Dec 26 2017

# The Techniques to Improve Capacitive Touch Sensing

We have published the article to talk some basic knowledge about knowledge of Capacitive Touch Sensing of Sensor. In this article,there are some basic cap-sense circuit configurations and discusses how to deal with low-frequency and high-frequency noise.

Measuring change

If you have read the article that I noticed at the first garagraph, you will know that the essence of capacitive touch sensing is the change in capacitance that occurs when an object approaches a capacitor.The presence of a finger increases the capacitance by

• 1) introducing a substance (i.e., human flesh) with a relatively high dielectric constant

• 2) providing a conductive surface that creates additional capacitance in parallel with the existing capacitor.

Right!  the mere fact that the capacitance changes is not particularly useful. To actually perform capacitive touch sensing, we need a circuit that can measure capacitance with enough accuracy to consistently identify the increase in capacitance caused by the presence of the finger. There are various ways to do this, some quite straightforward, others more sophisticated. In this article we will look at two general approaches to implementing capacitive-sense functionality; the first is based on an RC (resistor–capacitor) time constant, and the second is based on shifts in frequency.

RC Time Constant

When my first time to realized that higher math actually has some relationship with the exponential curve representing the voltage across a charging or discharging capacitor,I experienced a vague feeling of university nostalgia. There’s something about it—maybe that was one of the first times I realized that higher math actually has some relationship to reality, or maybe in this age of grape-harvesting robots there is something appealing about the simplicity of a discharging capacitor. In any event, we know that this exponential curve changes when either resistance or capacitance changes. Let’s say we have an RC circuit composed of a 1 MΩ resistor and a capacitive touch sensor with typical fingerless capacitance of 10 pF.

We can use a general-purpose input/output pin (configured as an output) to charge the sensor cap up to the logic-high voltage. Next, we need the capacitor to discharge through the large resistor. It’s important to understand that you cannot simply switch the output state to logic low. An I/O pin configured as an output will drive a logic-low signal, i.e., it will provide the output with a low-impedance connection to the ground node. Thus, the capacitor would discharge rapidly through this low impedance—so rapidly that the microcontroller could not detect the subtle timing variations created by small changes in capacitance. What we need here is a high-impedance pin that will force almost all of the current to discharge through the resistor, and this can be accomplished by configuring the pin as an input. So first you set the pin as a logic-high output, then the discharge phase is initiated by changing the pin to an input. The resulting voltage will look something like this:

If someone touches the sensor and thereby creates an additional 3 pF of capacitance, the time constant will increase, as follows:

The discharge time is not much different by human standards, but a modern microcontroller could certainly detect this change. Let’s say we have a timer clocked at 25 MHz; we start the timer when we switch the pin to input mode. We can use this timer to track the discharge time by configuring the same pin to function as a trigger that initiates a capture event (“capturing” means storing the timer value in a separate register). The capture event will occur when the discharging voltage crosses the pin’s logic-low threshold, e.g., 0.6 V. As shown in the following plot, the difference in discharge time with a threshold of 0.6 V is ΔT = 5.2 µs.

With a timer clock-source period of 1/(25 MHz) = 40 ns, this ΔT corresponds to 130 ticks. Even if the change in capacitance were reduced by a factor of 10, we would still have 13 ticks of difference between an untouched sensor and a touched sensor.

So the idea here is to repeatedly charge and discharge the capacitor while monitoring the discharge time; if the discharge time exceeds a predetermined threshold, the microcontroller assumes that a finger has come into “contact” with the touch-sensitive capacitor (I put “contact” in quotation marks because the finger never actually touches the capacitor—as mentioned in the previous article, the capacitor is separated from the external environment by solder mask and the device’s enclosure). However, real life is a little more complicated than the idealized discussion presented here; error sources are discussed below in the “Dealing with Reality” section.

Variable capacitor, Variable frequency

In the frequency-shift-based implementation, the capacitive sensor is used as the “C” portion of an RC oscillator, such that a change in capacitance causes a change in frequency. The output signal is used as the input to a counter module that counts the number of rising or falling edges that occur within a certain measurement period. When an approaching finger causes an increase in the capacitance of the sensor, the frequency of the oscillator’s output signal decreases, and thus the edge count also decreases.

The so-called relaxation oscillator is a common circuit that can be used for this purpose. It requires a few resistors and a comparator in addition to the touch-sensitive capacitor; this seems like a lot more trouble than the charge/discharge technique discussed above, but if your microcontroller has an integrated comparator module, it’s not too bad. I’m not going to go into detail on this oscillator circuit because 1) it is discussed elsewhere, including here and here, and 2) it seems unlikely that you would want to use the oscillator approach when there are many microcontrollers and discrete ICs that offer high-performance capacitive-touch-sense functionality. If you have no choice but to create your own capacitive-touch-sensing circuit, I think that the charge/discharge technique discussed above is more straightforward. Otherwise, make your life a little simpler by choosing a microcontroller with dedicated cap-sense hardware.

