forum.cncdrive.com

Hi,

Before it work stable and well, but recently i start have a problem, UCCNC trigger hard limit and reset, but only when i start spindle, and during on state it will trigger with very short repetions. My VFD is standartalone, not controled by AXBB-E. I checked all earth connections and they are OK at very PSU, i even disconnect the EMI filter and direct plug to mains but problem still happens the same.

Heres a video of the problem: https://youtu.be/yaaG_FVZW5w
In video some of hard limit triggers are not visible due lack of FPS, some are very very very short but they trigger the reset. Also it only happen when spindle is ON and everytime i turn on the spindle. While OFF everything stable.

Found the problem, it appear to be a limit switch (SN04-N), even not triggered nor switch light on it does the problem, strange thing, using isolated inputs, 3 of them in parallel sharing same line but only one causing problem

Edit: The problem presists even when i remove the last sensor and keep only the 2 that comes with the machine, the induction sensors are what causing the problem, but it didn't happen before, it start today without any other modification...
One of the sensors cables come along with the spindle cable on drag chains with about 3m length, the spindle cable have mesh and a ferrite core on cable, i also try put cores on sensor cable but same result.

Video here: https://youtu.be/iUDxRh9z7KE

how have you wired these sensors up?

You may need to install a higher current sink resistor in parallel with the AXBB-E input as inductive (and capacitive) proximity sensors have a leakage current which could be triggering your input signal

http://pliki.aksotronik.pl/sn04.pdf

http://www.riko.com.tw/download/pdf/sen ... sensor.pdf

lists current the leakage current as 2mA....
think of it like this.... 24V / 2mA = equivelent to 12kOhm

The manual for the AXBB-E lists the input ports as having 2.2kOhm
https://www.cncdrive.com/downloads/AXBB_E_manual.pdf

so this means that when the prox sensors is off it will feed a leakage current into a 2.2kOhm resistor.... and it's own off impedance is approx 12kOhm...
therefore 2.2/(12+2.2) * 24, means that the input voltage at the AXBB-E input appears to be approx 3.7V when the prox sensor is OFF!

I do not know what the high / low state signal of the AXBB-E is but they normally have a window of say below 6 or 8 Volts being off, and above 16 to 18Volts being ON.... the problem can be in the range between where if the last state was ON, the input will not trigger OFF status.

This problem gets worse when you have 2 sensors in parallel..... 1/ (1/R1 + 1/R2) is the equation for parallel resistance.... for similar resistances (12kOhm) this is the same as 1/2 the resistance as it acts in parrallel....

Now you have 2.2 / (2.2 + 6) * 24 = 6.4 V.....

Hence you install a second (3rd or 4th dependanding upon what wattage of resistor you have + its resistance) in parrallel with the AXBB-E input so that it acts as a current sink.

You also need to be careful that your sink does not exceed the current rating of ONE sensor being triggered (15mA)

Robertspark wrote:how have you wired these sensors up?

3 sensors in parallel NPN NO
Signal is GND when closed and 24V when open

Black: I5- (3 sensors there)
Blue: 24V-
Brown: 24V+ & L5+

As i need to wire 2 more sensors and have a total of 5, should i opt for NPN NC sensors and go serial connection? Would that avoid this problem and the lack of a sink resistor? What is the best option?

Have you used diodes on the outputs from the sensors?

https://www.google.com/url?sa=t&source= ... EHWZijLATK

I do not know what the high / low state signal of the AXBB-E is but they normally have a window of say below 6 or 8 Volts being off, and above 16 to 18Volts being ON....

"There is a serial 2.2kOhm resistor on each inputs which limits the current into the
optocoupler making the optocoupler to function between 10 and 30Volts input voltage"

Robertspark wrote:Have you used diodes on the outputs from the sensors?

https://www.google.com/url?sa=t&source= ... EHWZijLATK

No. Should i? I tried one diode at output between I5- and signal but not 1 per individual.

I think i found the major problem...
My machine is all made from steel, i tried to earth it now, problem is solved, no more soft light on signal nor trigger. If i use a thin wire at first contact with chassi when spindle is ON it sparkle on wire. I meassured AC from Neutral to Chassi, when spindle is OFF = 0VAC, when spindle is ON = 100VAC (un-earthed), them earth it and spindle ON gives 0VAC and problem gone!
As my sensores are screwed on machine the source of intreference comes from that EMI

Strange fact is that start to happen yesterday, to have such voltage on chassi can it be EMI? The only wire with AC on bundle is from spindle and that one is earthed and meshed, so i suspect connections on spindle jack are lose or touching, i saw the guy who made the eletrical joints and i can say that are pretty bad joints, almost cold solder join with a big ball, but as that work i never fix it myself, will have a look at that now.

PS: A 680ohm sink resistor worked too to supress the problem.

If it's a Chinese spindle, most don't have the ground pin connected to anything.

ger21 wrote:If it's a Chinese spindle, most don't have the ground pin connected to anything.

Yes it's generic 2.2KW spindle, i always wonder why they have a pin with earth but spindle body was isolated. Don't know what they do with the pin4(earth) from inside.

You need to take the top off, and connect a wire from the pin on the plug to the spindle cap or body. This is a common way to do it.