4-20ma ramp signal

Thank You all for replying. I forgot about using a 555 timer as a ramp generator. I just need to convert the voltage on pin 6 to a mA signal.

How accurate and repeatable does the signal and the timebase have to be?

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

And the linearity..

Several ways,

1) Use a small constant current source and charge a large capacitor then switch in a constant current sink and discharge it. Use that signal to drive a voltage to current converter typically availabe in a single chip form from TI (BB).

2) Easiest if you can program some firmware, get a microcontroller with a dac output (or PWM) and use that to drive a voltage to current converter.

#2 is the cleanest solution as #1 has analog stuff and more parts. If your not comfortable with analog Id suggest looking a microchip, one of their pics could surely do the job.

The question arises of whether any isolation may be needed, dependant on the nature of the target the current is to be sourced 'into'. A simple ramp generator, feeding one input of an op-amp, a resistor in the loop to sense the current, with the voltage fed into a differential amplifier, and just have the feedback arranged so that the output of the op-amp (with current buffering if it can't source 20mA) is adjusted to make the voltage across the sense resistor match the value from the ramp generator. For an isolated version, look at the application note 54, for the IL300 opto coupler, for a way of generating a voltage controlled current with isolation, and just feed this from the ramp.

Best Wishes

And what is the required voltage compliance?

Jon

Getting the 10-bit PWM to update without glitching around on the LS 2 bits.

It can be done with a single 8-cent LM324 and some discrete parts.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Hello Mook,

Ah, but where is the sportsmanship? Suggestion 1) should also be possible via voltage controlled current source/sink. Then there won't be a large cap, just a wee cap. If linearity and all that isn't critical it could be done w/o formal current steering.

Regards, Joerg

Hello Spehro,

Yes, that is also my favorite jelly bean amp. So far nothing has rivaled its price/performance ratio. Which is no small feat at two cents an amp. With some serious pondering the LM324 can be made to perform almost any control function because its input CM range includes ground. I have replaced whole uC solutions with a few of these, dropping the BOM total by almost a Dollar.

When we put our teeth in a jar people will probably still design it in. Unless the trend continues where universities let people off into the work force who can't even solder, let alone understand analog stuff.

Regards, Joerg

Didn't the OP say somewhere in the thread that ultralinearity isn't important? Howcome nobody's come up with the most obvious, the 555, yet? ;-P It's probably the cheapest approximation to a ramp you can get, and then just buffer & level-shift the ersatz "triangle" to do whatever you want. ;-)

Cheers! RIch

Two 555 timers, MOSFET solenoid drivers, two solenoid valves, a tank, and a pressure sensor with 4-20 ma output.

(ducking.....)

Hello Rich,

Doubt it's the cheapest. Spehro's version with an LM324 is likely going to be a couple cents less. I'd try totally discrete to shave off another penny or two ;-)

Regards, Joerg

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com