Tuching the Transistor Base

[lord loh.]

I recently encountered some problems in one of my projects where the transistor base was floating. Whenever I touched the transistor base, the circuit would trigger. Then I managed to correct thye trouble by placing a base to emitter capacitor.

In an electronics magazine, I saw a project that uses this effect to run a circuit.

My experience is that the triggering is unreliable.

Why does this occur? Static charge? And is there a way to make sure that the transistor gets on when one touches it?

I have heard something like human body capacitance model? Is it related here?

Thank you.

 

[k7elp60]

The base of a transistor is very sensitive. In simple terms take for example a 2N3904 with a hfe of 100. With a collector current of 10 Ma the base current would be 0.1 Ma. If the transistor does not have a biasing circuit to control it, who knows what the collector current will be? The collector current may well be controlled by the resistance in the collector circuit.

With my scope probe just lying on the work bench I can see a AC wave on the screen of about a volt. I move the probe close to my soldering station and the voltage on the screen increases.

Anytime the transistor (bi-polar) sees approximately 0.7 volts between the base and the emitter of the right polarity it's going to turn on.

 

[zachtheterrible]

What kind of transistor was it? Bipolar transistors wouldn't be as sensitive to this as FET transistors because FET transistors have a much higher input impedance.

 

[Someone Electro]

Most probobly a FET.An BJT will thurn on only the that person is toching the suply rails.

 

[akg]

Re: Toching the Transistor Base

I recently encountered some problems in one of my projects where the transistor base was floating. Whenever I touched the transistor base, the circuit would trigger. Then I managed to correct thye trouble by placing a base to emitter capacitor.

In an electronics magazine, I saw a project that uses this effect to run a circuit.

My experience is that the triggering is unreliable.

Why does this occur? Static charge? And is there a way to make sure that the transistor gets on when one touches it?

I have heard something like human body capacitance model? Is it related here?

Thank you.

electricaly human body ois like an antenna , collecting all those rf signal, that can produce a voltage at a high impedence ckt.

 

[Someone Electro]

Yep exsactly that.

FETs are voltage sensitive devices and have a near inifite inpadance.

 

[lord loh.]

It was a BJT BC546. Gain of 250.

Okay the transistor amplifies the RF that I packed up. But it does not work all the time. Is there a way to guarantee that the transistor conducts each time there is a touch. And also make sure that there is no false triggering ?

Thank you.

 

[Nigel Goodwin]

It was a BJT BC546. Gain of 250.

Okay the transistor amplifies the RF that I packed up. But it does not work all the time. Is there a way to guarantee that the transistor conducts each time there is a touch. And also make sure that there is no false triggering ?

Thank you.

There's no way to guarantee it - touch switches have always had a reputation for poor reliability!.

 

[lord loh.]

There's no way to guarantee it - touch switches have always had a reputation for poor reliability!.

Okay....
I remember seeing a touch brightening lamp at one of my father's colleague's place that was placed on a TV that changed states along with the TV Channels.

 

[TKS]

mhhh

i would just use an microcontroller in your project..

using pwm it could easyly trigger the fet bit by bit...

also with the ADC function you can capture....
(use 1 volt ref voltage)


and rocgnize things for example 2 speedy touches..

or 1long touch..

it makes software coming arround and with software the functions are endless..

Tks

 

[akg]

It was a BJT BC546. Gain of 250.

Okay the transistor amplifies the RF that I packed up. But it does not work all the time. Is there a way to guarantee that the transistor conducts each time there is a touch. And also make sure that there is no false triggering ?

Thank you.

using a very high impedence i/p might help. but then the probablity of false triggering may arise.

 

[lord loh.]

Re: mhhh

i would just use an microcontroller in your project..

using pwm it could easyly trigger the fet bit by bit...

also with the ADC function you can capture....
(use 1 volt ref voltage)


and rocgnize things for example 2 speedy touches..

or 1long touch..

it makes software coming arround and with software the functions are endless..

Tks

So can a ADC recognize touches at it's analog terminals? My body voltage might be a few hundred microvolts.... So What do I check with the Software? Duration of the touch? or the voltage level?

using a very high impedence i/p might help. but then the probablity of false triggering may arise.

Why a high impedence? I am using a BJT. And since it is a current operated device, wont is require some minimum amount of cuurent to saturate?

There's no way to guarantee it - touch switches have always had a reputation for poor reliability!.

Will using a darlington of BJT give a better response? I think It might.

 

[Nigel Goodwin]

Re: mhhh

There's no way to guarantee it - touch switches have always had a reputation for poor reliability!.

Will using a darlington of BJT give a better response? I think It might.

There have been many systems used over the years, it makes no real diference - the way they all work is VERY dependent on outside influences (RF fields, hum levels etc.)

Using higher gain makes it more sensitive, but it also makes it more liable to false triggering!.

 

[Roff]

Re: mhhh

My body voltage might be a few hundred microvolts....

Your body voltage might also be thousands of volts.

I think self-excited systems are more reliable. They rely on body capacitance instead of stray fields or static charge. Here are some examples I found from Discover Circuits:

https://circuitos.tripod.cl/schem/r110.gif
https://www.redcircuits.com//Page22.htm
**broken link removed**
https://www.paia.com/touchsw.htm

Disclaimer: I haven't tested any of them. :roll: