Ebay SINTRON Kossel Mini full kit

Quote
GrAndAG

Quote
hacker
IIRC, in particular it's much thinner?

The thickness is generally the same (the plastic one is even a bit thicker). Because they both should match the mounting slot at the hot-end.

Makes sense. I must've been confused.

Hello everybody,
I also purchased this kit. Unfortunately before I found this thread. I wouldn't if I had seen all the problems before.
After a lot of pain mounting and cabling the thing together with its useless manual its finally done with a lot of help from this thread
I used all parts provided in the kit and grinded down the rods so they move freely and don't bound and jam to each other and the mounting on the tower.

Now its calibration time witch drives me nuts. Basically its not working at all for me. Especially the autoprobe is a joke and constantly cashes the head onto the bed and on the sides.
I don't understand why Sintron just don't provide a working firmware where just the delta z for the height and the length of the rods needs to be adjusted. All other values should be the same or not? The manual says I have to measure (exactly) different parts where its not clear for me from where to where I have to measure.

My question to you guys is if anybody has a realy working firmware for me where I just have to correct the height and the rod length to finally start printing anything?

Any help much appreciated!

Sintron firmware is like the hole kit: borrowed and puzzled together without any beta testing. ( except the beta testers AKA customers )

I'm pretty sure, if one of the buyers will come up with a working firmware, they copy the files to their server and call it "solved "

Right now, I can't recommend a firmware for Ramps/mega that is "idiot proof" ( no offense )
Rich Cattels auto calibration FW-branch is two years old and no longer maintained.
The new guy working on debugging Marlins ABL ( thinkyhead ) doesn't even have a delta printer...

The picture I attached, helped me a bit to identify what's what, but still I stumbled over the effector offset. It is NOT the distance from effector-center to rod-center.
It is the distance between effector-center and the virtual axis between two rods.

I'm running duet-controller and DC42's firmware now and it was pretty easy to calibrate ( because I had all the data from weeks of Marlin calibration already )
-Olaf

Hi Mr PCatt
Use The Picture from o_lampe write in your measurements they should be accurate to 0.1 mm if you can
I have up loaded My Marlin_Delta Configuration.h file
in it I have commented the lines that must be customised with your measurements from the picture.
once you establish the Height it will take quite a few go s to do this then you make the bed level.
set you diameter to 170mm for the plate and circular in Pronrterface and or Cura
keep going it took me days to do this but worth it. I do not use the z probe..it attacked the heat bed plate like a rabid pit bull when i started

Further
In Pronterface always start with G28 then drop it down to about 10 mm before the bed then take 1mm steps then 0.1 steps untill it is about 1mm above ..do the maths then update config.h and recalibrate. again and again...until Y axis is zero and head is 1mm off plate this is your height

I've written macros in pronterface to probe the towers with a z-probe ( 3 probes to build an average )
You have to adapt the coordinates to your print radius.

; X-tower probing
G1 F4000 X-124.8 Y-72 Z20
g30
g0 z20
g30
g0 z20
g30
g0 x0 y0 z100

Let's say your z-probe offset is 5mm and you get:
4.1
3.95
4.05
as probe result.
Then you set M666 X-1.03 followed by M500 ( store eeprom )

Do a G28 and probe again until the values are in the 5mm area. Repeat this for all towers.
-Olaf

Guys, thank you for your help!
So i measured several things again and uploaded a new firmware, mounted a fresh roll of PLA and now I wanted to give it a try to see if I am even close to print something (without Z-Probe).
BUT
The filament is not coming out of the nozzle. Its just dripping ever so slightly, never ever enough to draw a line or so. The extruder motor is just stuttering because the pla is not going forward, something seems jammed or so. The screw for the pressure on the roll of the extruder motor is set to max. Temperature is set to 205. Fan is on all the time. I can pull the pla stripe back out, and it looks normal to me.
Omg, what is this now? Is the extruder jammed even before the first print? How can I fix this?

Quote
MrPCatt
Hello everybody,
I also purchased this kit. Unfortunately before I found this thread. I wouldn't if I had seen all the problems before.
After a lot of pain mounting and cabling the thing together with its useless manual its finally done with a lot of help from this thread
I used all parts provided in the kit and grinded down the rods so they move freely and don't bound and jam to each other and the mounting on the tower.

Now its calibration time witch drives me nuts. Basically its not working at all for me. Especially the autoprobe is a joke and constantly cashes the head onto the bed and on the sides.
I don't understand why Sintron just don't provide a working firmware where just the delta z for the height and the length of the rods needs to be adjusted. All other values should be the same or not? The manual says I have to measure (exactly) different parts where its not clear for me from where to where I have to measure.

