Folger Tech Kossel Delta 2020 Full 3D Printer Kit w/Auto-Level

Isn't the fan meant to cool the cylindrical heat sink above the hot end?

Quote
paul_delta
I had the same problem. With original fan mount it's hard to get stable temperature above 200C. I designed this sleeve (on the picture - the yellow one), just print it and put between the fan and fan mount: [www.thingiverse.com]

[attachment 67037 folgertech_kossel_fan_mod_view1.jpg]

Quote
Legrand
I had some problems getting above 210°C due to the hot end being too close to the circular radiator and fan (after I switched over to the 3mm head) so I created a small aluminum "foil" shield to keep the "wind" away form the hot end and am now printing at 230°C.

Yes, this fan cools heat sink, but without this adapter it blows also to the heater element and reduce the effectiveness/temperature . The adapter just focuses the air flow on the heat sink.

Quote
Legrand
Isn't the fan meant to cool the cylindrical heat sink above the hot end?

Quote
paul_delta
I had the same problem. With original fan mount it's hard to get stable temperature above 200C. I designed this sleeve (on the picture - the yellow one), just print it and put between the fan and fan mount: [www.thingiverse.com]

[attachment 67037 folgertech_kossel_fan_mod_view1.jpg]

Quote
Legrand
I had some problems getting above 210°C due to the hot end being too close to the circular radiator and fan (after I switched over to the 3mm head) so I created a small aluminum "foil" shield to keep the "wind" away form the hot end and am now printing at 230°C.

Edited 1 time(s). Last edit at 12/04/2015 08:46AM by paul_delta.

So then, in real life, you don't have it set up as shown in the CAD drawing you posted (pointing at the nozzle tip), you have it aimed upward to the heatsink instead.

Quote
paul_delta
Yes, this fan cools heat sink, but without this adapter it blows also to the heater element and reduce the effectiveness/temperature . The adapter just focuses the air flow on the heat sink.

Quote
Legrand
Isn't the fan meant to cool the cylindrical heat sink above the hot end?

Quote
paul_delta
I had the same problem. With original fan mount it's hard to get stable temperature above 200C. I designed this sleeve (on the picture - the yellow one), just print it and put between the fan and fan mount: [www.thingiverse.com]

[attachment 67037 folgertech_kossel_fan_mod_view1.jpg]

Quote
Legrand
I had some problems getting above 210°C due to the hot end being too close to the circular radiator and fan (after I switched over to the 3mm head) so I created a small aluminum "foil" shield to keep the "wind" away form the hot end and am now printing at 230°C.

I'm talking about the yellow adapter. The bottom one - is additional fan for filament.

Quote
Legrand
So then, in real life, you don't have it set up as shown in the CAD drawing you posted (pointing at the nozzle tip), you have it aimed upward to the heatsink instead.

Quote
paul_delta
Yes, this fan cools heat sink, but without this adapter it blows also to the heater element and reduce the effectiveness/temperature . The adapter just focuses the air flow on the heat sink.

Quote
Legrand
Isn't the fan meant to cool the cylindrical heat sink above the hot end?

Quote
paul_delta
I had the same problem. With original fan mount it's hard to get stable temperature above 200C. I designed this sleeve (on the picture - the yellow one), just print it and put between the fan and fan mount: [www.thingiverse.com]

[attachment 67037 folgertech_kossel_fan_mod_view1.jpg]

Quote
Legrand
I had some problems getting above 210°C due to the hot end being too close to the circular radiator and fan (after I switched over to the 3mm head) so I created a small aluminum "foil" shield to keep the "wind" away form the hot end and am now printing at 230°C.

D'oh - oops.

Guess I'll print that after I get done with the spool holder parts I'm making now...

Quote
paul_delta
I'm talking about the yellow adapter. The bottom one - is additional fan for filament.

Quote
Legrand
So then, in real life, you don't have it set up as shown in the CAD drawing you posted (pointing at the nozzle tip), you have it aimed upward to the heatsink instead.

Quote
paul_delta
Yes, this fan cools heat sink, but without this adapter it blows also to the heater element and reduce the effectiveness/temperature . The adapter just focuses the air flow on the heat sink.

Quote
Legrand
Isn't the fan meant to cool the cylindrical heat sink above the hot end?