The capacitive-sense peripheral in the EFM32 microcontrollers from Silicon Labs is an example of an integrated module based on the relaxation-oscillator approach:

The multiplexer allows the oscillation frequency to be controlled by eight different touch-sensitive capacitors. By quickly cycling through the channels, the chip can effectively monitor eight touch-sensitive buttons simultaneously, because the microcontroller’s operating frequency is so high relative to the speed at which a finger moves.

Reality

We must notice that a capacitive-touch-sense system will be afflicted by both high and low-frequency noise.

The high-frequency noise causes minor sample-to-sample variations in the measured discharge time or edge count. For example, the finger-less charge/discharge circuit discussed above might have a discharge time of 675 ticks, then 685 ticks, then 665 ticks, then 670 ticks, and so forth. The significance of this noise depends on the expected finger-induced change in discharge time. If the capacitance increases by 30%, the ΔT will be 130 ticks. If our high-frequency variation is only about ±10 ticks, we can easily distinguish signal from noise.

However, a 30% increase in capacitance is probably near the maximum amount of change that we could reasonably expect. If we get only a 3% change, the ΔT is 13 ticks, which is too close to the noise floor. One way to reduce the effect of noise is to increase the magnitude of the signal, and you can do this by reducing the physical separation between the PCB capacitor and the finger. Often, though, the mechanical design is constrained by other factors, so you have to make the best of whatever signal magnitude you get. In this case, you need to lower the noise floor, which can be accomplished by averaging. For example, each new discharge time could be compared not to the previous discharge time but to the mean of the last 4 or 8 or 32 discharge times. The frequency-shift technique discussed above automatically incorporates averaging because small variations around the mean frequency will not significantly affect the number of cycles counted within a measurement period that is long relative to the oscillation period.

Low-frequency noise refers to long-term variations in the fingerless sensor capacitance; these can be caused by environmental conditions. This sort of noise cannot be averaged out because the variation could persist for a very long period of time. Thus, the only way to effectively deal with low-frequency noise is to be adaptable: The threshold used to identify the presence of a finger can’t be a fixed value. Instead, it should be regularly adjusted based on measured values that do not exhibit significant short-term variations, such as those caused by the approach of a finger.

All in all,we notice that capacitive touch sensing does not require complex hardware or highly sophisticated firmware. It is nonetheless a versatile, robust technology that can provide major performance improvements over mechanical alternatives.

• Jerseys NFL Wholesale on 2018-3-15 6:19:42

These two objectives are the key ones to understand, though two others may also figure in.

• author on 2018-3-16 15:25:48
author

Re:

• Cheap China Jerseys on 2018-3-28 20:39:49

This is an electronic blog that provide useful electronic knowledge or the latest semiconductor news electronic engineers.Anyone who wants to use this platform to do unrelated advertisements is not allowed.You contact the author's mailbox directly for any idea. Anyway,thanks for your pages to provide a lot of valuable electronic knowledge. I like it.

• nika on 2018-3-29 10:39:55

According to the working principle of capacitive sensor, there are three basic types of capacitive sensors, namely variable distance (d) type (also called variable gap type), variable area (A) and variable Permittivity (epsilon) type. The variable gap type can measure the displacement, the variable area type can measure the linear displacement, the angular displacement and the size, and the variable dielectric constant can measure the liquid level and the material thickness.

• kamiatar on 2018-3-29 10:40:39

The determinant of the capacitance is: C=εS/4πkd. The epsilon is a dielectric constant, the S is the right pair area of the capacitor plate, the D is the distance of the capacitor plate, and the K is the constant of the static electric power.

• author on 2018-3-29 10:49:43
author

Re:

There is an page named "Comprehensive Explanation of Capacitors" about capacitors, such as what is capacitor,its clasidfication? Features? Types? Function and more,the page is comprehensive. You may like it.

• samitura on 2018-3-29 10:42:24

Compared with the resistance and inductance sensors, the capacitive sensor has the following advantages: 1. Good temperature stability 2. The structure is simple and the adaptability is strong. 3. The dynamic response is good 4. The non-contact measurement can be realized with the average effect

• author on 2018-3-29 10:45:17
author

Re:

Hi~samitura

U R Right,everything has its two sides.So~there are some shortcomings of capacitive sensors:

1. High output impedance and poor load capacity

2. Great influence of parasitic capacitance

3. output characteristic nonlinearity

• yakami on 2018-3-29 10:51:56

Since the end of the 70s, with the development of integrated circuit technology, a capacitive sensor which has been encapsulated with micrometer has been developed. This new type of sensor can greatly reduce the influence of distributed capacitance and overcome its inherent disadvantages. The capacitive sensor is one of the most widely used and promising sensors.

• sarah on 2018-4-9 18:02:40

GOOD

• Cheap NFL Jerseys Wholesale on 2018-4-24 21:51:34

It’s not enough, but that’s all we can afford.