My question to you guys is if anybody has a realy working firmware for me where I just have to correct the height and the rod length to finally start printing anything?

Any help much appreciated!

I understand how you feel. If I had to do it all over again I am confused what I would do. If you read Sintrons manufacturing it says they use UP 2 printer to print the Kossel frame? Not their own Kossel.

But I have talked to James over and over and for a Chinese facility NO other ebay seller would give us the support that they do. They are so huge that if you look up parts they are sold (Sintron) at Sears.com. If you have any issues, talk to their support and they will help you solve the issue in 24hrs. I used this forum and James to help me make my Kossel. So if I had to buy a kit on eBay it would be from Sintron.

Next, If I had to do it over again, I am stuck between spending several thousand on a resin printer or a fully assembled printer. Maybe should have bought UP2 ? since that is what Sintron uses. But XYZ, makerbot, etc use chips on their filament so you cant go and buy generic spools of the stuff like you can with the Kossel.

I would never know what I know now if I had just bought a pre assembled printer. I actually think I would be bored with the whole idea. But part of the fun and frustration is figuring out calibration and fine tuning.

So, with that said, your head crashing issue is related to your z height. You need to tinker with configuration.h. You will find this:
#define MANUAL_Z_HOME_POS 240
Change that value to something high. It will prevent your head from crashing.

Then to fine tune the head you need to tinker with:
#define Z_PROBE_OFFSET {0, 14.27, -2.2, 0}
which is also in configuration.h

Change the -2.2 value. I regularly change this value. The more neg means into the bed. The more positive means up to the top.

Also, I had a solid bed and when the head crashed it bent the nozzle tip. I had to ream it out with a wire to unclog it. So now I have a spring bed leveler. If the head crashes into the bed its not so bad as the springs absorb the shock and I can hit the panic button to shut the system off.

The z probe issue can be solved. I fixed mine. Do not mess with the z probe holder mechanism. I shaved my down to be smoother and it messed up the probe a bit. Clean it a bit but use it as it is. Next, when u mount the probe, in its deployed state it is really far down. You can mess with below in Marlin.main. in the marlin firmware:
void retract_z_probe() {
feedrate = homing_feedrate[X_AXIS];
destination[Z_AXIS] = current_position[Z_AXIS] + 75; // originally +20 at +40 z probe elevates an NO more scratching of heatbed
prepare_move_raw();

destination[X_AXIS] = -33; // 52.8
destination[Y_AXIS] = 95; // 77.5
destination[Z_AXIS] = 30;
prepare_move_raw();

See my notes above. The z probe was so low it gouged my heat bed. So I made it +75. When the probe goes into auto retract mode it elevates itself and does not crash into the bed anymore. the Destination x axis and y axis is where in space you have your z probe retraction mechanism. I got mine off of thingiverse. Its a block of plastic that hits the z probe to make it retract. The settings above are where it is placed on my frame. The only thing that I have to do now before I print is deploy the probe. If I forget to deploy the probe the machine spews filament all over the place when I hit print because it cant do a G29 correctly.

Issues with Resin printer make me happy with my Kossel. You have a small print area with a Resin printer. The resin is pricey. $145/ liter. The resin tray I hear you have to change after every 2 liters. I do not know what a tray is though. LOL. The printer can be bought for $1000-$3500. Items on thingiverse are not easy to print as you have to put holes in them so the resin can leak out.

Quote
MrPCatt
Guys, thank you for your help!
So i measured several things again and uploaded a new firmware, mounted a fresh roll of PLA and now I wanted to give it a try to see if I am even close to print something (without Z-Probe).
BUT
The filament is not coming out of the nozzle. Its just dripping ever so slightly, never ever enough to draw a line or so. The extruder motor is just stuttering because the pla is not going forward, something seems jammed or so. The screw for the pressure on the roll of the extruder motor is set to max. Temperature is set to 205. Fan is on all the time. I can pull the pla stripe back out, and it looks normal to me.
Omg, what is this now? Is the extruder jammed even before the first print? How can I fix this?

I had the same issue. I dont know if I caused the extruder to be plugged or it came that way. You need to take the nozzle off. Examine it. Or manually push filament into the heated hotened. If it curls up on itself then your nozzle is plugged. You have to do what I did above to fix it. You need to ream it with a very very small drill bit or wire. Also, when I got my Kossel. I just put it together. However, you really really should tightened the hot end nozzle with a wrench or you will leak out filament all around the sides and get burnt PLA around the hotened tip.

