Trying to make a transistor cancel another transistor

[David Lacroix]

Hey guys,

Thank you very much for all the quick responses! This was a learning experiment, I just started playing around with the most simple components so I might be a bit slow to get all that stuff. The idea behind it was that I thought of a complete project, other than just doing little circuits that means nothing. Basically, what I want to do is a simple thermometer with a series of 4 blue LEDS on one side then going to 4 red LEDS to indicate basic measure of temperature. It would be controlled by a thermistor somewhere in there. The experiment was just to play around with transistor as switches to make sure one LED would not work unless another one was working. Don't ask me how I'm going to do the thermometer project, I haven't figured it out yet!.

Trust me, next time I'm going to try my best drawing a clear schematic with annotations, it's going to make the next thread a bit more clear.

 

[Herschel Peeler]

Hey guys,

Thank you very much for all the quick responses! This was a learning experiment, I just started playing around with the most simple components so I might be a bit slow to get all that stuff. The idea behind it was that I thought of a complete project, other than just doing little circuits that means nothing. Basically, what I want to do is a simple thermometer with a series of 4 blue LEDS on one side then going to 4 red LEDS to indicate basic measure of temperature. It would be controlled by a thermistor somewhere in there. The experiment was just to play around with transistor as switches to make sure one LED would not work unless another one was working. Don't ask me how I'm going to do the thermometer project, I haven't figured it out yet!.

Trust me, next time I'm going to try my best drawing a clear schematic with annotations, it's going to make the next thread a bit more clear.

Super. Take it one step at a time. Four red LEDs and four blue LEDs. What voltage are you going to run it at? That will determine how they will be wired.

 

[David Lacroix]

Super. Take it one step at a time. Four red LEDs and four blue LEDs. What voltage are you going to run it at? That will determine how they will be wired.

Probably, around 4.5 Volts from 3 AA batt in series. How voltage would determine the wiring?

 

[CDRIVE]

Probably, around 4.5 Volts from 3 AA batt in series. How voltage would determine the wiring?

Typically, it doesn't but it's crucial in setting a base line that will dictate component values. For a example consider a simple circuit consisting of a Battery (Vbat), Resistor (R1) and an LED (Led1). If Led1 is to operate at 20mA the value of R1 will be dependent on the value of Vbat as well as the color of Led1.

That's just the basics. It's too broad a question to answer simplistically. I would imagine that asking a power grid engineer if voltage affects how things are physically wired he'd reply with ... "Well Duh!! We can't run cables carrying 500KV within a few inches of each other"!

Chris

 

[Herschel Peeler]

Probably, around 4.5 Volts from 3 AA batt in series. How voltage would determine the wiring?

Okay, first step forward. Four red and four blue LEDs running at 4.5 V from batteries. The size of the resistor determines the current through the LED. Running from batteries you may not want to drive the LEDs at 20 mA or 30 mA. This is one option of what the LED and driver might look like. More or fewer LEDs can be added or removed as desired.
Next step ... When do you want what LEDs to come on?

 

[David Lacroix]

Ok, so let's say I've got 10k NTC thermistor feeding the base of all 8 transistors connected in series to the LEDs. What I thought of, would be to have different increasing rating of resistors on the base of these transistors so when the first blue LED lights up, meaning it's goddamn cold, the next blue LED in line, meaning it's just damn cold, is not lighting up because the resistance on the base of it's transistor is not letting the current flow. It would be something like that so on and so on to the last red LED meaning it's scorching hot. I would be using a 2n5401 PNP transistor because....well, I have some.

 

[Herschel Peeler]

Ok, so let's say I've got 10k NTC thermistor feeding the base of all 8 transistors connected in series to the LEDs. What I thought of, would be to have different increasing rating of resistors on the base of these transistors so when the first blue LED lights up, meaning it's goddamn cold, the next blue LED in line, meaning it's just damn cold, is not lighting up because the resistance on the base of it's transistor is not letting the current flow. It would be something like that so on and so on to the last red LED meaning it's scorching hot. I would be using a 2n5401 PNP transistor because....well, I have some.

Ah, I see. Now that is something different. Lots of parts or an IC, LM3914.

 

[CDRIVE]

If you're more interested in learning, seeing and feeling (including failures) how a transistor works then stick with discrete circuitry. Chips are basically sh!t loads of solid state junctions (resistors too) all crammed into a black plastic thingy that looks like a multi-legged bug. You won't learn much about transistors and transistor circuitry by bypassing them in favor of IC's. Besides, chips don't usually drive things like LEDs and Relays directly. Transistors are used for that. Though there are all sorts of LED or Relay driver IC's available too.

