Our February column, the first in a series on specifying and creating data-acquisition systems, emphasized analog-to-digital-converter (ADC) resolution and accuracy.
The ADC modules and boards used for data acquisition (DAQ) offer bipolar input ranges of ±x volts, or unipolar ranges of 0-y volts. Just as digital voltmeters have several input ranges, so do ADC modules. But rarely do the sensor voltages completely “fill” these ranges. A 6.5mV sensor output, for example, connected to a 0-10mV ADC input leaves 35 percent of the range unused, as shown in the figure below. For simplicity’s sake, I’ll assume an ideal 14-bit successive-approximation ADC (214 = 16384), so the LSB has a step size of 10 x 10-3 V / 16383 steps, or 0.610 V.
Instead of using a small portion of an ADC’s input range for a sensor input, amplify the signal to better resolve it with smaller voltage steps.
An amplifier placed before the ADC input could boost the 6.5mV sensor output to a maximum of, say, 9.8mV, which comes close to full scale without exceeding it. By “spreading” the 6.5mV signal across a greater span of 16383 ADC steps, you increase the apparent resolution. A 9.8mV signal would represent 16055 steps for the ADC, so for the original sensor signal you now have:
6.5 x 10-3 V / 16055 steps or 0.405 V/steps
The ADC steps remain 0.610V, but each step represents 0.405V from the sensor. You can use the scaling ratio, 1.51 (9.8/6.5), later to calculate actual sensor voltages.
A home-brew amplifier circuit with a gain of 1.51 would boost the 6.5mV signal to 9.8mV, but commercial programmable-gain or instrumentation amplifiers do the job better. Programmable-gain amplifiers offer fixed-gain values, usually in steps of 1, 2, 4, 8 .., 1, 2, 5, 10 .., or 1, 10, 100 ... But none equals 1.51.
Thanks for a very interesting refresher article. I hope to see more of these in the future. It's amazing what one tends to forget when it is not being used everyday.
Yes, Paul, and I requested this correction a few weeks ago when another reader noted the incorrect units. Sometimes symbols don't translate well from a Word document to the HTML. I'll ask again--thanks for your reminder.
Hi, Christopher. You make a good point about sensor ranges, so I'll put it on my topic list for a column after I wrap up this series on data-acquisition. You remind me that it's always good to start on a high range with an instrument and then change to a lower scale as appropriate. I once saw a bent needle on a Simpson VOM someone used to measure line power with a low-voltage setting. It almost made me cry.
Glad to see you presenting this. I've often had to point this out to my junior engineers; now I can point them to this article.
On a parallel subject. Are you going to discuss selecting sensor ranges relative to the measured value in question? I've also encounter situations where the engineer selected something like a 0-100 psi pressure transducer to measure a varying pressure with a mean around 80 psi and then just assumed the occasional 100 psi spikes where as high as the signal got.... I was trained to select a sensor which put the nominal reading at ~50% of the sensor range if I was fairly sure of what I was measuring, and to use less of the range if there was more uncertainty in the measured quantity.
Very nice solution for utilizing the entire range available and increasing the apparent resolution – with a very cost effective and easy to implement solution. Thanks for another great article!
The Machinist Calc Pro computes speeds and feed rates for milling, turning, and drilling: cutting speed, spindle speed, feed rate (inches/minute), cutting feed, etc.
During a recent meeting with engineering-school faculty and alumni, Contributing Technical Editor Jon Titus talked about whether colleges should educate generalists or specialists. What do you think?
From Dell / Intel® New Paradigms in Design Work Scott Hamilton, vertical market strategist for Dell Precision workstations, 5/2/2013 5
Early in my career, I worked as a draftsman and remember the days of drawing on vellum with numbered pencils and Mylar with plastic lead. This was a fun experience in the sense that I ...
I've been using workstations for more than 10 years and love finding ways to get more performance from my system. With demanding professional applications that require more power each ...
A lasting memory from my first job as an engineer in an auto assembly plant is standing on hard concrete at six in the morning, vending-machine coffee clutched in hand, listening to ...
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This radio show will show what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.
To save this item to your list of favorite Design News content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
2 
