Slow current ramp generator

[martinochiro]

To test a digital thermometer, i need a circuit that generates a 4-20mA current ramp in 20min (1200 sec) in the up and down direction (also of 1200sec). The power supply available is 12V and the input resistance of the digital thermometer is 250 Ohm.

Thank You

 

[Prototyp_V1.0]

Because the time must be so accurate, using a opamp integrator would not obvoously not be recomended.

What abut a digital counter + DA converter + Opamp circuit for converting input voltage to current?

 

[crutschow]

I agree that an up/down digital counter driving a D/A seems the best approach.

What resolution/accuracy do you need?

For example, 16-bit resolution would give a step increase about once every second with a resolution of about 0.0015%.

 

[martinochiro]

A resolution of 12 bit (or more) is ok.

 

[crutschow]

Then use a 12-bit up/down counter and a 12-bit D/A.
12 bits will give you 4096 steps so, for a 1200s ramp, you need a 4096 / 1200 = 3.4Hz clock to the counter.

Depending upon the accuracy you need, the clock could be generated by a 555 timer connected as an astable oscillator.

A typical up/down counter is the 4-bit CD4029. You would need three of them connected in sequence.

A flip-flop (latch) could be used to change the up/down input of the counter when the counter reaches the max count and when it reaches the zero count.

Are you able to design such a circuit by yourself?

 

[martinochiro]

Thank You crutschow
a schematic would be much appreciated.
Regards

 

[crutschow]

Well I don't have a schematic handy.
Have you done any microprocessor programming?

 

[D.A.(Tony)Stewart]

1. verify the monotonicity, gain, offset error of your ADC using a better DAC of same resolution then calibrate and correct all errors.

2. Make an I to V convert with V to I convertor , calibrate and correct all gain, offset and nonlinear errors.

3. Combine 1 & 2 to simulate thermometer data usind DAC output and V to I converter at any speed.

Is this a synchronous data collection system? Sample rate?

 

[martinochiro]

Thank You for the suggestions.
This is a test circuit i will use just to simulate (slow) variations in temperature. Not a true data collection system.

 

[martinochiro]

No experience in micro programming.
Any suggestion about some micro family to begin in this area ?

 

[D.A.(Tony)Stewart]

Microchip AVR, Raspberry Pi, Arduino.....

 

[crutschow]

I have used a Stamp module for a small project, which is quite simple to use and is programmable in Basic.
I recommend you use a micro that is designed to use Basic, such as the Stamp or PICAXE, since I think it's the easiest programming language to start with.
Starting with a language like C (or its relatives like Java) is only for the masochistic out there as I see it. I think the arcane syntax of C is something only a programmer could love. ;-)
Obviously I don't fall into that category.

 

[D.A.(Tony)Stewart]

You dont need a MCU. Just a counter like cd4040 with clock and DAC with shift register in between if Serial DAC and suitable 8x or 12x clock and logic.

 

[crutschow]

You dont need a MCU. Just a counter like cd4040 with clock and DAC with shift register in between if Serial DAC and suitable 8x or 12x clock and logic.

No, of course you don't. But the design for this using a small micro might be easier then designing the clock generation, counter connections, and logic required to do this task.