Quote
MrPCatt
Guys, thank you for your help!
So i measured several things again and uploaded a new firmware, mounted a fresh roll of PLA and now I wanted to give it a try to see if I am even close to print something (without Z-Probe).
BUT
The filament is not coming out of the nozzle. Its just dripping ever so slightly, never ever enough to draw a line or so. The extruder motor is just stuttering because the pla is not going forward, something seems jammed or so. The screw for the pressure on the roll of the extruder motor is set to max. Temperature is set to 205. Fan is on all the time. I can pull the pla stripe back out, and it looks normal to me.
Omg, what is this now? Is the extruder jammed even before the first print? How can I fix this?

I wonder can you also push the filament? As in hand-extrude? I think the hotend came preassembled, so there's no way you could incorrectly assemble it, right? If only not sticking ptfe tube deep enough?

My hotend, BTW, got clogged pretty soon, I didn't even try to clean it yet, because I was going to replace it with another hotend I had lying around, so I can't really tell why… I do suspect unevenly cut ptfe or some such, but now that you have similar issue I also consider inadequate cooling using the supplied fan…

Not sure if that was of help except for hinting at possibility that there's nothing wrong you did ;-)

Quote
MrPCatt
Guys, thank you for your help!
So i measured several things again and uploaded a new firmware, mounted a fresh roll of PLA and now I wanted to give it a try to see if I am even close to print something (without Z-Probe).
BUT
The filament is not coming out of the nozzle. Its just dripping ever so slightly, never ever enough to draw a line or so. The extruder motor is just stuttering because the pla is not going forward, something seems jammed or so. The screw for the pressure on the roll of the extruder motor is set to max. Temperature is set to 205. Fan is on all the time. I can pull the pla stripe back out, and it looks normal to me.
Omg, what is this now? Is the extruder jammed even before the first print? How can I fix this?

Is it hard to pull the filament out?
And is it hard to put it back in?

I had a problem where the tip of the filament wasnt cut right and wasnt going into the heat area.
Pull it out a few times and gave it a good push and it went in straight.

Quote
DRTak
I would never know what I know now if I had just bought a pre assembled printer. I actually think I would be bored with the whole idea. But part of the fun and frustration is figuring out calibration and fine tuning.

Thats the reason I decided to go with a kit instead of buying something out of the box. Its all about introducing myself into the matter of 3D printing. The journey is the reward. Actually I don't really have a need for 3D printer parts at all.
But this kit is a bit overkill. I did not want to re engineer everything this with endless hours on google and knowledge bases

Quote
hacker
I wonder can you also push the filament? As in hand-extrude? I think the hotend came preassembled, so there's no way you could incorrectly assemble it, right? If only not sticking ptfe tube deep enough?

Quote
xile6
Is it hard to pull the filament out?
And is it hard to put it back in?

No, i could not push it in our by hand, no chance.
And yes, everything was preassembled....
And i sharpened the tip a little bit.

Not sure what the cause was, but I got it free for now.
Maybe something similar to what DRTak said, a head-crash bend the nozzle or so.
I took the noozle of and cleaned it with the help of a torch and a needle as good as I could.

Quote
DRTak
However, you really really should tightened the hot end nozzle with a wrench or you will leak out filament all around the sides and get burnt PLA around the hot end tip.

Yes I did tighten it with a wrench, and yes, it was still to little and now burned pla is all around the nozzle. Guess I have to take it out again



But guess what, I actually could print something out. What a reward...

Of corse its far from good. My 10mm calibration cube is x/y 11.2mm. , but at least finally something

Quote
MrPCatt

Quote
DRTak
I would never know what I know now if I had just bought a pre assembled printer. I actually think I would be bored with the whole idea. But part of the fun and frustration is figuring out calibration and fine tuning.

Thats the reason I decided to go with a kit instead of buying something out of the box. Its all about introducing myself into the matter of 3D printing. The journey is the reward. Actually I don't really have a need for 3D printer parts at all.
But this kit is a bit overkill. I did not want to re engineer everything this with endless hours on google and knowledge bases

Quote
hacker
I wonder can you also push the filament? As in hand-extrude? I think the hotend came preassembled, so there's no way you could incorrectly assemble it, right? If only not sticking ptfe tube deep enough?

Quote
xile6
Is it hard to pull the filament out?
And is it hard to put it back in?

No, i could not push it in our by hand, no chance.
And yes, everything was preassembled....
And i sharpened the tip a little bit.

Not sure what the cause was, but I got it free for now.
Maybe something similar to what DRTak said, a head-crash bend the nozzle or so.
I took the noozle of and cleaned it with the help of a torch and a needle as good as I could.

