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

Try to find some old dead desktop computers. They usually have a few hanging around in them somewhere.

Quote
Sven.L
Hi,

I cannot get anywhere those 2-pin connectors, I went to more than 15 shops earlier this day. But I purchased, in hope they can fit, these parts you see on the attached photo below. Let's see and hope, at least one of these two types can fit.

[attachment 72127 Photo7492.jpg]

Quick question, why is it that when I do a Height Map and set the Min and Max of both X and Y to 70 and -70 respectively, I can do the map no problem, but if I change the values to 80 and -80, it will scrap the bed as it checks the point.


P.S Just for show and tell, I've attached my height map for X Min: -70, Y Min: -70, X Max: 70, Y Max: 70 after I've gone through G29 S2 to add zCorr. It looks great to me, what's your thoughts on it?

---

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

Having my printable radius 70mm I have to set height map min/max coordinates to -48/48 to fit boundaries. Your coordinates are definitely out of max. available.

Your map height difference is almost 8mm (!). Probably because your measured square is too big. Try to measure -60/60.

Quote
Captain_Tim
Quick question, why is it that when I do a Height Map and set the Min and Max of both X and Y to 70 and -70 respectively, I can do the map no problem, but if I change the values to 80 and -80, it will scrap the bed as it checks the point.


P.S Just for show and tell, I've attached my height map for X Min: -70, Y Min: -70, X Max: 70, Y Max: 70 after I've gone through G29 S2 to add zCorr. It looks great to me, what's your thoughts on it?

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

PS: I recommend to disable z-corr (distortion correction) forever and spend more to calibrate your printer manually or with the help of Dc42 calculator.

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

With it set to 60 instead, this is the result.

---

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

Quote
Captain_Tim
With it set to 60 instead, this is the result.

Looks like you didn't calibrate your printer at all 7.69-4.52= 3.17mm height fluctuation across measured square...
For example here is one of my test heat maps, right after 15 min manual calibration and one DC42 calculator iteration.
[www.youtube.com] (look at results at 1:16 minute).
As you can see - the height fluctuation is 0.15mm across measured square. And this is not the best, actually.

Edited 1 time(s). Last edit at 02/08/2016 10:45PM by paul_delta.

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

PS: if you have standard FT probe mount with big offset from the nozzle - this is the results of effector tilt. As I said many times - FT probe mount is piece of junk. High probe offset on delta printer will never provide good calibration.

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

Quote
paul_delta

Quote
Captain_Tim
With it set to 60 instead, this is the result.

Looks like you didn't calibrate your printer at all 7.69-4.52= 3.17mm height fluctuation across measured square...
For example here is one of my test heat maps, right after 15 min manual calibration and one DC42 calculator iteration.
[www.youtube.com] (look at results at 1:16 minute).
As you can see - the height fluctuation is 0.15mm across measured square. And this is not the best, actually.

Well, i'm guilty of not really doing any manual calibrations. I had it calibrated manually last week, on Friday I believe, but because I had some extrusion covers near the base of the towers when I was probing, it messed up the readings, and those got saved to the EEPROM. I wasn't really home all weekend so I didn't have a chance to really redo my calibrations. I thought I'd just try this method, but it seems people disprove of the easy way. Tomorrow, I'll more then likely re-calibrate it. It didn't take me to long on Friday, about 45 minutes with everything new in the EEPROM. I shall report back tomorrow with a new height map... hopefully.

Also, is DC42's calculator this one: Calculator, if so, how do I specify the 7 points when I want to test it out? Is that just a case of moving the nozzle to the desired location, then just simply doing a G30, then take the nozzle Height Error from that reading, adding it to the calculator, and hitting the calculate button, rinse, wash, repeat?

---

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

Quote
paul_delta
PS: if you have standard FT probe mount with big offset from the nozzle - this is the results of effector tilt. As I said many times - FT probe mount is piece of junk. High probe offset on delta printer will never provide good calibration.

I don't have a way to make a new probe mount at the moment. My Two Up is out of commission till I fix the Y Drive (I think there's just to much tension on the belt, and the motor isn't getting enough juice).

