Hi, I'm new and I'm building some simple circuits for fun.
I'm looking for a component / a group of component that can function like a switch.
What I am thinking is something like a normal three pin switch, but instead of using mechanical methods to press the switch, you can use a small voltage to turn it on/off.
I've googled for it and transistor poped up. But I think its not the thing I'm looking for.
I'm looking for a switch that close when a voltage is applied. And the switch remain closed after the voltage is removed. And when a voltage is applied again, the switch becomes open until its triggered again.
Edit: should this be at circuits forum? I just notice that there is one after i post this but i couldnt delete this post and move it there

 

MOSFET or IGBT.

The gate is isolated so there is no current flow. Apply a small voltage to the gate and you can turn on an N Type MOSFET or IGBT. I think. I often get those backwards.

The thing is I had a MOSFET and knew nothing about it. Connected a supply and a motor (don't remember how) and touched the gate with my finger. The switch came on full power and the motor ran at full speed. Touching the battery negative terminal and touching the gate (I think) did make the motor stop completely.

I then connected a six inch length of insulated wire to the gate. I could grasp the insulation and the static from my body induced a voltage into the gate and the motor gradually increased speed until it was at full speed.

The motor was a small one out of a CD player and the MOSFET was out of a pool spa controller. Rated to handle large voltages and currents. I'd have to go into my lab and find all that stuff to be able to tell you exactly what I had and recreate the experiment just to be able to tell you exactly how I did that.

But the point is that if you charge the gate it should hold that charge for a long time. Depending on static levels and humidity the time will probably vary. And I'm NOT the expert on MOSFETS. I'm still trying to understand them myself.

 

MOSFET or IGBT.

The gate is isolated so there is no current flow. Apply a small voltage to the gate and you can turn on an N Type MOSFET or IGBT. I think. I often get those backwards.

The thing is I had a MOSFET and knew nothing about it. Connected a supply and a motor (don't remember how) and touched the gate with my finger. The switch came on full power and the motor ran at full speed. Touching the battery negative terminal and touching the gate (I think) did make the motor stop completely.

I then connected a six inch length of insulated wire to the gate. I could grasp the insulation and the static from my body induced a voltage into the gate and the motor gradually increased speed until it was at full speed.

The motor was a small one out of a CD player and the MOSFET was out of a pool spa controller. Rated to handle large voltages and currents. I'd have to go into my lab and find all that stuff to be able to tell you exactly what I had and recreate the experiment just to be able to tell you exactly how I did that.

But the point is that if you charge the gate it should hold that charge for a long time. Depending on static levels and humidity the time will probably vary. And I'm NOT the expert on MOSFETS. I'm still trying to understand them myself.

Thanks for your help It helped a lot It'll take forever if I have to find them from google LOL
I'll check out more information about the mosfet and igbt after i go home from school

 

One place I like to find specs is on Digikey (dot) com. But usually you have to have a part number in order to find a spec sheet.

My workbench is a mess (typically) but I'll see if I can find that MOSFET. If so I'll see about posting a link to the spec sheet, and maybe even recreate the motor experiment for you.

 

Hi DHMO, welcome to the forums.

Here is what I gather you wish to do: you have a black box with wires A B and C. Initially A to B is open. You alloy a voltage to C and B and the switch between A and B closes. Switch remains closed until you remove and again reapply the voltage from C to B.

That sounds like a logic gate to me, specifically a D type flip flop. These come as integrated circuits, usually at least two in each package. Then you need something to drive your output like a transistor or a MOSFET.

Using a MOSFET by itself will give you problems, and it will not turn off as you wish.

This is not a simple basic beginners circuit problem but anywhere you start is good. So.... What voltage will you use to turn this on and off, will it have an additional power supply, and what does the switch part need to do, how many volts and amps is it controlling?

