Runaway Y carriage - Sintron Kossel using Duet

I have moved this topic from: [forums.reprap.org]

After the request of several people.

I have had this problem for months. When I turn on the machine for the first time in the day, the Y carriage rams itself into the endstop and doesnt stop. It wants to go to the sky. After waiting 10 or more minutes, the problem goes away. I assume because the system has warmed up. But before I have the chance to hit the Emergency stop button, the Y tower is shaking like an earthquake.

My system is an upgraded Mini Kossel from Sintron. I have upgraded the board to a Duet 0.6 and Duex4 expansion board. I also have it hooked to a paneldue 9(version firmware 1.13). I have checked resistor R57 with a 10x loop and it looks intact. No solder cracks and no cracks in the resistor. I have examined the AMTEL chip and the micro solder joints all look good. I have tried tightening the belts and loosening the belts. No effect. I moved the Y endstop to the X carriage and the problem moves to the X carriage. I have replaced the Y wire and Y endstop with new ones. The problem does not go away. I have tried different permutations of the wire endstops. Does not help. I am lost on what to do besides wait 10min before using the system. My only issue with that is the problem does recur after a print. The print ends. The system shuts off all the motors and hotends and hotbed. Then when I come to the machine and try to auto calibrate and home it, it rams itself into the Y carriage again. I assume this will mess up my homing. I also assume that this is the reason that I had to replace my SIntron Printed Red carriages with ABS molded carriages since the printed carriages cracked in half. Another fix I tried was to reverse the polarity of the Y endstop so that low level is high and high level is low. Then put this new configuration into the config.g file. This seems to help the problem but it bugs me because this should not be an issue in the first place. The other fix that I tried was a homedelta.g code change. I made it so that during homing the carriages move up fast 1/2 way. Then the last half I made it slowly creep up to the carriages. This helps a little as well but only after the system has warmed up. Any ideas?

Software Information
Firmware Name: RepRapFirmware
Firmware Version: 1.09r-dc42 (2016-01-16)
Web Interface Version: HTML: 1.08, JS: 1.08
Web Interface by Christian Hammacher
Licensed under the terms of the GPL v2
File Upload for /sys and /www Directories


The web interface results are:
General
List Items
Configuration File
Machine Properties
Tools
Drive Configuration
Drive Endstop hit Minimum Maximum InstantDv Max Speed Acceleration Motor Current
0 No 0 mm 230 mm 20 mm/s 333.33 mm/s 1000 mm/s² 1000 mA
1 No 0 mm 210 mm 20 mm/s 333.33 mm/s 1000 mm/s² 1000 mA
2 No -2 mm 200 mm 20 mm/s 333.33 mm/s 1000 mm/s² 1000 mA
3 No n/a n/a 20 mm/s 60 mm/s 1000 mm/s² 500 mA
4 No n/a n/a 20 mm/s 60 mm/s 1000 mm/s² 500 mA
5 No n/a n/a 20 mm/s 60 mm/s 1000 mm/s² 500 mA
6 No n/a n/a 20 mm/s 60 mm/s 1000 mm/s² 500 mA
7 No n/a n/a 20 mm/s 60 mm/s 1000 mm/s² 500 mA

Edited 1 time(s). Last edit at 01/19/2016 10:57AM by DRTak.

Please post the M574 and M558 lines from your config.g file.

With everything cold and the machine properties displayed in the web interface, check whether the XYZ endstop states change to the correct value when you press the corresponding endstop switch. I presume the X and Z will be OK, but not the Y.

You are using 3 wire endstops, but you have connected only 2 wires. That could be the problem. Please post close up photos of both sides of the endstop board so that I can work out the right connections. Also tell me how you have them wired at present.

If you have a multimeter, disconnect the endstops at the Duet and measure the voltage between the 3V3 and STP endstop pins, for each of XYZ. Also tell me what state the endstops show in the web interface with the endstop cables disconnected.

---

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].

Please post the M574 and M558 lines from your config.g file.
LINE 24: M574 X2 Y2 Z2 S1.
LINE 52: M558 P1 X0 Y0 Z0


With everything cold and the machine properties displayed in the web interface, check whether the XYZ endstop states change to the correct value when you press the corresponding endstop switch. I presume the X and Z will be OK, but not the Y.
With the machine cold only X and Z work and say YES. Y says NO when I push the endstop. However, it will randomly say YES after I push it after 5min waiting. This will read YES but it is a delayed YES. The web interface slowly says YES. The X and Z endstops say YES rapidly. After 10min of waiting, if I push the Y endstop it will say YES faster but still slower than the X and Z. I left your original code line for the homedelta.g inside of the file. This way it will rapdily go to Home position. Ram itself into the endstop. Then cycle to G1 S1 X6 Y6 Z6 F250 slowly. This helps correct the issue and it will not ram itself into the Y endstop as much and eventually resolve. But only after waiting more than 10min.

