of the nozzle perfectly in center of the table if you wish. But it's absolutely not needed.

I'm sorry, I misunderstood. I don't think that tilt matters very much. I think it may be caused by the groove mount plate not being screwed tight up against the effector, so the effector is horizontal but the groove mount plate isn't.. May be there is a stray bit of plastic between them that needs to be trimmed off.

---

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].