Help setting up FSR on Maxv2 /Rambo

[Reikal]

Hi Folks,

I've been trying for a week to get the Ultibots FSR kit working on my max.
To date i've almost bricked the rambo, nearly broken the build plate and had a darn good stab at breaking the arms of the Z axis.

I have high hopes that somone here will be able to assist in getting the FSR to detect before i succeed in actually damaging the printer.

Below is the endstop config i've been trying, but cannot get the FSR to do anything.

Code: Select all

// ##########################################################################################
// ##                            Endstop configuration                                     ##
// ##########################################################################################

/* By default all endstops are pulled up to HIGH. You need a pullup if you
use a mechanical endstop connected with GND. Set value to false for no pullup
on this endstop.
*/
#define ENDSTOP_PULLUP_X_MIN true
#define ENDSTOP_PULLUP_Y_MIN true
#define ENDSTOP_PULLUP_Z_MIN true
#define ENDSTOP_PULLUP_X_MAX true
#define ENDSTOP_PULLUP_Y_MAX true
#define ENDSTOP_PULLUP_Z_MAX true

//set to true to invert the logic of the endstops
#define ENDSTOP_X_MIN_INVERTING true
#define ENDSTOP_Y_MIN_INVERTING true
#define ENDSTOP_Z_MIN_INVERTING true
#define ENDSTOP_X_MAX_INVERTING false
#define ENDSTOP_Y_MAX_INVERTING false
#define ENDSTOP_Z_MAX_INVERTING false

// Set the values true where you have a hardware endstop. The Pin number is taken from pins.h.

#define MIN_HARDWARE_ENDSTOP_X false
#define MIN_HARDWARE_ENDSTOP_Y false
#define MIN_HARDWARE_ENDSTOP_Z false
#define MAX_HARDWARE_ENDSTOP_X true
#define MAX_HARDWARE_ENDSTOP_Y true
#define MAX_HARDWARE_ENDSTOP_Z true

I've attached pictures of the wiring too.
tapping the build plate will light up 1-3 of the LEDs on the FSR board, and at the same time will mark the Endstop Min to On when looking at the z calibration screen on the rambo.

Despite this nothing takes any notice of the reading. In its current state every run of OpenDact stop 10mm above the build plate, runs a config in the air then subtracts 10mm from the Z Max and says it's done.

Any hints on how to get this FSR board to work?


(Yes im planning to upgrade to Duet, but that's not an option right now due to other financial commitments)

[DeltaCon]

Can you tell me where you found the documentation of how to connect these FSR's?
If it works like this, I am very interested. In what I thought I learned from this forum is that FSR endstops are not very known to work great on the Rambo, and you probably need a Repetier firmware that is higher in version than the standardly deliverd one by SeeMeCNC. But that would probably a minor problem.

Not knowing if it's going to be of any helpt to you, bu what I notice in your post is:

// Set the values true where you have a hardware endstop. The Pin number is taken from pins.h.
#define MIN_HARDWARE_ENDSTOP_X false
#define MIN_HARDWARE_ENDSTOP_Y false
#define MIN_HARDWARE_ENDSTOP_Z false

While you actually DO have hardware endstops. Besides that probably some pin assignments wil have to be corrected because of the different delta approach.

[enggmaug]

Hello,


I went into the process of installing FSRs last week end. So it's still fresh.

Your wiring looks good.

However, if I uderstand, you are trying to add a Zmin endstop, while you have a Z probe.
On deltas, Zmax and Zmin endstops are linked to the Z tower, not the Z dimension.

If you use the repetier firmware 0.92.8 configuration tool, you just need to click the Z probe option, and indicate your Z probe is plugged in Zmin slot, with Normally open state.

I hope it can help you out.

[DeltaCon]

@Reikal: sorry for hijacking your question but...
@ enggmaug: So you have working FSR's on the standard Rambo board with just an upgrade repetier firmware? Can you confirm that OpenDact works on this combo? If so, I will order these FSR's at once!

[magicmushroom666]

Just to reassure you, I'm running an early rambo, V1.1 I think from the indiegogo campaign with the older seemecnc repetier firmware (091) and the little adapter FSR board everyone uses. This works great with openDACT.

