Shifting, Magnifying Voltages

Posted by: @davee

↑

integrator

The fact that by convention, the powering + and - are not shown in schematics clears up the confusion of the parts.  In the above image... you're telling me that the Vo is area under the curve of the Vi as a f(time)???  I'll have to digest on that.  My first nature would always be to handle that in software.  You know the saying, "If you have a hammer, everything looks like a nail."  That's pretty cool! 😎

Posted by: @davee

↑

A complete PID computing circuit can be built by combining the above elements, for something like an optimised heater control system, without a digital processor or line of software!

I could see that being great for a low-end system to make for pennies sell at profit.  But, still I'd prefer to spend the $1.20 for a ESP that I can make WiFi configurable, have it OTA update the software with newer features. 

But I should ponder when I'd do such "calculations" in hardware.  I guess I already am if using the differential feature of the ADS1115.

Posted by: @davee

↑

Given your need to use your Tonka toys in the better weather, etc.

😉 😊 

Inq

Hi @Inq,

   I agree with the general tone of your message with

Am I being too optimistic to think that by using the Vref in this differential way, that at least any thermal effects of the Hall sensor will be eliminated by the Vref also moving so that the difference is always accurate consistent?

near the start, so I don't have much to add.

As for this specific question, the innards of the Hall Sensor remain a bit of an epoxy covered mystery, but as a minimum, I suspect it includes the 'obvious' Hall sensor, plus a voltage reference (like a low current linear voltage regulator, but hopefully chosen for reasonable voltage accuracy and temperature stability) and an operational amplifier "block", to increase and buffer the Hall effect voltage. There will also be the two part ferrite core to couple the magnetic field from the external current carrying wire to the actual sensing device. I have suggested that there are at least three electronic components, but two or even three might be integrated into a single device, although I suspect (with no evidence) they are discrete.

The differential measurement method should eliminate most, if not all, of any voltage drifts by the voltage reference, but there are several other circuit 'blocks' which will also have some temperature dependency. The data sheet states 1% accuracy, which I would assume gives a rough guide. For a 200A device, +/- 2A error is a simplistic expectation.

On the more optimistic side, at low currents, it is to be hoped that using the differential method will minimise the 'zero offset', so that when the real current is zero, then the indicated value will also be close to zero.

----------

So sorry, I can't give you any guarantees of accuracy, but I think the approach minimises one obvious drift area that will probably be of most practical concern.

Best wishes, Dave

Hi @inq,

  With regard to your message showing the op-amp integrator circuit, then the output voltage is indeed the time integral of the input voltage Vin.

For example, if you apply a constant voltage to Vin, then the output 'waveform' will be a straight line whose gradient is proportional to Vin.

In some ways, op-amps are a flashback to a time before digital computers were of a size and cost that were accessible to more than a very few people. But, as you note, even your current sensor project is making use of an op amp to perform the first step. Relatively few people would build an analogue PID system, for a simple heater control, given the low cost of a digital microcontroller, but op-amps and other analogue electronic circuits remain an essential part of the scene. I only mentioned the PID heater to give an idea of the capability, not as a recommendation for your next project.

Best wishes, Dave