Host Engineering Forum
General Category => Do-more CPUs and Do-more Designer Software => Topic started by: jcb on July 01, 2022, 11:56:36 AM
-
I was playing around with a BX-HSIO4 to interrogate and measure the PWM return on a Temposonics G series LVDT position sensor.
On the HSIO Card the return value on a 5 inches of travel rod is a difference of 180 in the .ACC which only gives me a resolution of .025 inches per unit which is far to gross for saw positioning. It would allow me to use the rods in low resolution applications for hold down or table positioning and I am going to test closed loop control with a proportional valve with rod feedback later this morning in the shop.
I have always used motion controllers for this and interface them to the PLC with Modbus/TCP. I figured the speed on the HSIO4 inputs should have no problem handling the measurements and I am able to use a PWM output for interrogation and a Edge Trigger Input to measure the duration of the feedback signal.
The calculation is: Resolution (length / increment) = 1/TimerFreq (usec/increment) * ReturnSpeed (length/usec)
On my Motion Controller the timer register contains the feedback. It is in the units of cycles of the high-frequency clock I then do the math based on the IC's Clock rate of 117.96Mhz.
(1 uS /117.96 increment) * (1 in / 9.0uS(Gradient)) = 0.00094193888 inches per ct / Recirculations
I see in the filter settings the clock rate is 75mhz which means if I use raw counts I should be able to achieve ~1.5 thousands resolution.
(1 uS / 75) * (1 in / 9.0uS) = 0.00148148 inches per ct / Recirculations
Am I missing something?
Thanks
John.
-
The Unit of Time in the BX-HSIO4 Edge Triggered Accumulator is in Microseconds.
Is it possible to increase the resolution of this timer in the future via software or is this a hardware limit?
Thanks
John.
-
The Unit of Time in the BX-HSIO4 Edge Triggered Accumulator is in Microseconds.
Is it possible to increase the resolution of this timer in the future via software or is this a hardware limit?
Thanks
John.
Probably technically possible, but not trivial. Would require FPGA changes.
-
Probably a simple question, but I am trying to use the HSIO4 for counting encoder pulses. My encoder has the following pin designations:
Vcc
GND
A
B
I have tried several ways, but cannot figure out how to wire the HSIO4 module to read the encoder pulses from A and B. Probably super simple, but perhaps I have been looking at it too long. I am supply 5V from the Y0 output of the HSIO4.
-
I have done a bit more trouble shooting
Confirmed 5V from Y0 to output Common
Confirmed 3.5V (High) and 0V (Low) on A when measuring A to Common
Confirmed 3.5V (High) and 0V (Low) on B when measuring B to Common
However, when I wire A and B into the HSIO4 (and configured for Quad Input in DMD) I get no reading from the PLC. A is wired to X0 + and B is wired to X1 + on the HSIO4 card. What am I missing??
-
I assume you are using a ZIPLink terminal and cable.
ZIPLink module: ZL-RTB40 or SL-RTB40-1
ZIPLink cable: ZL-BX-CBL40-S or ZL-BX-CBL40-1S
With the above equipment and guessing at your encoder (without seeing its specs):
Encoder A --> IN 0+ (on ZIPLink)
Encoder GND --> IN 0- (on ZIPLink)
Encoder B --> IN 1+ (on ZIPLink)
Encoder GND --> IN 1- (on ZIPLink)
And, of course, you'll have to hook your +5 VDC power supply to your encoder's Vcc, and the minus (-) terminal of your power supply to the GND of your encoder.
As I said, this is just my first guess based on the info you provided.
-
Greg, thanks for the suggestion but it did not work.
I am using the 5VDC supply as supplied from the card (HSIO4) and the output on Y0. Perhaps I need to use an external supply.
-
Greg, thanks for the suggestion but it did not work.
I am using the 5VDC supply as supplied from the card (HSIO4) and the output on Y0. Perhaps I need to use an external supply.
You certainly do. The max current Y0 can output is only 20mA. I seriously doubt your encoder will draw that little. A typical encoder, I would guess, would need probably 3x to 4x that (i.e. 60-80mA).
-
Yes sir! That did it...thanks Greg!