[Reikal]

Just to reassure you, I'm running an early rambo, V1.1 I think from the indiegogo campaign with the older seemecnc repetier firmware (091) and the little adapter FSR board everyone uses. This works great with openDACT.

Thank you for the reassurance, but would you mind sharing how you got it to work?

----------

I've now also tried setting it as the z-probe and mapping the z-probe pin to 33 (default for z-min) where it is connected on the board.
Result was hitting the bed with considerable force and nearly flipping it of the printer.

This really is not much fun to set up.

[mhackney]

I'd like to make a bunch of comments here...

Firstly, slow everything down while you are testing to get this working. If you are slamming into the bed that fast, you ARE going to break something. Slow down, relax, it is not a race. Also, KEEP YOUR FINGER ON THE POWER SWITCH when you are first testing. That way you can turn things off it it goes horribly wrong.

There is a story I like to tell... A man goes into the doctor's office and says "Doctor, my arm hurts when I do this (he demonstrates flexing his arm wildly)." The doctor replies, "Well, then don't do that!" The point being, don't keep slamming your hot end into the bed over and over, do something different and less hurtful. (by the way, make sure to check the nozzle orifice as impacts that violent could distort the soft brass)

Secondly, remove the glass plate so you don't break it! Ideally, if you have an aluminum heat dissipator, use it. If not, find anything - wood, a sheet of cutting board, whatever, to place over the Onyx so you don't break it.

Thirdly, I was never able to get OpenDACT to work with my FSR setups with an Azteeg X3 Pro running repetier. Hopefully magicmushroom can provide the info you need. I've since moved on and no longer run repetier on any of my deltas. I do however, have FSRs with Repetier on my Taz 4 printer. But auto calibration for Cartesian printers is pretty good on repetier whereas not good for deltas.

Fourthly, most delta printers (the RMax for sure) home to the TOP of the printer. So, you need to have the MAX_HARDWARE_ENDSTOP_"N" set to true - where "N" is X, Y and Z. It is confusing because Repetier uses X, Y and Z for tower names on a delta even though they are not Cartesian. It would have been better to call them alpha, beta and gamma but we have what we have.

Fifthly, it is common practice to hook up a Z probe as the MIN_HARDWARE_ENDSTOP_Z with it set to true (the others set to false).

Sixtly, it's a nit but any perceived issue with FSRs on RAMBo are not true. RAMBo has no problem (see above), it is the FIRMWARE that is the problem. At this time, the only "sort of working" option that I know about for 8 bit Arduino-based delta firmware that supports auto calibration is the RC branch of the Marlin firmware. Of course, I have not kept current with Repetier firmware for the last year but if it had implemented reliable auto calibration, someone on this forum would know about it and be a hero for telling others about it and there would be no need for OpenDACT.

I hate to get on a soapbox (but I will)... It is 2016, the Arduino based firmwares have not been able to keep up with delta printer auto calibration requirements. There are many reasons for this. There are now controllers and firmware that are purpose-built to get great results with deltas. Once you've tried one of them you will wonder why you wasted your time banging your head against the wall. While the concept is commendable, OpenDACT is a bandaid and not a "no brainer" to get to work for everyone (whereas the other controller/firmware options are no brainers to get to work). Many of us calibrated our Rostocks manually for years before auto calibration was a reality. It really isn't that difficult or time consuming to learn to do. You might find you can calibrate manually a lot faster (and keep your sanity) than struggling with primitive firmware.

[off soapbox]

[Reikal]

Thanks mhackney

Soapbox is entirely justified, i have run into the deficiencies of using Atmel processors on previous projects far less complicated than this.

I've dropped my testing back to just the G31 command for now, it seems the root issue is that the z-probe does not trigger.
Depending on the settings i try it's either stuck on low or high.
i'm not really clear on what the pullup setting does on the endstops, right now im just trying different combinations of settings but it's not exactly a viable way to get this working.

Does anyone on here have a complete set of firmware changes to get the FSR board to register either as a Zmax or a Z probe?
Im using version 0.91 of the firmware just because the seemecnc premade has worked before and the 0.92 configuration builder asks a lot of questions i simply cant answer.




On a side now, mhackney; your blog posts are great, please keep up the awesome work. I'll be following it closely myself next month when i change up the electronics. It's still a great read though and im sure will save many frustrating hours of learning in the dark for many.

Main reason for this is i want the ethernet connectivity, might finally get this off the dining room table then!

[enggmaug]

My setup with FSR on Rambo does work. However, OpenDact does not for now.
I believe I have the same issue other had : my computer is set up for french format of numbers, not US.
I'll try to change that this week-end and try OpenDact again.

I'll keep you updated on that.

However, G31 does work.
I hooked up the FSR the same way as shown in the first post. But I did not config a Z min endstop. I configured a Z probe in the online conf tool.

I can share my configuration if you want, in order to have something to start with.

[mhackney]

I should add, you absolutely DO want to use the latest Repetier firmware. The version that ships with the RMax support for probing does not work. Learn to use the online tool, it isn't bad and once you do it is a great way to keep current. enggmaug, maybe you could post your config tool .json file - that can be loaded into the online tool as a starting point and tweaked for anything unique with the target printer.

[enggmaug]

Ok, I am back home.

Here is my config file for repetier.

It is only a starting point for you to set up your firmware. I am not liable for any problem that could arise following use of that configuration.

Now that I am home I can confirm that everything works just well with :

Rostock V1 with Rambo board v1.2
Repetier firmware 0.92.8 with above configuration.
FSR module plugged in Z min Endstop slot.
Windows 7 or 10 - US numbers formatting.


Make sure the Zprobe height is set to 0 in your EEPROM (I had issues with OPENDACT running my hot end into the bed because of that).

[mhackney]

Excellent, that should hep folks a lot enggmaug, thanks.

[magicmushroom666]

Ok my Z probe settings from 0.91 configuration.h are as follows:

Code: Select all

/* Z-Probing */

#define FEATURE_Z_PROBE true
#define Z_PROBE_PIN 11  //was 63
#define Z_PROBE_PULLUP true
#define Z_PROBE_ON_HIGH true
#define Z_PROBE_X_OFFSET 0
#define Z_PROBE_Y_OFFSET 0
// Waits for a signal to start. Valid signals are probe hit and ok button.
// This is needful if you have the probe trigger by hand.
#define Z_PROBE_WAIT_BEFORE_TEST false
/** Speed of z-axis in mm/s when probing */
#define Z_PROBE_SPEED 7
#define Z_PROBE_XY_SPEED 150
/** The height is the difference between activated probe position and nozzle height. */
#define Z_PROBE_HEIGHT 0 //was 39.91
/** These scripts are run before resp. after the z-probe is done. Add here code to activate/deactivate probe if needed. */
#define Z_PROBE_START_SCRIPT ""
#define Z_PROBE_FINISHED_SCRIPT ""

I used Arduino Pin 11 for my signal from the FSR board, which I believe is the Ymin endstop connector (see http://reprap.org/wiki/Rambo_development). All my max endstops are set to false, just using that pin for the IO. Got power and ground from the same endstop connector to keep things straightforward.

Code: Select all

/* By default all endstops are pulled up to HIGH. You need a pullup if you
use a mechanical endstop connected with GND. Set value to false for no pullup
on this endstop.
*/
#define ENDSTOP_PULLUP_X_MIN true
#define ENDSTOP_PULLUP_Y_MIN true
#define ENDSTOP_PULLUP_Z_MIN true 
#define ENDSTOP_PULLUP_X_MAX true
#define ENDSTOP_PULLUP_Y_MAX true
#define ENDSTOP_PULLUP_Z_MAX true

//set to true to invert the logic of the endstops
#define ENDSTOP_X_MIN_INVERTING true
#define ENDSTOP_Y_MIN_INVERTING true
#define ENDSTOP_Z_MIN_INVERTING true
#define ENDSTOP_X_MAX_INVERTING false
#define ENDSTOP_Y_MAX_INVERTING false
#define ENDSTOP_Z_MAX_INVERTING false

// Set the values true where you have a hardware endstop. The Pin number is taken from pins.h.