In the case of your Thermistor / LED Thermometer (experiment, project?) I can think of quite a few transistor designs that would accomplish what you described. I'm not saying that IC's won't do it better or even that you'll be able to achieve the accuracy, repeat ability and temperature stability that can be attained with today's myriad of modern chips. What I will garantee you is that you'll learn a hell of a lot more and be doing so in the correct order of importance. After all, a chip will remain a magic monolith without solid understanding and (hands on) experience with transistors.

Chris

 

[Herschel Peeler]

Ok, so let's say I've got 10k NTC thermistor feeding the base of all 8 transistors connected in series to the LEDs. What I thought of, would be to have different increasing rating of resistors on the base of these transistors so when the first blue LED lights up, meaning it's goddamn cold, the next blue LED in line, meaning it's just damn cold, is not lighting up because the resistance on the base of it's transistor is not letting the current flow. It would be something like that so on and so on to the last red LED meaning it's scorching hot. I would be using a 2n5401 PNP transistor because....well, I have some.

Attached is a next step. Eight LEDs, 4 blue, 4 red, each turned on individually, On one section two transistors are added (Q9, Q10, R17, R18, R19) are shown. R17 and R18 are a voltage divider. When the voltage applied to R19 reaches the voltage at the voltage divider plus about 0.6 V Q9 turns on, which turns on Q10, which turns on Q8 which turns on that LED.
You need to add an equivalent circuit of Q9, Q10, R17, R18 and R19 to each of the other LED drivers. Selecting the resistors at the voltage dividers to set at what voltage that LED will turn on.
Now all we need to do is make our temperature sensor circuit that gives us a varying voltage according to the temperature. (Coming up next)

This gives us a bar graph display that lights up the LEDs matching the temperature. I haven't figured out a way to blank the lower LEDs yet.

Or we can use an LM3914 and avoid a lot of these parts.

 

[hevans1944]

Whatever happened to the "good old days" when everyone was trying to improve on the AM-band crystal set receiver by using them thar newfangled and expensive transistor thingamabubs? Audio amplifiers, regenerative RF amplifiers, and such? Lots of fun trying to figure out bias and trading off gain for distortion with nary a text book much in sight for the hobbyist. And no Internet with all its information and misinformation to sort through. Do I miss it? Well, not so much... and Forrest M. Mims III is still around, now with an Internet presence, without whom some of us would probably be watching television by candlelight. No, wait, that was Edison. Or was it Farnsworth? Tesla? <sigh>

 

[Herschel Peeler]

Ok, so let's say I've got 10k NTC thermistor feeding the base of all 8 transistors connected in series to the LEDs. What I thought of, would be to have different increasing rating of resistors on the base of these transistors so when the first blue LED lights up, meaning it's goddamn cold, the next blue LED in line, meaning it's just damn cold, is not lighting up because the resistance on the base of it's transistor is not letting the current flow. It would be something like that so on and so on to the last red LED meaning it's scorching hot. I would be using a 2n5401 PNP transistor because....well, I have some.

What temperature range were you interested in monitoring? How much does your sensor change per degree? Do you hae a part number we can look up?

 

[Harald Kapp]

Or we can use an LM3914 and avoid a lot of these parts.

Imho the only sensible and reliable way to go.

 

[Herschel Peeler]

Imho the only sensible and reliable way to go.

Agree. This is a learning experience for him.

 

[Herschel Peeler]

Probably, around 4.5 Volts from 3 AA batt in series. How voltage would determine the wiring?

One section like the three transistors on each LED section. Lots of parts ... or an LM3914 which is designed to do what you want lighting one LED at a time.

 

[CDRIVE]

Really (LM3914) guys? I went back and re-read from post 1.

Hey guys,

Thank you very much for all the quick responses! This was a learning experiment, I just started playing around with the most simple components so I might be a bit slow to get all that stuff. The idea behind it was that I thought of a complete project, other than just doing little circuits that means nothing. Basically, what I want to do is a simple thermometer with a series of 4 blue LEDS on one side then going to 4 red LEDS to indicate basic measure of temperature. It would be controlled by a thermistor somewhere in there. The experiment was just to play around with transistor as switches to make sure one LED would not work unless another one was working. Don't ask me how I'm going to do the thermometer project, I haven't figured it out yet!.

Trust me, next time I'm going to try my best drawing a clear schematic with annotations, it's going to make the next thread a bit more clear.

I then looked at David's schematic again. Remember this?