Quote
DRTak
However, you really really should tightened the hot end nozzle with a wrench or you will leak out filament all around the sides and get burnt PLA around the hot end tip.

Yes I did tighten it with a wrench, and yes, it was still to little and now burned pla is all around the nozzle. Guess I have to take it out again



But guess what, I actually could print something out. What a reward...

Of corse its far from good. My 10mm calibration cube is x/y 11.2mm. , but at least finally something

As someone told me. Give up on the calibration cube. My is alwasy 10.5mm. Then I did the nickle calibration test and my nickle would not fit or it was too loose. I gave up. I got it close enough to what I want. The true trick was what someone else told me. Make sure your Delta Rod is the correct length.

I got sick of calibration (dont have suitable screws yet to screw on top of the cariages) and decided to get the center calibration as close as possible and melt some plastic.

My first try at a 1x1x1 cube resulted in a huge blob.... turns out i had the filament diameter setting set to 3mm in slicer.
the second try i still get a ton of overextrusion, and a saggy cube. also my extruder is skipping a lot of steps and i dont think it should try to extrude that much plastic.... i cant really figure out whats happening and i feel like i missed some obvious settings... anything in marlin regarding the extruder that should have been edited?

Quote
ThaReaper
anything in marlin regarding the extruder that should have been edited?

E-Steps probably?

The good thing is, you can adjust E-steps in Cura without reflashing the controller. It's in the machine adjustment menu.
-Olaf

I used that sites method... i told the printer to reverse 30mm of filament. The result was 251mm reversed. I think i found out what caused my extruder to skip and my cube to become a blob

Quote
Tha_Reaper
I used that sites method... i told the printer to reverse 30mm of filament. The result was 251mm reversed. I think i found out what caused my extruder to skip and my cube to become a blob

Sounds like you may have the microstepping for the extruder motor set incorrectly on your electronics.

---

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].

I have set the esteps from 760 to 91 and it extrudes pretty close to perfect now. My 10x10x10 cube was 10.2x10.1x10.4 mm and great layer adherance. As far as I know all jumpers were preinstalled on the electronics and I didn't alter that.

91 steps/mm is very low. What sort of extruder do you have? If it is a direct drive (non-geared) extruder, then 91 steps/mm might just be correct. If it is a 3:1 geared extruder then the steps/mm should be more like 400, in which case an actual value of 91 means you have the microstepping set too low.

---

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].

IIRC, Marlin's stock Configuration.h has 760*1.1 for extruder steps? Could it be that you changed it to 91*1.1? That would make 100.1 which is closer to 104.something I got after calibrating. Because 91 indeed sounds a bit low.

P.S. Installed duet today, haven't yet managed a single successful print, but couldn't dedicate much time to it either. The duet produces totally different sounds when moving, I wonder if that is the motor current (which I haven't changed from whatever was the default for minikossel) or is that the result of the lack of segmentation in reprapfirmware?

Quote
dc42
91 steps/mm is very low. What sort of extruder do you have? If it is a direct drive (non-geared) extruder, then 91 steps/mm might just be correct. If it is a 3:1 geared extruder then the steps/mm should be more like 400, in which case an actual value of 91 means you have the microstepping set too low.

The kit has direct drive extruder.

Quote
hacker
IIRC, Marlin's stock Configuration.h has 760*1.1 for extruder steps? Could it be that you changed it to 91*1.1? That would make 100.1 which is closer to 104.something I got after calibrating. Because 91 indeed sounds a bit low.

That's correct. As I don't know a lot about marlin and coding I decided just to change the first part. Didn't know if the *1.1 was some kind of required part...

I still need to fix a couple of things... Like a good bed leveling mechanism and adjustment screws on top of my carriages... I can't find screws that are suitable for that. Also my z probe sucks. It just doesn't want to work and gets stuck every bloody time. I will try more tinkering but I really expect that I'll need to ask here for more help on that soon.

Quote
Tha_Reaper

Quote
hacker
IIRC, Marlin's stock Configuration.h has 760*1.1 for extruder steps? Could it be that you changed it to 91*1.1? That would make 100.1 which is closer to 104.something I got after calibrating. Because 91 indeed sounds a bit low.

That's correct. As I don't know a lot about marlin and coding I decided just to change the first part. Didn't know if the *1.1 was some kind of required part...

I still need to fix a couple of things... Like a good bed leveling mechanism and adjustment screws on top of my carriages... I can't find screws that are suitable for that. Also my z probe sucks. It just doesn't want to work and gets stuck every bloody time. I will try more tinkering but I really expect that I'll need to ask here for more help on that soon.