---

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

I communicated with Folger Tech, and they told me to use the triple wire (white, red, black) to connect to the servo triple wire (orange, red, brown), and to cut off then the white wire, so only the red and black wire are connected to the switch. While the PDF manual speaks about a 2-pin connector, in reality it is a 3-pin connector and a white-red-black wire (not the black-red wire as mentioned in their PDF). With such wrong information, it is no wonder that I had all these problems...

But the resulting wire length of both the connected orange-red-brown and white-red-black wire is now 50 inch. Still wondering if that makes sense...

Quote
Folger Tech
For the needed sensor cable, you would use the 3pin Red,White, and black cable (...).
Following the wiring diagram, the white will be cut and taped as only 2 wires are needed.

Edited 2 time(s). Last edit at 02/09/2016 12:51AM by Sven.L.

The picture you provided shows the servo connected to extension cable, that must be connected to RAMPS servo port as is - without any cutting - all 3 pins.
If you will cut the white wire on this picture, the servo will not work...

PS: As I told you two pages ago, you need to connect the servo directly to RAMPS servo port and mechanical switch to the ground (-) and Signal pins of RAMPS's Z-min

Quote
Sven.L
I communicated with Folger Tech, and they told me to use the triple wire (white, red, black) to connect to the servo triple wire (orange, red, brown), and to cut off then the white wire, so only the red and black wire are connected to the switch. While the PDF manual speaks about a 2-pin connector, in reality it is a 3-pin connector and a white-red-black wire (not the black-red wire as mentioned in their PDF). With such wrong information, it is no wonder that I had all these problems...

But the resulting wire length of both the connected orange-red-brown and white-red-black wire is now 50 inch. Still wondering if that makes sense...

Quote
Folger Tech
For the needed sensor cable, you would use the 3pin Red,White, and black cable (...).
Following the wiring diagram, the white will be cut and taped as only 2 wires are needed.

[attachment 72143 Photo7493.jpg]

Edited 3 time(s). Last edit at 02/09/2016 01:27AM by paul_delta.

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

Again, check RAMPS Wiki - it explains mechanical endstops pretty clear, with pictures [reprap.org]

Edited 1 time(s). Last edit at 02/09/2016 02:05AM by paul_delta.

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

For Rev. B inductive sensor wiring .

I think I know what is going on with your wiring. That extra cable is intended to be cut. You use the small female end for connecting to your board, the other female end (the one in your pic is not used) First you strip back some of that thick black cable coming from your inductive bed sensor, then you solder the two resisters, I think 10k&15k ohm, between the black and the red (look at the diagram to be sure of that setup) , the white is not used and can be taped back the blue and brown run to your 12v power supply. Brown is + and blue is -. You use the small male end with its Frankenstein resistor combo to connect to the board with the white wire (check for sure it was late when I completed that step) to the outside facing away from the servos, it goes between the two end stop sensor connections . I was confused until I figured it out. My sensor lights up and senses the bed well with this setup. Hopefully my explanation will help you make sense of the folgertech wiring diagram, some of the trace lines are kinda muddy and difficult to read.

Quote
Sven.L
I communicated with Folger Tech, and they told me to use the triple wire (white, red, black) to connect to the servo triple wire (orange, red, brown), and to cut off then the white wire, so only the red and black wire are connected to the switch. While the PDF manual speaks about a 2-pin connector, in reality it is a 3-pin connector and a white-red-black wire (not the black-red wire as mentioned in their PDF). With such wrong information, it is no wonder that I had all these problems...

But the resulting wire length of both the connected orange-red-brown and white-red-black wire is now 50 inch. Still wondering if that makes sense...

Quote
Folger Tech
For the needed sensor cable, you would use the 3pin Red,White, and black cable (...).
Following the wiring diagram, the white will be cut and taped as only 2 wires are needed.

[attachment 72143 Photo7493.jpg]

Edited 1 time(s). Last edit at 02/09/2016 02:46AM by Gingerbreadman.

Read the thread before making an advice. Sven.L has Rev.A with the servo/mech.switch, not inductive probe.

