Simulator

Husker

I am trying to build simulator for a pressure decoder at work. The decoder monitors a pressure signal. A device at distance from the sensor creates pressures spikes in a Morse code type of format. The decoder recognizes these spikes in a timed sequence and provides readable data.

The pressure sensor is a transducer that creates variances in resistance. Right not I am using the AD5220 to try to replicate the signal. So far I have been able to create the signal but with flaws. When I use software to perform a FFT it comes out completely erratic. I have tried several options. Adding capacitors to the circuit and adjustments in programming have helped smooth out the signal so at least it isn't a square wave.

Is there something I can add to the circuit or anything in particular I should look at in the programming? Any help is appreciated.

Note:
-Decoder reads current from 4-20mA
-Decoder reads changes in pressure, not the magnitude (it flat lines at the same point whenever the pressure signal settles)

*The top FFT below is the real signal and the bottom FFT is the simulated signal

D = T/21
Z = 0

DEBUG "Total Time=", DEC(T),CR
DEBUG "Data Slot=", DEC(D),CR
DEBUG "Pulse Time=", DEC(P),CR,CR

Main:

'Drop to lowest possible resistance
HIGH 0
HIGH 1
LOW 14
LOW 15

FOR x = 1 TO 100 'Approximate drop of 150*80=12000ohms (Max is 10K)
LOW 1
HIGH 1
NEXT

'Survey Routine
DO
PAUSE 10000 'Allows SIU to stabilize
GOSUB Sync
'FID Pulses
GOSUB DATA0
GOSUB DATA0
'MFCX Pulses
GOSUB Data0
GOSUB DATA0
'AZUX Pulses
GOSUB DATA0
GOSUB Data0
GOSUB DATA0
'INCX Pulses
GOSUB DATA0
GOSUB DATA0
GOSUB DATA0
'MDAX Pulses
GOSUB DATA0
GOSUB DATA0
'BSPX Pulse
GOSUB DATA0
'PBVX Pulses
GOSUB DATA0
LOOP

Sync:
FOR x = 1 TO 4
GOSUB Pulse
PAUSE (24*D-(4*P))/4
NEXT
RETURN

Data0:
PAUSE 0
GOSUB pulse
PAUSE T-P
RETURN

DATA1:
PAUSE D
GOSUB pulse
PAUSE T-D-P
RETURN

DATA2:
PAUSE D*2
GOSUB pulse
PAUSE T-D*2-P
RETURN

DATA3:
PAUSE D*3
GOSUB pulse
PAUSE T-D*3-P
RETURN

DATA4:
PAUSE D*4
GOSUB pulse
PAUSE T-D*4-P
RETURN

DATA5:
PAUSE D*5
GOSUB pulse
PAUSE T-D*5-P
RETURN

DATA6:
PAUSE D*6
GOSUB pulse
PAUSE T-D*6-P
RETURN

DATA7:
PAUSE D*7
GOSUB pulse
PAUSE T-D*7-P
RETURN

DATA8:
PAUSE D*8
GOSUB pulse
PAUSE T-D*8-P
RETURN

DATA9:
PAUSE D*9
GOSUB pulse
PAUSE T-D*9-P
RETURN

DATA10:
PAUSE D*10
GOSUB pulse
PAUSE T-D*10-P
RETURN

DATA11:
PAUSE D*11
GOSUB pulse
PAUSE T-D*11-P
RETURN

DATA12:
PAUSE D*12
GOSUB pulse
PAUSE T-D*12-P
RETURN

DATA13:
PAUSE D*13
GOSUB pulse
PAUSE T-D*13-P
RETURN

DATA14:
PAUSE D*14
GOSUB pulse
PAUSE T-D*14-P
RETURN

DATA15:
PAUSE D*15
GOSUB pulse
PAUSE T-D*15-P
RETURN

Pulse:
Z = Z + 1
DEBUG "Pulse", DEC(Z),CR
LOW 0
LOW 1
HIGH 15
PAUSE P/2
HIGH 1
HIGH 0
LOW 1
LOW 15
PAUSE P/2
HIGH 1
RETURN

stamptrol

Can you show us a schematic or gives us a bit more information on the transducer and decoder?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk

http://www.siskconsult.com
·

Husker

I have included the schematic and the pressure transducer data sheet. Remember the transducer is being replaced by my board.

stamptrol

OK,

Your transducer produces an anaolg signal proportional to pressure applied. That 4 - 20 mA is measured in some instrument, controller or logger.

The AD5220 is nothing more than a potentiometer whose wiper position is controlled by sending the chip pulses and an up/down direction command.

I can't tell from your chip schematic exactly what you're doing as it looks like you've got the pot in series with the 100K. What is the "I" representing? For sure you are not simulating a 4 - 20 mA signal.

Assuming the instrument you're trying to drive is really looking for a voltage of 1 to 5 volts, the pot chip may work if you reconnect it as a potentiometer ( a to +, b to 0v, w as wiper). Then your simulated waveform will be limited by the fact that you only have 128 possible steps and the transitions will affect your FFT.

You might be better off to use a 12-bit d/a which would give you 4096 steps. Then, if you're instrument really wants 4 - 20 mA, just feed your 1 - 5 v d/a signal into a commercial signal converter to give you a real 4 - 20 mA signal.

Cheers

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk

http://www.siskconsult.com
·