 

this sounds like a job for a relay

but... can you provide more info about your applications:
1. if this AC or DC circuit?
2. what kind of voltage and current range are expected for switched circuit?
3. is the "voltage signal" supposed to be galvanically isolated from switched circuit?
4. what kind of voltage and current range are expected for controlling signal?

 

Welcome to AAC!

I'm looking for a switch that close when a voltage is applied. And the switch remain closed after the voltage is removed. And when a voltage is applied again, the switch becomes open until its triggered again.

You're asking for an alternate action type of switch. You need more than a transistor or MOSFET. You need to add a flip flop to get the alternate action.

 

A self latching relay might do this all by itself. I've never used one myself so I don't know offhand if a single input can make it flip and flop easily.

 

Here's what I am finding: I have a 2SK1277 MOSFET. Data sheet is http://pdf1.alldatasheet.com/datasheet-pdf/view/60808/FUJI/2SK1277.html

The diagram below shows a 6 volt battery (4 AA), a MOSFET and a Motor. Touch with your fingers POINT A & C. The motor comes on. Why? Because the gate now has a charge. Touch POINT B & C and the motor stops. Why? Because you've removed (or reversed) the charge on the gate.

IF you touch ONLY C while the motor is running, it's possible the motor will slow down to a stop. If not - rub your feet on the floor. My motor stops when I rub my bare feet on the carpet. Who knows what yours will do - it depends on whether you're building a negative charge on your body or a positive one. IF you are building a positive charge then if the motor is not running and you touch POINT C the motor may begin running.

By briefly touching B & C I can slow the motor down so much that it's easily visible which way the motor is spinning. But I also stuck some wire insulation and some cotton fibers on the shaft so it's easy to see when it's running and when it's not.

 

A self latching relay might do this all by itself.

Self latching (the ones I have) require voltage on one coil to activate it. To deactivate it you need to apply a voltage to the OTHER coil.

 

Because of the mis-match between components and voltages I'm never turning my MOSFET fully on, so the motor never runs at full speed. But the interesting thing about this circuit is that you can experiment with static charge. I just sat down after handling several plastic boxes. Boxes that are NOT supposed to present any static danger to sensitive components. Sitting down in my chair and just touching C made the motor run faster than if I touched A & C. That meant I was putting a higher charge on the gate than what was available from my small battery.

From what you asked, it seems like you want to touch a button once and turn something on, then touch it again and turn it off. No doubt this is a logic circuit because it's depending on a single input to toggle from off to on and back to off again. My circuit doesn't do that. But it also doesn't incorporate a switch. All it needs is a touch pad and a close ring so that when your finger touches it you make contact between A & C. Another touch pad and close ring for contacting B & C to shut it off. Two different touch points, but they are not switches.

The thing about the MOSFET I have is that it came out of a pool spa controller. It's supposed to manipulate 220 VAC. Maybe not directly, but I am not familiar with how the system was set up. Nevertheless, my MOSFET requires a higher gate voltage to achieve full on. Hard grounding should open the gate sufficiently to not present any significant drain to any batteries - but again, with FET technology I'm still learning a lot about it. So others may have more valid points than what I say. All I can say is that on my bench, the circuit is working as advertised. If need be I'll post a video.

Here's what I mean by "Touch Pad" and "Close Ring". You would need two of these to turn your circuit on and to turn it off. Depending on which pad and ring you touch. (with one finger)

 

This will do what you want:

Do you need a circuit description?

 

@emfields: Stop being so elegant.

 

This will do what you want:
View attachment 121622
Do you need a circuit description?

Sure can you describe how a CD4013 works without any power? Or why many many input lines are not connected? Or why an unknown input voltage will trigger this circuit once and only once? How did you conclude the switch output will always be one polarity and low enough to be handled by the 2N2222?

Oh, and what is R2 for?

(This is why we get the full requirements before we design a circuit.)

@tony: OP asks for a voltage input, not a touch point. Such is possible but highly unreliable if not done properly, such as directly exposing a gate to an unknown human potential. At best the circuit may kinda work until a dry winter day when ESD blows the gate right out.

