Ebay SINTRON Kossel Mini full kit

Quote
dc42
If the bed is matt black and you have no glass on top of it, then you may not get good results from the sensor, because matt black paint will probably not reflect enough IR. The sensor will work on a dark grey surface such as BuildTak, but not black. I recommend matt black paint on the aluminium when you will be putting glass on top of it, because then the sensor can see the reflection from the top surface of the glass without being swamped by a stronger reflection from the aluminium below.

Remember, the only purpose of the H parameters is to account for different trigger heights of the Z probe at different points on the bed. So only use H parameters if you measure the probe trigger height at each point, as I described at [forums.reprap.org]. However, if you get differences in trigger height greater than about 0.2mm then it probably means that you have serious effector tilt, and you need to fix that first.

.
Uh oh. Well. If I paint it with white oven paint? Will that help

Or I can use glass or buildtak. I read your blog and you said u will not use buildtak out of anger? What happened? I have the same thought with some places. If they are not nice. I wont buy from them again? In my experience I think I will use glass. I have issues with my nozzle grazing the build plate (when I had the analog z probe). Thats why I have moved away from Kapton like you did. So I would have to repaint the plate Gloss black? and add glass to the top?

I noticed an error with the web interface today. After I made my own macros under Default G codes. +Add g code. I lost web connection. The macro I made was basic. It was G1 X0 Y0 Z1
I named it HomeZero. I hit apply. Used the gcodes after one calibration session. Then lost web interface. It said cannot connect to duet firmware? So I deleted the cache in Chrome and it worked again. When the web interface opened up. The macros were deleted I assume when I deleted the chrome web cache.

With auto calibration x 3. I have it down to (Z height above bed): Nozzle at center 0.2, X tower: 0. Y tower: 0. Z tower: .05. The nozzle height at center never changes. It is always 0.2. The other heights change by up to 0.1mm. Any way to fix the z height at center? Add another g30 probe point at say x2 y2?

Quote
DRTak
... I read your blog and you said u will not use buildtak out of anger? What happened? I have the same thought with some places. If they are not nice. I wont buy from them again? In my experience I think I will use glass. I have issues with my nozzle grazing the build plate (when I had the analog z probe). Thats why I have moved away from Kapton like you did.

Sorry, I used colloquial English without considering whether non-native speakers of English would understand it. To do something "in anger" means to have a proper attempt at doing it. So what I meant was that I only used BuildTak very briefly, not enough to consider it a proper test.

Quote
DRTak
I noticed an error with the web interface today. After I made my own macros under Default G codes. +Add g code. I lost web connection. The macro I made was basic. It was G1 X0 Y0 Z1
I named it HomeZero. I hit apply. Used the gcodes after one calibration session. Then lost web interface. It said cannot connect to duet firmware? So I deleted the cache in Chrome and it worked again. When the web interface opened up. The macros were deleted I assume when I deleted the chrome web cache.

Maybe you need to update the files in /www on the SD card? The latest version is 1.07 available at [github.com].

Quote
DRTak
With auto calibration x 3. I have it down to (Z height above bed): Nozzle at center 0.2, X tower: 0. Y tower: 0. Z tower: .05. The nozzle height at center never changes. It is always 0.2. The other heights change by up to 0.1mm. Any way to fix the z height at center? Add another g30 probe point at say x2 y2?

The calibration routine tries to minimise the sums of the squares of the errors over all the probe points. If it is leaving a large error at the centre, this means that it can only reduce that error by increasing the errors at other points. For now I suggest you use 7-point calibration and reduce the probing radius by e.g. 50mm, and see if that gives better results. You can use the calculator at [escher3d.com] to generate the new probe points.

---

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].

dc42,

I have a question. How does the IR sensor know where the nozzle is in space. I have spaced the IR board 1.5mm above the nozzle tip. But how would the IR board know that?

What if I placed it 2mm above? How would it know?

I guess what Im asking is.. did I overlook that simple equation. Type in some value of what the nozzle height is in relation to the IR board?

Quote
DRTak
dc42,

I have a question. How does the IR sensor know where the nozzle is in space. I have spaced the IR board 1.5mm above the nozzle tip. But how would the IR board know that?

What if I placed it 2mm above? How would it know?

I guess what Im asking is.. did I overlook that simple equation. Type in some value of what the nozzle height is in relation to the IR board?

I think you mean M558's H-parameter in your config.g (Oops, not quite right)

Edited 1 time(s). Last edit at 12/12/2015 05:40PM by hacker.