You are using 3 wire endstops, but you have connected only 2 wires. That could be the problem. Please post close up photos of both sides of the endstop board so that I can work out the right connections. Also tell me how you have them wired at present.
I am only using 2 wires. The outer wires. Picture attached. The outer wires are connected to GND and STP on the Duet 0.6

If you have a multimeter, disconnect the endstops at the Duet and measure the voltage between the 3V3 and STP endstop pins, for each of XYZ. Also tell me what state the endstops show in the web interface with the endstop cables disconnected
Endstops disconnected. They all read YES on the web interface. Picture attached. Voltage across the STP to GND is 2.5v. STP to 3V3 is Zero. 3V3 to GND is 6volts. I have a really basic multimeter. I need to buy a better one to read DC current. But that is what I have so far.

Edited 1 time(s). Last edit at 01/20/2016 03:39PM by DRTak.

Thanks. However, the photos I wanted were of the whole board, so that I can follow the PCB traces and work out how the components on the endstop boards are connected.

Two more tests to try with the board cold:

1. When you short STP to 3V3 on the Y endstop,does the web interface show the endstop as triggered?

2. When you short STP to GND on the Y endstop (or when you connect the Y endstop switch and it is not triggered), does the Y endstop LED illuminate, with the same brightness as the other two?

---

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].

1. When you short STP to 3V3 on the Y endstop,does the web interface show the endstop as triggered?
If I disconnect the Y endstop wire from the endstop trigger and the trigger is NOT activated. The web interface reads YES. When I short out STP to 3V3 the web interface still says YES but NO RED LIGHT goes on for the Y endstop.
If the Y endstop wire is connected to the endstop trigger and the triger is NOT activated. The web interace reads NO and the RED LED Y endstop is on and red. When I short out STP to 3V3 the web interface still says NO but ALL of the RED LEDs go out for all of the motors.


2. When you short STP to GND on the Y endstop (or when you connect the Y endstop switch and it is not triggered), does the Y endstop LED illuminate, with the same brightness as the other two?
If the Y endstop wire is connected to the endstop trigger and the triger is NOT activated. The web interace reads NO and the RED LED Y endstop is on and red.

If the Y endstop wire is disconnected to the endstop trigger. The RED LED Y endstop light is OFF. Web interface reads YES. If I now short out STP and GND. The web interace reads NO and the RED LED Y endstop is on and red.The brightness matches the other LEDs.

I have attached a close up picture of the endstop. You should be able to mag it up with Photoshop etc. It is a Sintron V1.2.

Edited 1 time(s). Last edit at 01/20/2016 10:50PM by DRTak.

From your measurements, it sounds to me that the Duet Y endstop connector is working correctly. You have configured the endstops as active high, which means that the red LEDs on the Duet should be lit when the endstop switch is connected but not triggered.

So I think the problem is that you have not connected the endstop switches correctly. From your photos, and confirming with a schematic I found elsewhere, I believe the correct connection is as follows:

Endstop Pin 1 (as indicated by the figure '1' on the PCB in your first photo) - connect to STP on the Duet endstop connector
Pin 2 or pin 3 - connect one of these to GND on the Duet endstop connector
Pin 4 - connect to 3V3 on the Duet endstop connector

You will need to use S0 in your M574 command, because these endstops are designed to produce active low outputs.

Please try this connection scheme with either the X or the Z endstop connector. If it works, then try it with the Y endstop connector too.

Also make sure that there is nothing metallic underneath the endstop boards that could be shorting connections together.

Edited 3 time(s). Last edit at 01/23/2016 07:06AM 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
From your measurements, it sounds to me that the Duet Y endstop connector is working correctly. You have configured the endstops as active high, which means that the red LEDs on the Duet should be lit when the endstop switch is connected but not triggered.

So I think the problem is that you have not connected the endstop switches correctly. From your photos, and confirming with a schematic I found elsewhere, I believe the correct connection is as follows:

Endstop Pin 1 (as indicated by the figure '1' on the PCB in your first photo) - connect to STP on the Duet endstop connector
Pin 2 or pin 3 - connect one of these to GND on the Duet endstop connector
Pin 4 - connect to 3V3 on the Duet endstop connector

You will need to use S0 in your M574 command, because these endstops are designed to produce active low outputs.

Please try this connection scheme with either the X or the Z endstop connector. If it works, then try it with the Y endstop connector too.

Also make sure that there is nothing metallic underneath the endstop boards that could be shorting connections together.

THANK YOU THANK YOU!!!! I hope anyone who is having this issue also reads this and gets help.
Its fixed and then some. I have never been able to get my Sintron endstop LEDs to light up when activated. I know they were supposed to because other said it did. Now they do. I changed the M574 command to be S0. After that, I did exactly as you said. X and Z first. and it worked. It registered on the web interface as YES. Then I crossed my fingers and tried the Y endstop. It also works. It all works immediately now. No waiting 10min or more for a warmup. So now the Duet LED's are off. When endstop is trigged, the endstop themselves turn RED and the DUET board LEDS light up red. I am so happy. After all of my upgrades I was thinking I had a bad machine that I sunk a lot of cash into. But now its perfect (except for bridges/bridging. Haha).

Thanks again so much DC42. You are a genius.

Edited 1 time(s). Last edit at 01/23/2016 01:45PM by DRTak.

I was having issues with my new duet Ethernet. My old Duet 0.6 went bad. I couldnt figure out why my endstops were not working. The machine would go in reverse. Went I wanted it to go down, it would go up etc. It was related to the fact that I had to change the Shipped Duet Ethernet Config G file from reading LOW to "My" Normal of reading HIGH. For those of you who have issues.