 

Sure can you describe how a CD4013 works without any power?

It can't, of course, but in all of my years in this business I've understood that it's generally accepted that power and ground connections to digital chips, on conceptual schematics, aren't shown but are de rigueur and are accepted as being there.

Or why many many input lines are not connected?

Since it was only necessary to show one D type flip-flop in order to illustrate the concept of a divide-by-two circuit which would meet the OP's expectations, I felt it unnecessary to include the other flip-flop in the package, only to have to spare it out.

Or why an unknown input voltage will trigger this circuit once and only once?

While the OP didn't specify an input trigger voltage, my graphic clearly shows it as being equal to about half of the supply voltage and, with a noiseless transition, would toggle the dflop only once for every input pulse.

Or did you mean something else?

How did you conclude the switch output will always be one polarity and low enough to be handled by the 2N2222?

I don't understand that one. Can you elaborate, please?

Oh, and what is R2 for?

R2 is the load which is being switched ON and OFF by Q1

(This is why we get the full requirements before we design a circuit.)

Thank you for your insights. I do suggest, however, that if you have constructive criticism to offer you do it, in future, by posting a schematic/circuit diagram instead of a thousand words.

 

It can't, of course, but in all of my years in this business I've understood that it's generally accepted that power and ground connections to digital chips, on conceptual schematics, aren't shown but are de rigueur and are accepted as being there.



Since it was only necessary to show one D type flip-flop in order to illustrate the concept of a divide-by-two circuit which would meet the OP's expectations, I felt it unnecessary to include the other flip-flop in the package, only to have to spare it out.



While the OP didn't specify an input trigger voltage, my graphic clearly shows it as being equal to the supply voltage, and with a noiseless transition, would toggle the dflop only once for every input pulse.

Or did you mean something else?



I don't understand that one. Can you elaborate, please?



R2 is the load which is being switched ON and OFF by Q1



Thank you for your insights. I do suggest, however, that if you have constructive criticism to offer you do it, in future, by posting a schematic/circuit diagram instead of a thousand words.

You may have such experience, I certainly do, but do you expect our new friend DMHO to also have such experience? I don't, which is why we all should be careful in how we answer him. That prompted my 1,000 word response when I saw the thread heading to calamity.

Now if DMHO comes back and answers the questions I posed to him way back in post #5 we may be able to make some real headway in helping him solve his problem, and teaching him something too as a bonus.

 

I'm looking for a switch that close when a voltage is applied. And the switch remain closed after the voltage is removed. And when a voltage is applied again, the switch becomes open until its triggered again.

Plenty of solutions depending on the load but as mentioned a "D" Flip Flop or a latching relay which uses a momentary pulse to latch followed by another momentary pulse to un latch. Again, the best and most practical solution depends on the load.

Ron

 

OP asks for a voltage input, not a touch point. Such is possible but highly unreliable if not done properly, such as directly exposing a gate to an unknown human potential. At best the circuit may kinda work until a dry winter day when ESD blows the gate right out.

So very true. Hadn't considered that.

On my test bench, the device I am running, the motor runs at best - half speed. Most likely because I'm not giving enough gate voltage to turn the FET all the way on. I have noticed that when I stand up while touching the C point I can make the motor run at closer to full speed. So yes, I am introducing some static field potential into the gate. And when I rub my feet on the carpet I generate negative voltages which close the gate down. So yes, you're right about ESD. Funny, I made 30 years of living dealing with ESD. How could I have overlooked that this time?!

Goes to show the younger ones that everyone can make mistakes.

 

Tony: MOSFET gate capacitance is a funny thing and does have its uses. Way back I used this simple DMM as proof of life on a fet: short G to S, read ohms D to S, look for an open. Then read ohms G to S, and look for conduction D to S. this worked only because the ohmmeter could charge the gate enough to get some conduction. It worked as proof of life because I am seeing the device work both on and off as a simple switch.