Quote
DRTak
dc42,

I have a question. How does the IR sensor know where the nozzle is in space. I have spaced the IR board 1.5mm above the nozzle tip. But how would the IR board know that?

What if I placed it 2mm above? How would it know?

I guess what Im asking is.. did I overlook that simple equation. Type in some value of what the nozzle height is in relation to the IR board?

You have to measure the trigger height (i.e. nozzle height above the bed when the probe triggers) and enter that value as the Z parameter the G31 command in config.g. See the Testing & Commissioning section of [miscsolutions.wordpress.com].

Edited 1 time(s). Last edit at 12/12/2015 04:43PM by dc42.

---

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].

Quote
dc42

Quote
DRTak
dc42,

I have a question. How does the IR sensor know where the nozzle is in space. I have spaced the IR board 1.5mm above the nozzle tip. But how would the IR board know that?

What if I placed it 2mm above? How would it know?

I guess what Im asking is.. did I overlook that simple equation. Type in some value of what the nozzle height is in relation to the IR board?

You have to measure the trigger height (i.e. nozzle height above the bed when the probe triggers) and enter that value as the Z parameter the G31 command in config.g. See the Testing & Commissioning section of [miscsolutions.wordpress.com].

Ah. That is right. The G92 Z0 command. Then trigger height. Thus the IR probe knows where it is in space.

I have been getting some odd USB sounding resets on my PC. I also could not heat the hotend and bedplate AND calibrate at the same time. It sounded like a grounding issue or miss wiring of some sort. Sometimes it would disconnect me from the web interface randomly. So I removed the duet0.6 from the case. Rechecked all the wiring. Turns out (I dont really know if this is the issue) but BED_TEMP GRND is on the left of the board NOT right as pictured here: [blog.think3dprint3d.com]

So I have moved the Flylead OUT to BED_TEMP GRND (on the left side NOT the right) to the DUEX4 to the GRND VSSA lead. Now i just need to reassmble it all and test it. I think that the flylead hooked to BED_TEMP BED R might have been the issue. But according to the above picture that is how they said to connect it to the duex4.

I also upgrade the web interface to 1.07. But I cannot use any macros or it crashes still. Odd. Especially this one: G1 X-72 Y-42 Z1. That macro will make the web interface not work anymore. Only after I delete the Chrome cache it works.

Edited 1 time(s). Last edit at 12/13/2015 02:58PM by DRTak.

Did you check the housing of the USB socket? Some reported bad solder jobs on duet0.6 from Replikeo. ( Mine was OK and obviously resoldered by hand.)
-Olaf

Quote
o_lampe
Did you check the housing of the USB socket? Some reported bad solder jobs on duet0.6 from Replikeo. ( Mine was OK and obviously resoldered by hand.)
-Olaf

Yes my usb socket is ok. Its from think3dprint3d. I thought maybe it was a bad connection of the usb also. But in the end. It was the GND issue. I have tested it now. It all works. The ground issue fix also fixed the web interface issue so far. I can make my own macros and they work. No more crashes. Oddly. The grounding issue did ruin my 2gb SD card that came with my duet0.6. I replaced it. Thank goodness I backed up my files.

Also, I was having color hue issues with my paneldue. But the correct gnd flylead sort of fixed it. Its better but still there.

Now my issue is with my 3 thermistors. I have a hotBED therm, hotend therm, and a backup hotend therm. I think the main hotend therm is not correct. I got it from ebay. Its a 104GT2. I put their R values in and it didnt look right. So I adjusted them below:
M305 P0 T10000 B3950 R1000 H50 L0 ; Put your own H and/or L values here to set the bed thermistor ADC correction.
M305 P1 T100000 B4267 R1000 H-5 L0 ; Put your own H and/or L values here to set the first nozzle thermistor ADC correction
M305 P2 T100000 B3974 R1000 H30 L0 ;

I feel that the hotbed temp is correct. So I tried to match the hotend temp to the hotbed temp. But something isnt right.. Maybe my negative H value?

Next issue is steps/mm for the duet0.6. Default is 663 for M92. I tested it and just like with the RAMPS its 100. Am I wrong? I did the basic check. Mark filament. Push 10mm extrude. With M92 at 100 the amount extruded is around 10mm per my calipers.

Edited 2 time(s). Last edit at 12/13/2015 04:54PM by DRTak.

Regarding the thermistors, in theory all 3 channels should need the same H parameter. Any other discrepancies are likely to be due to thermistor tolerances. It's normal for the hot bed and hot end to give readings a degree or three different at room temperature.