Quote
Gingerbreadman
I think I know what is going on with your wiring. That extra cable is intended to be cut. You use the small female end for connecting to your board, the other female end (the one in your pic is not used) First you strip back some of that thick black cable coming from your inductive bed sensor, then you solder the two resisters, I think 10k&15k ohm, between the black and the red (look at the diagram to be sure of that setup) , the white is not used and can be taped back the blue and brown run to your 12v power supply. Brown is + and blue is -. You use the small male end with its Frankenstein resistor combo to connect to the board with the white wire (check for sure it was late when I completed that step) to the outside facing away from the servos, it goes between the two end stop sensor connections . I was confused until I figured it out. My sensor lights up and senses the bed well with this setup. Hopefully my explanation will help you make sense of the folgertech wiring diagram, some of the trace lines are kinda muddy and difficult to read.

Quote
Sven.L
I communicated with Folger Tech, and they told me to use the triple wire (white, red, black) to connect to the servo triple wire (orange, red, brown), and to cut off then the white wire, so only the red and black wire are connected to the switch. While the PDF manual speaks about a 2-pin connector, in reality it is a 3-pin connector and a white-red-black wire (not the black-red wire as mentioned in their PDF). With such wrong information, it is no wonder that I had all these problems...

But the resulting wire length of both the connected orange-red-brown and white-red-black wire is now 50 inch. Still wondering if that makes sense...

Quote
Folger Tech
For the needed sensor cable, you would use the 3pin Red,White, and black cable (...).
Following the wiring diagram, the white will be cut and taped as only 2 wires are needed.

[attachment 72143 Photo7493.jpg]

Edited 1 time(s). Last edit at 02/09/2016 02:00AM by paul_delta.

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

I lost my resistors and decided to just try hooking all 3 wires up to the 5V next to the other endstop leads. Worked like a charm.

Quote
Gingerbreadman
I think I know what is going on with your wiring. That extra cable is intended to be cut. You use the small female end for connecting to your board, the other female end (the one in your pic is not used) First you strip back some of that thick black cable coming from your inductive bed sensor, then you solder the two resisters, I think 10k&15k ohm, between the black and the red (look at the diagram to be sure of that setup) , the white is not used and can be taped back the blue and brown run to your 12v power supply. Brown is + and blue is -. You use the small male end with its Frankenstein resistor combo to connect to the board with the white wire (check for sure it was late when I completed that step) to the outside facing away from the servos, it goes between the two end stop sensor connections . I was confused until I figured it out. My sensor lights up and senses the bed well with this setup. Hopefully my explanation will help you make sense of the folgertech wiring diagram, some of the trace lines are kinda muddy and difficult to read.

Quote
Sven.L
I communicated with Folger Tech, and they told me to use the triple wire (white, red, black) to connect to the servo triple wire (orange, red, brown), and to cut off then the white wire, so only the red and black wire are connected to the switch. While the PDF manual speaks about a 2-pin connector, in reality it is a 3-pin connector and a white-red-black wire (not the black-red wire as mentioned in their PDF). With such wrong information, it is no wonder that I had all these problems...

But the resulting wire length of both the connected orange-red-brown and white-red-black wire is now 50 inch. Still wondering if that makes sense...

Quote
Folger Tech
For the needed sensor cable, you would use the 3pin Red,White, and black cable (...).
Following the wiring diagram, the white will be cut and taped as only 2 wires are needed.

[attachment 72143 Photo7493.jpg]

My apologies Sevn L . Despite (or because of) reading through the entire thread twice I missed yours being an A build. I hope I did not create too much confusion as that was certainly not my intent.

Can I ask the merits of your method for connecting the Rev B induction sensor, besides sounding a lot simpler. Are there any relative drawbacks/compromises? Just looking for a better understanding. This was my first build and Im still stunned my printer didn't burst into flames (given my nonexistent arduino/ramps board skills) when I plugged it in. Thanks

Quote
spiffcow
I lost my resistors and decided to just try hooking all 3 wires up to the 5V next to the other endstop leads. Worked like a charm.

The primary advantage for me was not having to try to dig the resistors out of the dumpster.. Not sure if there are any major drawbacks. Maybe the proximity detected is slightly reduced? The read values are very consistent to down to 0.01mm. They could have just included a voltage converter instead of the resistor wires. I see several on ebay for ~$1.50.

Quote
Gingerbreadman
Can I ask the merits of your method for connecting the Rev B induction sensor, besides sounding a lot simpler. Are there any relative drawbacks/compromises? Just looking for a better understanding. This was my first build and Im still stunned my printer didn't burst into flames (given my nonexistent arduino/ramps board skills) when I plugged it in. Thanks

Quote
spiffcow
I lost my resistors and decided to just try hooking all 3 wires up to the 5V next to the other endstop leads. Worked like a charm.