Aha, 91*1.1 is 100.1, which is much closer to my calibrated value of 104.027. For what I understand this value is allowed to deviate from theoretical, though I'm not sure how far. I think yours might be still a bit off, maybe you still want to do some calibration when you're done with the rest.

I wanted to add adjustments screws on top of the carriages, but given it up and fell back to adjusting offset in software. First in repetier (I believe Marlin also has endstop offsets) and then the whole thing is much easier with DC's implementation for duet board.

I also had to rework the whole z-probe, not it works with the allen key I ad with diameter of 3 and a bit further from the hotend. I'd share my STL, but it's not compatible with sintron's hotend (and I'm not sure if you can print it yet). In short, yes, original z-probe is a bit of disaster, though as far as I can tell by the forum it's possible to make it work. I haven't heard of anyone making it work with original plastic part, though, people tend to employ metal part instead.

Edited 1 time(s). Last edit at 10/30/2015 06:24AM by hacker.

Everything (except for the z-probe) seems to be working now. Im surprised how fast this printer is... Bed gets to 60 degrees in less than a minute. hotend takes a little over 1 minute to hit 200 degrees.
Printing at 120mm/s is no problem, infill at 150. I need to tweak the sizes a little bit as everything seems to be 0.1-0.3 mm too big in random directions.
Now i just need to wait for extra fans to install under the heatbed. on my first bigger print the X tower motor just shut down. probably overheated. my IR thermometer measured 68 degrees in that area. Until i have that i'll just print without a heated bed.
This everything is on the stock 10 amps PSU by the way. It sure get hot, but seems to be working just fine. Is it really necessary to upgrade replace that?
This certainly is an insane upgrade compared to my M3D micro. and that for less money.

What would you all say is the weakest parts of this printer, and i should get some spares just in case they break? I was thinking the stepper drivers... but i don't really have experience with it... for all i know its perfectly possible that they are close to indestructible.

Ok, i take back the things i said about the PSU. It started to randomly shut down of my when it has to power both the bed, and the hotend. I'm going to contact sintron about this. If they dont provide a solution (which i dont really expect) then i'll have to buy a new PSU. 15 or 20 Amps recommended?

I think the PSU has just enough power. When I power my heatbed and hotend it takes a bit of time for ABS. But for PLA its fine.

I ran into a big problem today. My RAMPs got fried. I tried using NO PC just the LCD. It was fine the other day when it was really cold 44 deg in the garage. But today it heated up a bit and I forgot that without a Laptop to run the Kossel you have to manually turn on the fans to the printer. So remember that. When you use the Kossel without a laptop make sure you turn on the fans manually by the LCD screen. Then start the print. Otherwise you will get fried electronics like me. The MOSFETs melted together.

It's well known that an Arduino Mega often struggles to provide enough 5V power to run the backlight of a graphical LCD display. The result is that the voltage on the 5V rail drops, the mosfets don't get enough gate drive and end up frying. I have listed workarounds in several different posts on these forums. The best one is not to use Arduino/RAMPS on the first place.

---

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].

So If I dont use an Arduino Ramps. How do I use my Kossel? What other system can run it.

I already changed from Pronterface to Repetier software wise.

But the pre made Marlin firmware with Arduino works. I asked James from Sintron what other system I could use and he didnt respond?

If there was a more reliable system I would gladly use it.

I really liked the idea that its a stand alone system though. ABS stinks really bad. And like the others I get a headache from it when I use it in the house. So the ability to use it in the garage is a must.

Quote
DRTak
So If I dont use an Arduino Ramps. How do I use my Kossel? What other system can run it.

Arduino/RAMPS has a number of issues, of which the lack of enough current from the 5V rail is the most frequently encountered. My recommendation is to use a modern 32-bit controller board such as Duet or Smoothieboard or one of their clones. These board are available from $50 + shipping including the drivers, although well-made ones not built in China cost more (just like Arduino and RAMPS).

For delta printers, I particularly recommend the Duet 0.8.5 which is available from Think3DPrint3D. It supports a fast auto delta calibration, segmentation-free delta movement, and a built-in web interface. See [miscsolutions.wordpress.com] for how I upgraded my delta from RAMPS to Duet. The calibration instructions in that series are out of date because it's completely automatic now if you have a Z probe.

If you want to spend less, you can get a Replikeo Duet 0.6 for $50 + shipping, although the quality seems to be variable.

If you want to continue using Arduino/RAMPS, see [forums.reprap.org] for how to work around the 5V power problem.

---

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].

So If I upgrade to a duet then I have to buy your Panel DUE to get LCD function right?

Edited 1 time(s). Last edit at 11/01/2015 03:41PM by DRTak.