The extruder steps/mm of 663 in the config.g file you are using is about right for the mini geared extruder used on the T3P3 Mini Kossel kit. If your extruder is direct drive, then 100 would be about right.

If you are getting colour hue issues with PanelDue, and it is not because you are viewing the display from a wide angle, then I suggest you check the 5V power feed into the PanelDue in case the voltage is 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].

Quote
dc42
Regarding the thermistors, in theory all 3 channels should need the same H parameter. Any other discrepancies are likely to be due to thermistor tolerances. It's normal for the hot bed and hot end to give readings a degree or three different at room temperature.

The extruder steps/mm of 663 in the config.g file you are using is about right for the mini geared extruder used on the T3P3 Mini Kossel kit. If your extruder is direct drive, then 100 would be about right.

If you are getting colour hue issues with PanelDue, and it is not because you are viewing the display from a wide angle, then I suggest you check the 5V power feed into the PanelDue in case the voltage is too low.

Well all 3 thermistors are different. I bought a screw in type thermistor for my Kossel hotend on ebay. Thats the 104gt2. The specs say its 100k and the R value given. It just broke. It was attached with a very thin cable. I just ordered some more. I ordered more. One screw on type from a different supplier. And spare bare thermistors from another vendor.

The hotplate thermistor is from what I can tell 4.7k pullup 10k resistance. I guessed the B value by looking at similar thermistors online. I decided to calibrate it with the ice water test. I put a spare exact 10k thermistor (not compatible with temps over 140deg C. So cant use it on hotend) in ice water. Used a digital meat thermometer and it read 2deg C. Calibrated the H value until it read 2.2deg C on the web interface for the duet.

The 2nd backup nozzle thermistor is from Sintron. It came with the kit. I dont know the specs on it. I guessed its 100k. I had the correct info for the RAMPS but cant get it set correctly for Duet. From the Marlin firmware in arduino. I found this: 1 is 100k thermistor - best choice for EPCOS 100k (4.7k pullup). I am using this info for the backup thermistor. It is pretty erratic in its measurements. Will replace with the spares I bought.

The Extruder motor is from Sintron. It uses GT2 gears. MK8 extruder.

I hooked up 12v to the duex4. Did not help with the odd color changes on the paneldue.

Edited 3 time(s). Last edit at 12/15/2015 08:15AM by DRTak.

Hey @all,

has anyone a tipp for me: the pom-rollers on the carriages have a lot of play on my kossel mini. how to solve that? they have so much play, that the force from the bowden tube tilt the effector.
other bearings? other wheels? order some linear-rails?

thanks

Quote
Rossini
Hey @all,

has anyone a tipp for me: the pom-rollers on the carriages have a lot of play on my kossel mini. how to solve that? they have so much play, that the force from the bowden tube tilt the effector.
other bearings? other wheels? order some linear-rails?

thanks

If it is the usual printed carriage with three wheels, there is a tensioning screw going right through the carriage near the top that you can tighten up. Also check that the wheels are wide enough so that the V-part does not sink completely into the slot of the extrusion.

Edited 1 time(s). Last edit at 12/15/2015 07:29AM by dc42.

---

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].

Quote
dc42

Quote
Rossini
Hey @all,

has anyone a tipp for me: the pom-rollers on the carriages have a lot of play on my kossel mini. how to solve that? they have so much play, that the force from the bowden tube tilt the effector.
other bearings? other wheels? order some linear-rails?

thanks

If it is the usual printed carriage with three wheels, there is a tensioning screw going right through the carriage near the top that you can tighten up. Also check that the wheels are wide enough so that the V-part does not sink completely into the slot of the extrusion.

Dave

Like a lot of the very cheap Chinese Delta kit they seem to use cheap Shower door type roller wheels on the carriages with only a single bearing.

From what I read in here it is a common problem with them wearing out quite quickly and a lot of user's either give up in frustration or upgrade to decent roller's such as the Openbuilds mini Vee wheels and Robotdigg do a cheap copy of them for around $1.50 each Mini Vee Wheels

and along with the eccentric spacers make for a easy adjustable carriage design.

Doug

well, i tightend the carriage. But the play comes from the bearings i think. yes it's a single bearing.
I considered the mini-v-wheels but are they compatible with the sintron extrusions, as they are not really v-slot extrusions and shipping costs from robotdigg are very expensive for austria.