#define MIN_HARDWARE_ENDSTOP_X false
#define MIN_HARDWARE_ENDSTOP_Y false
#define MIN_HARDWARE_ENDSTOP_Z false
#define MAX_HARDWARE_ENDSTOP_X true
#define MAX_HARDWARE_ENDSTOP_Y true
#define MAX_HARDWARE_ENDSTOP_Z true

On the FSR board I have IO 3 jumpered to ground to give NC action, and IO1/SN2 jumpered to ground to give maximum sensitivity.

[enggmaug]

I should add, you absolutely DO want to use the latest Repetier firmware.

[Jimustanguitar]

Is there a practical limit to the weight that the JohnSL auto-calibration program works at?

I'm curious because I'm going to try it with a 3/8" slab of aluminum as my heated bed / heat spreader plus the glass. I estimate that it weighs about 8 pounds, and most of the documentation talks about much less weight. https://github.com/JohnSL/FSR_Endstop

[mhackney]

It has nothing to do (at least, not much) with weight (actually, mass), the FSRs measure force. I put 3# on my Rostock bed and tested the FSRs and they function perfectly. As long as you don't have so much weight that it distorts the bed or mounts or actually compresses the FSRs, it will work fine.

[Jimustanguitar]

And what's your thought on the little silicone pads? I noticed that you had some rubber washers in one of your pictures...

[mhackney]

You need some sort of mushy material (but hard mushy) on the FSRs to spread the force. I prefer the silicone pads. The washers were for a set of FSRs I bought before the UltiBots kit became available. They work fine too.

[enggmaug]

It has nothing to do (at least, not much) with weight (actually, mass), the FSRs measure force.

The applied weight on the bed changes the force read at the FSR level.
However, what the interface board does, is to trigger a signal when there is a positive difference on the force applied onto one of the FSRs.
That means that if you apply a constant force on any FSR, the signal is not triggered.

If you sit a cow on your Onyx, the board will trigger a signal for a couple of seconds only.
And when your hotend hits the cow, the signal will be triggered again as the cow jumps in pain while you brand it.

[DeltaCon]

Guess the cow should sit very quietly...

[Jimustanguitar]

I was just concerned because of the curve... The wiki says that pressure changes the value from 5meg to under 50 ohms when the hot-end touches it (http://reprap.org/wiki/FSR) and then the board is triggered when "the voltage drops below 0.92 of the ambient value" (from JohnSL GitHub https://github.com/JohnSL/FSR_Endstop)... 8 pounds of aluminum heated bed is a lot more pressure than the touch of the hotend, and depending on whether the resistance curve is exponential or logarithmic, I was scared that there wouldn't be any range left on the FSR.

[geneb]

Wouldn't the point load be 8lbs / # of FSRs?

g.

[Jimustanguitar]

Yes it would, so really it's between 2 and 3 # each...

I'll know how well it works soon. Might as well experiment with adding and subtracting weight to test the sensitivity. I'll report back when I know more

[mhackney]

Yes it would Gene and (*) again, it is force, not pressure, that triggers the FSRs. The wiki says "when the hotend pushes down on the print surface"- no mention of pressure. It goes on to say "depending on the force". I'm sorry to be on a soapbox here but I have received in the neighborhood of 100 PMs and emails from folks asking questions about setting up FSRs. "None" of them (2 did) understand the distinction between pressure, mass and force. They think these are little scales that measure mass or little pressure sensors that measure pressure. They are neither, they measure force!

Jim, I have now tested with a gallon of 2% milk on my bed - 8lbs - and it triggers just fine. 3/8" 6061 weighs 5.292 lbs/square foot. A 12" disk would weigh 4.2 pounds. The cow example is perfect. As long as the cow is sitting there and not moving, it exerts pressure but there is no force component. If you lower the cow onto the bed slowly, there is force until the cow stops moving (cow at rest). Similarly, if you probe the cow you are exerting force through the probe. If the hot end is really hot, the cow gets a micro brand. If it is cold, the cow gets annoyed.

My point above was, if you put enough mass on the bed you can eventually damage the FSRs by compressing them and of course you change the sensitivity if the amount of triggering force stays constant. But all of this is well within the range of what you are trying to do with the aluminum bed on FSRs.

Don't get me started on the difference between mass and weight!




(*) over educated PhD scientist afoot

[geneb]

So basically, we're talking cow hooves per square ounce, right?


*runs FAST*



g.