Quote
paul_delta
I had the same problem. With original fan mount it's hard to get stable temperature above 200C. I designed this sleeve (on the picture - the yellow one), just print it and put between the fan and fan mount: [www.thingiverse.com]

[attachment 67037 folgertech_kossel_fan_mod_view1.jpg]

Quote
Legrand
I had some problems getting above 210°C due to the hot end being too close to the circular radiator and fan (after I switched over to the 3mm head) so I created a small aluminum "foil" shield to keep the "wind" away form the hot end and am now printing at 230°C.

Edited 2 time(s). Last edit at 12/04/2015 09:04AM by Legrand.

Quote
paul_delta
Here is my video.

Just curious, but could you tell me what your current accel/jerk settings are? Want to compare with mine, as I just used similar settings to my cartesian printer as a baseline, but since the physics are very different with a delta I do not think they are optimal. Also curious because I know if they are too aggressive it can cause issues with top speed never being attained.

Quote
paul_delta
Here is my video.

I hope that when my printer is up and running that ti prints that smoothly.

---

Owner of a QUBD Two Up
And
Folger Kossel 2020
And
Tevo Tarantula

Finally I printed this 20x20mm model at 120mm/s. I inspected my Eeprom configuration and realized that my acceleration was too low (300mm/s2) - after experiments to make the printer quieter for night time I forgot to set it back

Also I tried to increase acceleration to find the maximum that printer can handle. For my printer and Folgertech recommended vRef driver voltage it's safe to set acceleration up to 4000 mm/s2. 5000 and higher - 100% head crash because of missed steps.

Update: 4000 mm/s2 acceleration and 35 jerk is too much for Folger Kossel. It's OK if you print in the center of bed, but too much for max available coordinates because it looses steps.
I had to set it to 1000 to print 158mm size model.

The first video, where the speed was 60mm/s (not 120mm/s as expected) the acceleration was only 300mm/s2, Max. jerk 20. Here is the new video, with acceleration 4000 mm/s2 and max. jerk 35.
[youtu.be]

Quote
Vlerherg
Just curious, but could you tell me what your current accel/jerk settings are? Want to compare with mine, as I just used similar settings to my cartesian printer as a baseline, but since the physics are very different with a delta I do not think they are optimal. Also curious because I know if they are too aggressive it can cause issues with top speed never being attained.

Edited 1 time(s). Last edit at 12/05/2015 08:27PM by paul_delta.

Quote
paul_delta
Finally I printed this 20x20mm model at 120mm/s. I inspected my Eeprom configuration and realized that my acceleration was too low (300mm/s2) - after experiments to make the printer quieter for night time I forgot to set it back

Also I tried to increase acceleration to find the maximum that printer can handle. For my printer and Folgertech recommended vRef driver voltage it's safe to set acceleration up to 4000 mm/s2. 5000 and higher - 100% head crash because of missed steps.

The first video, where the speed was 60mm/s (not 120mm/s as expected) the acceleration was only 300mm/s2, Max. jerk 20. Here is the new video, with acceleration 4000 mm/s2 and max. jerk 35.
[youtu.be]

Quote
Vlerherg
Just curious, but could you tell me what your current accel/jerk settings are? Want to compare with mine, as I just used similar settings to my cartesian printer as a baseline, but since the physics are very different with a delta I do not think they are optimal. Also curious because I know if they are too aggressive it can cause issues with top speed never being attained.

Nice! I had very much the same thing happen to me, Experimenting and then promptly forgot that I had changed settings. That's why I mentioned it on the off chance. Sounds like you figured it out before anyway! I'll have to experiment on mine. I have a bigger non-stock dual-fan print cooler, so more mass. Likely will miss steps at lower accel/speeds.

Has anyone tried this kind of effector [www.thingiverse.com] ?
Very interesting solution, where the tip works as z-probe.

Quote
paul_delta
Has anyone tried this kind of effector [www.thingiverse.com] ?
Very interesting solution, where the tip works as z-probe.

I think I've seen something very similar before. It's an interesting concept and idea. I would be very interested in others opinion about it as well.

---

Owner of a QUBD Two Up
And
Folger Kossel 2020
And
Tevo Tarantula

In the Google Drive where the manual is for this printer. I found two sets of firmware, one entitled "KosselRevB" and the other "KosselRevBPE". What's the difference?

---

Owner of a QUBD Two Up
And
Folger Kossel 2020
And
Tevo Tarantula

They both are outdated. The only difference - 2 changes in Configuration.h:

B
#define INVERT_E0_DIR true // for direct drive extruder v9 set to true, for geared extruder set to false

BPE
#define INVERT_E0_DIR false // for direct drive extruder v9 set to true, for geared extruder set to false

B
#define MANUAL_Z_HOME_POS 315// For delta: Distance between nozzle and print surface after homing.