Quote
dougal1957

Quote
dc42

Quote
Rossini
Hey @all,

has anyone a tipp for me: the pom-rollers on the carriages have a lot of play on my kossel mini. how to solve that? they have so much play, that the force from the bowden tube tilt the effector.
other bearings? other wheels? order some linear-rails?

thanks

If it is the usual printed carriage with three wheels, there is a tensioning screw going right through the carriage near the top that you can tighten up. Also check that the wheels are wide enough so that the V-part does not sink completely into the slot of the extrusion.

Dave

Like a lot of the very cheap Chinese Delta kit they seem to use cheap Shower door type roller wheels on the carriages with only a single bearing.

From what I read in here it is a common problem with them wearing out quite quickly and a lot of user's either give up in frustration or upgrade to decent roller's such as the Openbuilds mini Vee wheels and Robotdigg do a cheap copy of them for around $1.50 each Mini Vee Wheels

and along with the eccentric spacers make for a easy adjustable carriage design.

Doug

Good to know, thanks. Forewarned is forearmed, although my carriages are alright so far, guess I should be prepared

The mini V wheels are available from many different suppliers, for example think3dprint3d in the UK. Search ebay or google for "delrin v-wheel". You don't need V-slot to use them, they are OK with standard slot extrusion. I reckon 6mm is the widest slot width they would work with.They need two 623ZZ bearings each.

Edited 2 time(s). Last edit at 12/15/2015 07:57AM by dc42.

---

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].

Quote
dc42
The mini V wheels are available from many different suppliers, for example think3dprint3d in the UK. Search ebay or google for "delrin v-wheel". You don't need V-slot to use them, they are OK with standard slot extrusion. I reckon 6mm is the widest slot width they would work with.They need two 623ZZ bearings each.

Dave they don't all use 3mm bore bearing's the Openbuild ones (and the Robotdigg ones) use a 5mm bore 10mm od bearing 4mm thick bearing such as an MR105zz. AFAIK only the T3P3 ones use the 623 bearing's

Having thought a little more about this I realise that 623zz's will fit the Robotdigg/Openbuilds ones as well Doh a brain fade moment mind I think I would rather have a 5 mm bore and the correspondingly stronger/more rigid M5 screw/bolt (Flathead ones in this case and of course a 1mm thick shim between the bearing's)

Doug

Edited 1 time(s). Last edit at 12/15/2015 08:37AM by dougal1957.

Quote
Rossini
well, i tightend the carriage. But the play comes from the bearings i think. yes it's a single bearing.
I considered the mini-v-wheels but are they compatible with the sintron extrusions, as they are not really v-slot extrusions and shipping costs from robotdigg are very expensive for austria.

That might be my effector tilt issue. My x and y tower have loose carriage. I tightened them to the max. But couldn't tighten them anymore. Still had play in them. I have issue with one of the three wheels in each carriage. The same wheel that is near the carriage wheel tensioner screw. But thought it was my imagination. Also tightened the wheel bolt but and it still had play in it. I checked Sintrons ebay description. It says Inner diameter 6mm. OD 22.5mm and width of 7mm. Bearing 696zz.

Edited 2 time(s). Last edit at 12/15/2015 01:06PM by DRTak.

Quote
Rossini
well, i tightend the carriage. But the play comes from the bearings i think. yes it's a single bearing.
I considered the mini-v-wheels but are they compatible with the sintron extrusions, as they are not really v-slot extrusions and shipping costs from robotdigg are very expensive for austria.

there is a drawing of the Openbuilds ones at Click on the Drawings tab Robot dig also has the same drawing. They will run in anything between 6mm to 8.5 mm wide slots.

I believe the Think3Dprint3D ones will run in smaller slots Dave (DC42) would be better placed to answer that and they may be a better fit depending on what extrusion Sintron use I suspect it will be the usual so called OpenBeam (Not to be confused with OpenBuilds).

Ooznest in the UK do the wheels as well but they are around 3 times the price of the Robotdigg ones which I use in conjunction with Motedis Type B 20x20 series extrusion's and also with The Openbuilds stuff for which they are really designed for (Ooznest also sell that).

HTH

Doug

Quote
Rossini
Hey @all,

has anyone a tipp for me: the pom-rollers on the carriages have a lot of play on my kossel mini. how to solve that? they have so much play, that the force from the bowden tube tilt the effector.
other bearings? other wheels? order some linear-rails?

thanks

I had this problem as well.