Read some more of David's replies, especially this one..

Ok, so let's say I've got 10k NTC thermistor feeding the base of all 8 transistors connected in series to the LEDs. What I thought of, would be to have different increasing rating of resistors on the base of these transistors so when the first blue LED lights up, meaning it's goddamn cold, the next blue LED in line, meaning it's just damn cold, is not lighting up because the resistance on the base of it's transistor is not letting the current flow. It would be something like that so on and so on to the last red LED meaning it's scorching hot. I would be using a 2n5401 PNP transistor because....well, I have some.

I'm not going to repeat everything I said in my last post as it would appear to be pointless. This should suffice......

Good morning class and welcome to Transistors 101. We're going to begin our studies by demonstrating how the basic bipolar transistor is a linear device. We will also demonstrate how bipolar transistors lend themselves well to switching and control systems.

Please turn to page 1000 where you'll find the LM3914 Dot / Bar Display Driver.

Is Hop my only Ally?

There was a lot to admire about Kris Blue (RIP ) but what I loved the most was his love of designing discrete circuitry. I used to tease him about it while hoping he'd never stop drawing them.

Chris

 

[hevans1944]

I do agree with Chris on this: you have to learn to crawl before you can stand up and walk through the forest that is modern electronics. To jump from "foolin' around" with 2N5401 PNP transistors (because you have a boatload of 'em) to using integrated circuits doesn't require a leap of faith (witness all the junk on Instructables and other hobby sites), but I doubt you can learn much about electronics that way. So, if the goal is to learn about transistors and electronics, start with basic electricity and learn how Ohm's Law and Kirchoff's Laws are used to solve for currents and voltages in simple resistor networks. Once you have those tools hanging from your tool belt proceed to learn about capacitors and inductors. Then add diodes to the mix and, eventually, transistors. Well, that's how I began to learn electronics in the 1950s... maybe someone has discovered some shortcuts by now, but somehow I doubt that.

 

[CDRIVE]

Here's a discrete concept you may like or not. It uses transistors in their linear mode. It should not be considered a finished circuit but does demonstrate a discrete method of lighting LEDs in a bar graph fashion. IE, when the next led lights the previous led does not go off. This better mimics a mercury thermometer anyway.

All transistors are configured as Common Collector Amplifiers (Emitter Followers). In this configuration a transistor cannot provide voltage gain but does provide current gain. They also provide a low output impedance (Emitter to GND) as well as relatively high input impedance (Base to GND). P1 is set at 100%. P2 is set to 80% and P3 is set at 60%.

As you can see from the LED current curves each LED gets progressively brighter until the next LED up the ladder begins to turn on also.

If learning to design Transistor circuitry is what you desire then give it a go. As I said earlier, I'm not claiming it's the best choice around but it certainly demonstrates transistors as current amplifiers.

Chris

 

[Herschel Peeler]

Here's a discrete concept you may like or not. It uses transistors in their linear mode. It should not be considered a finished circuit but does demonstrate a discrete method of lighting LEDs in a bar graph fashion. IE, when the next led lights the previous led does not go off. This better mimics a mercury thermometer anyway.

All transistors are configured as Common Collector Amplifiers (Emitter Followers). In this configuration a transistor cannot provide voltage gain but does provide current gain. They also provide a low output impedance (Emitter to GND) as well as relatively high input impedance (Base to GND). P1 is set at 100%. P2 is set to 80% and P3 is set at 60%.

As you can see from the LED current curves each LED gets progressively brighter until the next LED up the ladder begins to turn on also.

If learning to design Transistor circuitry is what you desire then give it a go. As I said earlier, I'm not claiming it's the best choice around but it certainly demonstrates transistors as current amplifiers.

Chris

View attachment 29389
View attachment 29390

A lot simpler than mine. Can you do it with PNP's like he wanted?

 

[(*steve*)]

If you were really keen you could drive the NTC thermistor from a constant current source, buffer the voltage with an emitter follower, use this to drive a series of long tailed pairs comparing the voltage to a reference and driving a LED.

You would use a shed-load of transistors in almost every configuration known to modern man.

Just to reduce repetition, I'd probably not want to build 8 stages. But hey, more power to you if you want to do it.

Unless I really wanted to just use bipolar transistors, I would probably use a jfet (rather than 2 transistors and 2 resistors) as a current source.

 

[(*steve*)]

A lot simpler than mine. Can you do it with PNP's like he wanted?

Sure, you just turn it on it's head.

Essentially, swap the power supply pins and turn the LEDs around.