Hello Guest it is December 04, 2021, 01:12:32 PM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - briksay

Pages: 1
I am using the UC100 through a BOB.  The BOB is where I "Was" getting the 5v power for the sensors.  THey are recommended to be 8vdc minimum up to 30vdc.
SO, my divider got me to 10V for the sensors.  The signal from the sensors to the BOB inputs pulls the ~9v down to 0.7vdc when sensor is activated.  Doesn't the BOB isolate that from the UC100?  I know its works currently.  I found that the BOB was putting the 9.3V OUT of the inputs.  if you pull that down to 0, then the Mach3 registers an output.

the 5V supplies on the BOB were OUT.  So, I built a voltage divider for the 24V supply to the BOB.  Powered the sensors at 10VDC.  They worked perfectly!!!  Problem solved!!

Well, a couple more steps of diagnostics.
the sensor inputs of the BOB are free hanging around 8-9V when NOTHING is connected.   I tested each limit with a simple wire to GND.  XYZ worked fine reading all the way to Mach3.  So, system works.
The prox sensors were being powered by the 8-9V sensor voltage backfeeding through the inputs.  The sensors are supposed to be powered by "source" power.  They were getting their only voltage through the signal wire.  Their power was connected to the 5V on the BOB which was putting out nothing.  So, tonight, move power source to the 24V source.  See if they work now.

This "being powered through the signal" problem explains why if it was triggered by a proximity to metal, the other 2 sensors would drop their voltage too. 

Worst case: new BOB;
Best case: 5V is bad and have to power them from another source.

Either way, I'm closer now than ever before.

Tonight, will be testing with DRY mechanical switches.  No NPN NO circuits.

my BOB is fed with a 24VDC PS.  they are fed from same.
the sensors are labeled PN SN04-N, found their specs.  Supply is 5-36VDC, 0V NPN N.O. output.
when I put metal in front of them, their signal line goes from 9.3V down to 0.7v.  Which "SHOULD" be enough to detect to zero for the Mach.  BUT!!!
The other channels drop by from 9.3 down to 8.3 simultaneous.  Even though I am not activating that sensor.

So, I don't know the history of the sensors by previous owners that hacked.  I am a electrical engineer and been doing sensors for 25yrs.  So, i understand and know what they SHOULD be doing.
And I know that these are acting weird.  So fairly convinced that the issues are in the sensors.  I will create the 3 switches w/ microswitches to GND and diagnose from there out.

if they are disconnected, they are high.  No problem.
If I pull them low by shorting to ground, No problem.  All on Mach seems to read this fine.

I just cannot fathom the coincidences that would require 3 non-contact sensors to go funky within a few days by changing a PC and adding the UC100.  Never touched the sensors.  but now my signals are cross fed and only work on rare occasions. 

It is not fixed 100% yet, as I have a REGULAR job that keeps me away from the fun stuff other than in the evenings.
I am new to CNC and learning.  It was a CHinese CNC machine from an auction.  I got it working.  The older Dell PC was just flaking due to age.  So, new PC and figuring out the missing components by going to WIN 7/ 64 meant the UC100.  NONE of this explains how the sensors are now working funky.  I had it working 100% on the older PC. 

Partial Sort:  the XYZ limit switch inputs were going from 9.3v down to 0.7v when bringing metal in range on the non-contacts.  I am also getting crossfeed.  If one limit is activated, the others drop their signal voltage by approx 1v.  This is SO weird.  No hardware changes on the machine.  Just a new PC and the introduction of the UC100.  But now limit switches are flakey.  I am going to build a simple microswitch bank to test w/o the proximity switches.  If this solves the prob, then the prox switches were it, but why now?  why when a UC100 was introduced?  Twilight ZONE???

I have narrowed down that the inputs to the UC100 WORK on Estop.  Just not the limit switches.  Since the inputs are basically the same on the BOB, it may be the sensors.  My limits are non-contact inductive / capacitive.  I will test the actual voltage to the input during the Hi/Lo cycles of the switches going into the BOB.

Had Mach3 through LPT talking to BOB.  No issues other than old PC dying.  So, New PC Win 7 / 64Bit.  Purchased UC100 from CNCDrive.  Installed all drivers and adjusted plugins.  I can control the XYZ through JOG, but the input switches are now ignored completely. Checked the Config Pins/Ports: IN on 13,12,15 like before.  When I put metal onto limits, the switch light comes on but the UC100 diag and the MACH diag page don't indicate switches at all.  The Mach3 shows all switches closed, even when they are not.  Please help.

PS, to get it to Jog, either disable limits or reverse input polarities.  but the limit activations don't make ANY of the indicator lights switch.

Pages: 1