06-24-10
V-Sensor Candidates
This list has a ceiling: There aren’t that many companies making them!
At the top of the pile is the Texas Instruments TPS3803-01. It says right on the DS that it’s applicable to ‘Intelligent Instruments’! There’s been some digging around and I’m coming up for air to report.
Most of these devices work with a couple of vital internal circuits. The first is something called a ‘Reference Voltage’ which is contained within a sharp cornered block diagram square with the words ‘Reference Voltage’ printed inside. That’s the proprietary part. My guess is it has something to do with Mr. Zener… but what do I know. The other common block is a comparator (I just assume that these behave pretty much like a 741, and don’t ever have to deal with high frequency). The one above is Shmitt-y. The Reference Voltage is connected to the comparator (+,-, it varies) and usually serves as a very low V ref, likely around 1.2V. The other end of the comparator input is connected to an internal V divider, and the V Input of the device plugs in to the top of that V divider. The resistor values on the divider determine the threshold of the particular part number in the series. Internal hysteresis (advertised as making the part ‘immune to noise’) is derived from the Shmitt-y-ness of the comparator. It’s useful, and some device datahseets will walk you through, enhancing the hysteresis with some external C or RC at the output or input.
Anyhoo, that’s basically what you get when you order a voltage supervisor under a buck. The trick I’m trying to pull, of course, is to monitor my own power source and act accordionly. Last time I used the STM device, I was fighting its internal voltage divider, and that’s why my trigger voltage wouldn’t get down as far as the 3.1 stated on the package. It would not be the end of the world if the situation called for working with the ST part, but since we traveled all this way to get here….
To the rescue, TPS3803-01! To quote from the Almighty datasheet:
…whereas the TPS3803-01 has an adjustable SENSE input that can be configured by two external resistors.
Viola.
Imagine if you will, a digital potentiometer wiper pin attached to the SENSE pin above. Now, you can detect a voltage live on the fly anywhere from 1.226V to 7.0V (max rating).
It comes only in a 5 pin SC-70 package (2,4mm x 2.15mm extent) and that’s an eye popping bummer. but hey, at $0.75 each if that’s the biggest problem? we’re doing ok.
On top if it all, the output is ‘open drain’, which means that it’s connected to the business end of a Metal Oxide Semiconductor Field Effect Transistor – MOSFET. Beautiful thing the MOSFET. If you want to change the state of the output, all you have to do is charge (or discharge, depending. By the looks of the comparator that it’s tied to, I’d say we’re discharging here) a teeny tiny capacitor inside the Gate pin. A very simple process that doesn’t effect anything down stream. No leakage current and no weird voltages on the output that are not all the way to GND. Any Voltage on Vdd lower than the threshold applied to the Sense pin will cause the device to output LOW, as in a direct electrical connection to GND (some resistance applies). That LOW is what they call the asserted state. It is, after all, a POR/BOR device. It will require an external pull up resistor in order to feel the deasserted state (logic HIGH) because then the MOSFET looks like the cathode end of a diode: high impedance. There it is. This is not to say that there is not some creeping voltage drop on transitioning this open drain MOSFET between its states. I will take a closer look at that falling edge in a future post.
Thanks for playing. Next up, Recharging Circuitry, I promise.