BPE
#define MANUAL_Z_HOME_POS 1120// For delta: Distance between nozzle and print surface after homing.

Quote
Captain_Tim
In the Google Drive where the manual is for this printer. I found two sets of firmware, one entitled "KosselRevB" and the other "KosselRevBPE". What's the difference?

PS: ... which is another fact that Folger Tech doesn't care what a mess they create...

Is there any newer FW by Folger, or is that the latest that they have? Also, I tried what I believe to be your repetier FW paul_delta, and in Repetier host, nothing moves. I'm not sure why. The testing code to make all the motors turn works though.

---

Owner of a QUBD Two Up
And
Folger Kossel 2020
And
Tevo Tarantula

The problem is that all gurus moved out from Mega 2650 to 32bit systems
As far as I know - there are couple of forks of Marlin - with autolevel, extended configuration options, etc. You may try them but they are slower that Repetier, especially if you want to attach LCD/SD to your Mega.

I have a LCD to hook up to it, but I have yet to add it, as I want the printer to be functional first.

---

Owner of a QUBD Two Up
And
Folger Kossel 2020
And
Tevo Tarantula

Actually, if you assembled your Kossel using rev B manual - my configuration of Repetier should works without any changes (you will need to adjust delta radius, diagonal and Z height only). Load my version and try to run G28 (home).

PS: and endstops first, of course

Update: 4000 mm/s2 acceleration and 35 jerk is too much for Folger Kossel. It's OK if you print in the center of bed, but too much for max available coordinates because it looses steps.
I had to set it to 1000 to print 158mm size model.

Quote
paul_delta
Also I tried to increase acceleration to find the maximum that printer can handle. For my printer and Folgertech recommended vRef driver voltage it's safe to set acceleration up to 4000 mm/s2. 5000 and higher - 100% head crash because of missed steps.

Newcomer here.

I recently got the Folger Tech Kossel printer. I was able to put it together and it mostly works.

However I did hit a roadblock that I cannot get past. The steppers seem to move slightly more up than down.

If I home the printer, come down to Z 1 and step down to 0 it works every time and it is accurate to 0.1 mm.

If I go up to say Z 5 and then come back down to 0, it always ends up above the bed surface. This problem seems to exist in *any* movement up so naturally positioning is completely off once the printer makes a few moves.

I am using Repetier 0.92 and Repetier host.

Any help would be highly appreciated guys, thanks in advance!

Did you check your drivers vRef voltage? Looks like there is no enough current to lift everything and you have skipped steps.
Before checking your drivers, set your travel acceleration in eeprom to very low value, like 5-10 and check if there is any change in movement precision.

Edited 3 time(s). Last edit at 12/06/2015 04:10PM by paul_delta.

Paul, thanks for the response. I did not check the voltages before so I just set them after your suggestion.

Before they were all around 650 mV, now they are X = 356, Y = 360, Z = 358 and the extruder 558 mV. (That potentiometer is not sized right for these low voltages, it is very hard to not change a lot with the slightest movement.)

I changed the acceleration from 20 down to 10 and 5. The problem at the bottom remains, if I go down to Z = 0, come up and go back down, I end up ~0.1 mm above the surface.

I did another - more accurate - test with a dial caliper and surprisingly I do not see the same problem with that. Naturally I cannot get the caliper at Z = 0 so I was doing experimentation around Z = 125 mm.

If I move the hot end up and down, it comes back to within +/- 0.001 inch of the original position. So the problem may have nothing to do with the servos. (This is true for 20 mm/s^2 acceleration as well.)

For the setup please see this picture:



Is it possible that the Z probe is somehow interfering near the plate? I mean when it triggers, can that have an effect on how low will the hot end go?

I think 0.02mm accuracy is good for this kind of construction. Some ball joints backlash may affect accuracy - I had similar problem near tower Y exactly because of ball joint. I used rubber bend wrapped around diagonal rods and it fixed the problem (almost). Pulley slipping may affect accuracy too. I used dremel to grind all motor shafts slightly to avoid slipping (but it may bring more play if you will not perfectly align the screw with the flats).

There is nothing wrong with z-probe (in this case). It doesn't affect regular operations because firmware checks z-probe condition only during calibration process/commands.