Just finished my build (hopefully) last night, and am curious what the current forum thoughts on pros/cons of Marlin vs Repetier for firmware on the FT2020 Kossel?

So how do I find out the Nozzle Height Error at a specific point? I'm doing the calculator now, and I'm not sure how to find out that measurement.

---

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

Does anyone else find it hard to feed the filament into the extruder? This is my first Bowden setup, so maybe it's something I'm doing. I only managed to get it into the PTFE tubing by removing the push to fit connector from the extruder, as well as the PTFE tubing from the push fit. Then I feed the filament through the extruder gears and through the hole, manually thread it through the push fit connector, then add the tubing. Once I did that and pushed it all the way to the hotend, it didn't go any further either. Either 190C isn't hot enough, or it wasn't going in correctly. Any advice on this?

Also, my PSU's fan keeps revving up really high every like 30 seconds when I have the hotend and/or the bed heating. I'm assuming it's because it's getting warm with the power draw, but I just wanted to see if anyone else has this happening to them.

---

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

Quote
dc42

Quote
spiffcow
Is your calculator code available? I'm wondering what would happen if I just plugged 100+ points into it

It's written in Javascript, so you can view the source in your browser. The C++ source it is derived from is in my RepRapFirmware github repo.

So I added your calibration algorithm to my Repetier Z-correction normalization app.. It takes the Z-correction map (G33 L0 in Repetier to list the points) and calculates the error based on that Looks like it makes a huge difference! I attached an image of what I got on my printer that had previously been calibrated over 5 iterations with the 10 point calculator.. The new calculation involves 100 datapoints. All I have to do is run G33, kill some time, and paste in the results

Anyway, I'm guessing that my concerns before over getting caught in local minima were most likely a shortage of data points. Can you tell me your license for the javascript code? I might want to share my app when its completed..

The Javascript is derived from the C++ code that I added to RepRapFirmware, which is GPL.

---

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].

Hi, thanks for all the responses I received here, I have read them carefully and will continue to do so My full appreciation for the good will here. But since there is so much confusion and I end up having more and more questions and less self-confidence, I decided to make some more photos and screenshots, I hope they help to clarify the situation a little bit. I made photos of the servo part of the Kossel I received. I am going to send the same photos to Folger Tech by e-mail.

I followed the instructions, and connected the white-red-black triple wire with its triple connector to the triple connector of the orange-red-brown wire coming from the servo, and also cut off the final last inch of the white wire.

But receiving so much controversy response here in the forum and more questions in the e-mail communications with Folger Tech, I prefer to send you these photos and hope that the instructions I received were correct. Here in the forum, some voices are saying, that having connected these both triple wires is totally wrong (in opposition to the Kossel manual) and that instead I should have made some connection to the RAMPS board.

I am a novice with electrics and electronics, and really need help and self-confidence in what I am made doing. Alright. For someone in a Third World country, this is really a huge investment. :-)



Edited 2 time(s). Last edit at 02/10/2016 04:27AM by Sven.L.

Quote
dc42
The Javascript is derived from the C++ code that I added to RepRapFirmware, which is GPL.

Okay cool. I had guessed as much, but I didn't see a license statement and wanted to be sure.

Thanks again for making the calculator available. I'd probably still be trying to get a test cube to print without it.

Quote
Sven.L
I followed the instructions, and connected the white-red-black triple wire with its triple connector to the triple connector of the orange-red-brown wire coming from the servo, and also cut off the final last inch of the white wire.

How many 3-pin black-red-white extension cables do you have in your kit?

---

My [www.thingiverse.com] stuff, my configuration of Repetier 0.92.6: [forums.reprap.org] (use Arduino 1.6.5 to compile)

The difficulty or ease with which you can feed the filament in will vary. Since the filament comes on coils and develops a circular "set", I normally have to bend the end a bit more straight before feeding it in. Many times I have to run it back and forth to get it past some edge of an internal part. That aluminum block by the extruder motor is the first problem. Maybe I'll design a funnel opening for it. Once the filament is in near the hot end, then it can get caught internally there as well. Again, feeding back and forth sometimes works, if not then you have to pop the Teflon tube off to help line it up.

