View Full Version : FlexiForce Drive Circuit: Power Supply

10-25-2009, 03:20 PM
Hello, I am an amature at electronics and was hoping that someone may be able to help out with my problems!

Rigth now, I am using 18 FlexiForce A201-25 force sensors to instrument a portable sensor glove to measure grip forces experienced during daily living.

I am looking for information/technical assistance regarding the drive circuit to gain the signal from the 18 sensors.

As this is to be a portable system, all of the drive circuits are to run off one supply. The battery has yet to be selected, and want your input/ recommendation on the amount of voltage to power 18 force-to-voltage circuits.

I have examined the recommended drive circuit that Tekscan provides and want to know if this can be applied to 18 sensors simultaneously, and how.

What is the suggested opamp for such a system? Tekscan suggests the MCP6001 opamp, but this has a limited range of 1.8V - 5.5V. Is a power supply that provides 5.5V enough to drive 18 sensors, and their required force to voltage circuit?

Note: the application requires the circuitry to be lightweight, compact and portable.

Any help would be hugely appreciated!!!!



10-26-2009, 03:15 PM
I have examined the recommended drive circuit that Tekscan provides and want to know if this can be applied to 18 sensors simultaneously, and how.No, you need one amp for each FlexiForce sensor... else you won't know which sensor you're reading.

The FlexiForce sensors are just variable resistors. Instead of an op-amp you could create an RC circuit and count the discharge or charge time. That just means count the time it takes for a microcontroller pin to detect that the input voltage has dropped to a logic 0 threshold or up to a logic 1. All you need is a micro and 18 resistors, 18 caps, and 18 FlexiForce sensors. Micro speed and RC determine accuracy. Plus you'll need to tune you timing code for the micro voltage thresholds. This kind of circuit does not take much power.

What is the suggested opamp for such a system?Use whatever op-amp that fits your setup. Tekscan has the op-amp configured as a simply inverting amplifier where gain is Rf/Rs.
As Rs approaches Rf gain gets closer to 1 (5V) . I image you'd want send the output of the op-amp to an A/D then read the A/D output with a micro? This circuit would take a little more power and area. Read the MCP 6001 datasheet http://www.datasheetcatalog.org/datasheet/microchip/21733d.pdf for more information.

What are you doing with the FlexiForce output? Where does it go? How is it processed?

10-27-2009, 09:49 AM
Thanks for the reply MikeG,

As for the use of an RC circuit, i dont really wanna go there (lol). been working on the opamp configuration, and i think that it is going to work for my application. to give a better background on my work, it goes something like this:

-i am instrumenting a glove with 18 Tekscan Flexiforce A201-25 force sensors to measure grip forces encountered in every-day living. The system must be portable (and lightweight), and be able record measurments for up to six hours.

-the 18 sensors will be driven using an inverting-amplifier set up. 5 quad op-amps will be used (allowing for the use of up to 20 sensors).

-the op-amp outputs (18 of them) will be run to five 4:1 multiplexers (the mux are actually available in a 2x 4:1 configuration, so only three physical units are required).

-the multiplexers will be controlled by an appropriate microcontroller to control sampling (i'm thinking of using the ATtiny13a microcontroller).

-multiplexer outputs are recorded using a Logomatic v2 SD Datalogger.

Here are the issues i'm currently facing:

-my biggest problem is power supply. the datalogger is taken care of, as it is going to be run off a seperate battery, so no worries there. However, i have NO CLUE on the battery to select to run the sensor drive ciruit, op-amps, multiplexers and microcontoller.

-op-amp selection is pretty much based on power supply (i assume)....i've been considering using the suggested MCP6001, but its range is limited to 1.8-5.5V (anything over, and its toast); using an LM348 and its range is +-18V; or using the LM124A low power quad op-amp, with a supply range of 3V to 32V. any suggetions here?

-The mux i'm using is the MPC509a from Texas instruments, and i'm not exactly sure what supply voltage it requires.

-The same for the microcontroller....thinking of using the ATtiny13a (unless you have other suggestions), and don't really know what kind of supply voltage it needs.

>if you have ANY suggestions on opamps, mux, or microcontrollers, let me know....and espacially power supply/battery suggestions!!!

Thanks so much!!

10-27-2009, 09:51 AM
also, processing is gonna be done in MatLab....but gotta build it first!

10-27-2009, 10:24 PM
I imagine that Tekscan selected the MCP6001 because of its 1.8V to 5.5V supply voltage specification which fits nicely in a TTL circuit.

Supply voltage for an op-amp defines its output limits. So let’s say you dust off a 741 and connect V+ to 15V. Then connect V- to -15V. Now you have the opportunity to amplify an input signal up to +/- 15V. Let’s also say you have the op-amp configured as an inverting amplifier with a gain of 10. Rf = 100k and Ri = 10k so -Rf/Ri = Gain. Rf is the feedback resistor and Ri is the input (or sensor). If the input voltage at –v (the voltage across Ri) is -1V the output voltage at Vout is 1*10 = 10V.

Flexiforce sensors have a large default resistance. If Ri is a Flexiforce, applying pressure causes the sensor’s resistance to decrease and approach Rf. That means that the gain approaches 1 or +/- the input voltage which is across Ri.

So why did I explain this? Well two reasons… I’m not sure that you understand op-amp supply voltage. +/-15V means you need 30V total. That’s a lot of batteries. Two, I’m not sure you understand gain. I could be wrong though… Just ask my wife.

I would simply measure the time it takes for a cap to charge/discharge and pick up a micro with more IO. Your supply voltage would be 3.3 to 5V depending on the micro.

10-28-2009, 03:28 AM
There's basically two reasons to choose low-voltage opamps.. First is power supply limitations, obviously, and second is speed.

The less voltage swing that needs to be overcome, the faster it'll work. In wide lines, sort of the reason computer microprocessors have been using lower and lower voltages over the ages, although it poses a whole new array of problems.

Personallly, I'd shoot for a 4.8v power supply. Close enough to 5v, which is easy to work with, and you can get more amp/hours out of the same volume as you'd get with a higher voltage.

You could go even lower though, and opt for a single cell lipo. At 3.2v you should be ok with most 3.3v circuits, but you'll have to take into account the larger influence of interference of all sorts.