Quote
Granite
If I move the hot end up and down, it comes back to within +/- 0.001 inch of the original position. So the problem may have nothing to do with the servos. (This is true for 20 mm/s^2 acceleration as well.)
...
Is it possible that the Z probe is somehow interfering near the plate? I mean when it triggers, can that have an effect on how low will the hot end go?

Did your kit come with belt tensioning springs? Where did you install them?

I had a similar problem, and found that the belt was tighter from the carriage down then from the carriage up, especially when the effector was at Z0. In other words, the motor is pulling at the carriage to keep the belt tight, and the belt is looser as it goes up to the top of the frame where it goes over the bearings and back down the other side.

The tensioner on mine is on the belt just above where the belt attaches to the carriage so it isn't in the way when at Z MAX.

Rick

Quote
llamatrails
Did your kit come with belt tensioning springs? Where did you install them?

I had a similar problem, and found that the belt was tighter from the carriage down then from the carriage up, especially when the effector was at Z0. In other words, the motor is pulling at the carriage to keep the belt tight, and the belt is looser as it goes up to the top of the frame where it goes over the bearings and back down the other side.

The tensioner on mine is on the belt just above where the belt attaches to the carriage so it isn't in the way when at Z MAX.

Rick

I put my tensioning springs under the carriages, just under where you loop the belt around it.

---

Owner of a QUBD Two Up
And
Folger Kossel 2020
And
Tevo Tarantula

Quote
Captain_Tim

Quote
llamatrails
Did your kit come with belt tensioning springs? Where did you install them?

I had a similar problem, and found that the belt was tighter from the carriage down then from the carriage up, especially when the effector was at Z0. In other words, the motor is pulling at the carriage to keep the belt tight, and the belt is looser as it goes up to the top of the frame where it goes over the bearings and back down the other side.

The tensioner on mine is on the belt just above where the belt attaches to the carriage so it isn't in the way when at Z MAX.

Rick

I put my tensioning springs under the carriages, just under where you loop the belt around it.

That is where I had them originally, as the build docs for my B version didn't even mention them. When at Z0 there isn't much flex in the belt from the carriage to the motor as the distance is very short compared to the distance from the carriage to the top of the frame which has lots of flex in the belt. The spring is to help minimize that backlash.

Rick

I added paul_delta's cold end fan shroud, as I couldn't get my hotend over 200 with the fan blowing over everything. I printed a few things at 210, and while I still had stinging, they turned out amazing. I haven't tried to print anything above 210- that's the point where my Hatchbox PLA seems to instantly liquefy. I had to reverse about 20mm to keep it from just oozing out of the nozzle at idle. It's a tricky first couple layers at 210 as it is. Above that will be on my list.

I then printed a combo fan duct (here), hoping that active cooling would help my stringing- nope. make it worse. Instead of fiberglass fine strings, I would get blobs on the edges instead. So I ended up taping up that bottom port, and will be going back to paul's shroud.

I went down in temp as a test as well. Printed a few test pieces at 170- that helped stringing more than going up to 210, but then had issues with line width.

The battle continues....

I'd suggest to run PID calibration procedure after fan modification. The PID coefficient changes significantly after installation and re-calibration helps to make the temperature as stable as possible.

Quote
fattmann
I added paul_delta's cold end fan shroud, as I couldn't get my hotend over 200 with the fan blowing over everything. I printed a few things at 210, and while I still had stinging, they turned out amazing. I haven't tried to print anything above 210- that's the point where my Hatchbox PLA seems to instantly liquefy. I had to reverse about 20mm to keep it from just oozing out of the nozzle at idle. It's a tricky first couple layers at 210 as it is. Above that will be on my list.

I then printed a combo fan duct (here), hoping that active cooling would help my stringing- nope. make it worse. Instead of fiberglass fine strings, I would get blobs on the edges instead. So I ended up taping up that bottom port, and will be going back to paul's shroud.

I went down in temp as a test as well. Printed a few test pieces at 170- that helped stringing more than going up to 210, but then had issues with line width.

The battle continues....

I tried this fan duct before I decided to make my own. It doesn't work - the draft from open door works better

Quote
fattmann
I then printed a combo fan duct (here), hoping that active cooling would help my stringing- nope. make it worse. Instead of fiberglass fine strings, I would get blobs on the edges instead. So I ended up taping up that bottom port, and will be going back to paul's shroud.

Edited 1 time(s). Last edit at 12/08/2015 01:39PM by paul_delta.