As for your PSU fan, you're correct, that's normal as it heats up with the additional draw from the heat bed and extruder.

Quote
Captain_Tim
Does anyone else find it hard to feed the filament into the extruder? This is my first Bowden setup, so maybe it's something I'm doing. I only managed to get it into the PTFE tubing by removing the push to fit connector from the extruder, as well as the PTFE tubing from the push fit. Then I feed the filament through the extruder gears and through the hole, manually thread it through the push fit connector, then add the tubing. Once I did that and pushed it all the way to the hotend, it didn't go any further either. Either 190C isn't hot enough, or it wasn't going in correctly. Any advice on this?

Also, my PSU's fan keeps revving up really high every like 30 seconds when I have the hotend and/or the bed heating. I'm assuming it's because it's getting warm with the power draw, but I just wanted to see if anyone else has this happening to them.

I'll give that a try later tonight, thanks.

Could someone help me to understand how to use dc42's calculator? I know you input the dimensions, but what do you do with the generated points to get the calculations? I've manually calibrated my printer, and everything is really close. I could do some more adjustments on the offsets, but for right now, it's ready to go, in my opinion. I'd just like to run through this calculator to get it even closer.

Quote
Legrand
The difficulty or ease with which you can feed the filament in will vary. Since the filament comes on coils and develops a circular "set", I normally have to bend the end a bit more straight before feeding it in. Many times I have to run it back and forth to get it past some edge of an internal part. That aluminum block by the extruder motor is the first problem. Maybe I'll design a funnel opening for it. Once the filament is in near the hot end, then it can get caught internally there as well. Again, feeding back and forth sometimes works, if not then you have to pop the Teflon tube off to help line it up.

As for your PSU fan, you're correct, that's normal as it heats up with the additional draw from the heat bed and extruder.

---

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

For each suggested point perform a move to that point and then probe.. For example, 'G1 X86.6 Y50 Z40' moves to 86.6/50 on the X/Y plane, 40mm above the bed. Then perform a 'G30' to probe the point. Record the results in the calculator, and set the EEPROM settings indicated in the calculator. Be sure to read carefully regarding the sign of the Z error on the calculator. You may need to reverse the sign of the recorded value.

My suggestion would be to take your hot end out of the effector and install the probe in its place, and make a script for probing all the points. Keep running the script and inputting the values until you get a very small error, then put your hot end back in the effector.

Quote
Captain_Tim
I'll give that a try later tonight, thanks.

Could someone help me to understand how to use dc42's calculator? I know you input the dimensions, but what do you do with the generated points to get the calculations? I've manually calibrated my printer, and everything is really close. I could do some more adjustments on the offsets, but for right now, it's ready to go, in my opinion. I'd just like to run through this calculator to get it even closer.

Quote
Legrand
The difficulty or ease with which you can feed the filament in will vary. Since the filament comes on coils and develops a circular "set", I normally have to bend the end a bit more straight before feeding it in. Many times I have to run it back and forth to get it past some edge of an internal part. That aluminum block by the extruder motor is the first problem. Maybe I'll design a funnel opening for it. Once the filament is in near the hot end, then it can get caught internally there as well. Again, feeding back and forth sometimes works, if not then you have to pop the Teflon tube off to help line it up.

As for your PSU fan, you're correct, that's normal as it heats up with the additional draw from the heat bed and extruder.

Quote
spiffcow
For each suggested point perform a move to that point and then probe.. For example, 'G1 X86.6 Y50 Z40' moves to 86.6/50 on the X/Y plane, 40mm above the bed. Then perform a 'G30' to probe the point. Record the results in the calculator, and set the EEPROM settings indicated in the calculator. Be sure to read carefully regarding the sign of the Z error on the calculator. You may need to reverse the sign of the recorded value.

My suggestion would be to take your hot end out of the effector and install the probe in its place, and make a script for probing all the points. Keep running the script and inputting the values until you get a very small error, then put your hot end back in the effector.

Could you give me an example of the output? From memory, as I'm not home to test it at the moment, the result shows a few values. A value for X and Y, and another variable that is for Z? So would the Nozzle Height Error be Z result - Z Height before G30?

Example
Point - G1 X86.6 Y50 Z40
Result - X86.6 Y50 Z41

Nozzle Height = 1mm

---

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