I made some videos and sent them to Sintron, they sent me some more wheels and I used the 9 best bearings on my printer. I still wasn't happy with that, so hooked up three rubber bands between the carriages so that they preloaded the bearings. This appeared to work OK, but me being me I wasn't totally happy with it.

Then a mate was upgrading his delta and had some surplus RobotDigg linear slides so I've put those on now. The slides are great, they use tiny little ball bearings and there is no, repeat zero, slop. I'm still getting a bit of ringing now, but I think that this is more to do with elasticity in the belts than anything else now.

That said, the biggest upgrade you can make is to add a PEI bed.

After that, play with every setting in your slicer that it's possible to tweak. Every single one will affect your print, for better or worse, and just about every default wasn't right for me.

Good luck with your journey.

Quote
Rossini
well, i tightend the carriage. But the play comes from the bearings i think. yes it's a single bearing.
I considered the mini-v-wheels but are they compatible with the sintron extrusions, as they are not really v-slot extrusions and shipping costs from robotdigg are very expensive for austria.

Im curious what you are going to do. I believe that I have the same issue. But if I upgrade to the robotdigg bearings or delrin. They are 5mm ID. Which means that the current carriage from sintron would have a lot of play in them? And therefore, I would have to buy the whole carriage kit (carriage and wheels and bearings). That is the same cost as the linear rails ( I would have to buy the matching carriage though to attach to the linear rail bearings).

I've got the t3p3 rollers and printer new carriers to fit the smaller M3 screws.
I don't like it now for two reasons:
The M3 screw is bending outwards under the pressure of the carrier tensioner.
The bearings don't come with a shim between, so you can't tighten them without severe bearing binding.
I had to use Nyloc nuts and adjust play vs. binding individually.

BTW: The t3p3 rollers are made for Nut-5 extrusions. When I changed my setup to robotdigg corners, I had to use wider Nut-6 extrusions ( or dremel the slots to fit in the corners )
Now the rollers are to narrow.
-Olaf

Quote
o_lampe

BTW: The t3p3 rollers are made for Nut-5 extrusions. When I changed my setup to robotdigg corners, I had to use wider Nut-6 extrusions ( or dremel the slots to fit in the corners )
Now the rollers are to narrow.
-Olaf

What are the sintron extrusions? Now i am confused. Are the Robotdigg metal corners compatible with the Sintron extrusions or not. Do i have to dremel the slots?


For the rollers: well i think i buy the mini-v-rollers and print matching carriages. Maybe these: [www.thingiverse.com]

Quote
Rossini

Quote
o_lampe

BTW: The t3p3 rollers are made for Nut-5 extrusions. When I changed my setup to robotdigg corners, I had to use wider Nut-6 extrusions ( or dremel the slots to fit in the corners )
Now the rollers are to narrow.
-Olaf

What are the sintron extrusions? Now i am confused. Are the Robotdigg metal corners compatible with the Sintron extrusions or not. Do i have to dremel the slots?


For the rollers: well i think i buy the mini-v-rollers and print matching carriages. Maybe these: [www.thingiverse.com]

Measure the width of the slot in your extrusions if it is 5 mm you will need to dremel them if it is 6mm wide then it will fit as is

crap i hope they are 6 mm (can't measure them now)


checked ... they are 6 mm puuh

Edited 1 time(s). Last edit at 12/17/2015 05:16AM by Rossini.

Quote
Rossini
crap i hope they are 6 mm (can't measure them now)

Mine seem to be 6. But do check ;-)

Mine definitely are 6mm. Because I'm also using Robodigg's corners. And they fits without any "Dremel work".

Well my Y carraige is snapped in half and barely hanging onto the 2020 rails. I tested the wheels. 3 of the wheels have bad bearings. This led me to buy a V roller wheel on ebay that includes molded ABS carriages. I realize that it is one step down from metal and one step up from printed parts. But my printer is out of commission until it arrives from another country.

I am getting closer and closer to a complete upgrade of the Kossel. Soon it will be metal effector and metal corners. LOL. And then metal carriages.

In the interim, I have decided to move all the electronics to the top of the kossel. So I have included the pictures. I also put in a picture of my wooden enclosure. But it is too big to put anywhere except in the garage and is missing a door.

Edited 1 time(s). Last edit at 12/20/2015 11:35AM by DRTak.

Hello,
Thank you for providing a link to this part on Thingiverse. I had a problem where one of the diagonal-rod ball-ends binding with the original built-in fan-mount. I printed this one, on my other 3D printer, and it installed fine, and that solved my binding problems.
regards.