• Wholesale Jerseys on 2018-5-6 18:46:11

It is a doable thing if you just use common sense.

• China Jerseys on 2018-5-23 4:32:36

Merchants in other nations are not required to submit this information, but Google has indicated that it may become required in the future.

• Jerseys From China Cheap on 2018-5-28 23:08:03

So, what we tell customers is, ‘Here’s what you can do today.

• Leonie on 2018-6-11 22:52:09

I enjoy the article

• where can i buy nfl jerseys on 2019-2-27 19:30:32

Thanks for another excellent article. Where else could anyone get that kind of information in such a perfect way of writing? I've a presentation next week, and I am on the look for such information.

• cheap real jerseys nfl on 2019-3-26 13:40:06

Thank you for another great article. Where else could anybody get that type of information in such a perfect way of writing? I have a presentation next week, and I am on the look for such information.

• nfl pro jerseys on 2019-3-30 11:43:00

Attractive section of content. I just stumbled upon your web site and in accession capital to assert that I acquire in fact enjoyed account we supply cheap nfl jerseys your blog posts. Anyway I will be subscribing to your augment and even I achievement you access consistently fast.

• Sol on 2019-4-5 18:57:30

It works very well for me

• where can i buy cheap jerseys on 2019-4-24 12:31:09

We're a group of volunteers and opening a new scheme in our community. Your site provided us with valuable info to work on cheap nfl jerseys china. You have done a formidable job and our whole community will be thankful to you.

• nfl jersey shop wholesale on 2019-4-25 4:37:34

I do agree with all the ideas you have presented in your postcheap jerseys. They are very convincing and will certainly work. Still, the posts are too short for novices. Could you please extend them a little from next time? Thanks for the post.

• game worn nfl jerseys on 2019-4-25 15:59:49

I am extremely impressed with your writing skills postwholesale nfl jerseys and also with the layout on your weblog with cheap nfl jerseys. Is this a paid theme or did you customize it yourself? Anyway keep up the nice quality writing, it is rare to see a nice blog like this one today

• cheap nfl wear on 2019-4-27 17:53:33

I do agree with all the ideas you've presented in your postcheap jerseys. They are very convincing and will definitely work. Still, the posts are very short for novices. Could you please extend them a little from next time? Thanks for the post.

• discount jerseys online on 2019-4-28 18:03:41

I believe that is among the most vital info for me. And i'm glad reading your article. However want to commentary on some basic issues, The web site style is wonderful,postcheap jerseys the articles is in reality great D.Just right job, cheers

• buy sports jerseys cheap on 2019-4-29 2:54:00

I feel this is one of the most vital information for me. And i am happy studying your article. However should commentary on few basic issues, The site taste is ideal,postcheap nike nfl jerseys the articles is actually nice D.Excellent activity, cheers

• cheap retro jordans on 2019-4-29 11:44:19

Very nice post. I just stumbled upon your blog and wanted to say that cheap nfl jerseys china I've truly enjoyed surfing around your blog posts. After all Il be subscribing to your rss feed and I hope you write again very soon!postwholesale nfl jerseys.

• cheap replica nfl jerseys on 2019-4-29 20:55:12

We are a group of volunteers and starting a new scheme in our community. Your web site offered us with valuable information to work on cheap nfl jerseys china. You have done an impressive job and our whole community will be thankful to you.

• redskins jersey cheap on 2019-5-2 2:13:18

Hi there, just became aware of your blog through Google, postcheap nfl jerseysand found that it's really informative cheap nfl jerseys. I gonna watch out for brussels. I will be grateful if you continue this in future. Many people will be benefited from your writing. Cheers!

• cheap nfl jerseys wholesale on 2019-5-2 16:18:18

I feel that is one of the such a lot significant information for me. And i'm satisfied studying your article. However should statement on few basic things, The web site style is perfect,postcheap nike nfl jerseys the articles is actually nice D.Excellent activity, cheers

• cheap jordan shoes on 2019-5-3 2:59:09

I believe that is one of the such a lot important info for me. And i am glad reading your article. However wanna statement on few basic issues, The website style is great,post the articles is in reality great D.Just right process, cheers

• pro sports jerseys cheap on 2019-5-3 22:25:58

Very nice post. I just stumbled upon your blog and wanted to say that cheap nfl jerseys china I've really enjoyed browsing your blog posts. In any case I will be subscribing to your feed and I hope you write again soon!postwholesale nfl jerseys.

• discount nfl jerseys online on 2019-5-5 18:26:18

Wow, fantastic blog layout! How long have you been blogging for? you make blogging look easy to cheap nfl jerseys china. The overall look of your website is excellent, postwholesale nfl jerseysas well as the content!

• bulk nfl jerseys on 2019-5-6 10:30:24

Hi there, just became alert to your blog through Google, postwholesale nfl jerseysand found that it's truly informative cheap nfl jerseys. I going to watch out for brussels. I will appreciate if you continue this in future. Many people will be benefited from your writing. Cheers!