How to limit current

[alexalex]

Hey Eric, you got me to thinking about the other transistors, PNP. Ignoring the resistor values, the below schematic should provide 12V to the motor?

**broken link removed**

Then all I have to do is add the diodes and the other three transistors for the other direction. Is it right to say that Q1 doesn't need a resistor on the base because it's actually on the emitter? Also it appears that the hFE of Q1 doesn't matter either as it will only make a very small change because we are dealing with small currents (20mA for Q1 C and E)?
Depending on the resistor on Q1 emitter it changes the mA of the Q2 base, how can we know the maximum mA that Q2 base can accept?

 

[Grossel]

If you omit the R2, then Q1 and Q2 will probably die. As the schematic is constructed, R2 needs to be there. Remember that voltage between emitter and base cannot exeed about 0.7 volts. Above that will kill the transistor.

 

[alexalex]

Thanks Grossel, so as far as you can see it all should be fine because R2 is there? Do you think any improvements could be made to this circuit or is it good enough as is?

 

[Grossel]

Only improvement I can think of must be to adjust the resistors values down a bit to increase the efficiency factor if power supply is from batterier.

Also I read a place (cannot remember where) that placing a little cap in parallell with the base resistor on a transistor used as switch may decrease time it takes to turn on. If that would improve this circuit, I don't know. I think somebody have to simulate it to prove it or not.

 

[alexalex]

Cool, thanks. Yep I've done it in a circuit simulator and it works fine.

So lets say I choose this PNP: https://www.st.com/stonline/products/literature/ds/5064/2n6036.pdf
It mentions the Emitter-base voltage as 5V. In Q2 PNP, the base is 11.2V and Emitter is 12V. Does it mean that the max difference between emitter and base can be 5V, so in my case I'm fine? Or as it's emitter-base voltage and I'm using it for collect-emitter, this 5V doesn't matter?

Also the Vce, it says maximum is 3V, is this the maximum loss of voltage between collector and emitter that occurs when at 4A (so my 12V would be 9V) or is it something else?

 

[audioguru]

So lets say I choose this PNP: https://www.st.com/stonline/products/literature/ds/5064/2n6036.pdf

it is not a pNP transistor. It is a PNP darlington transistor that is two transistors.

It mentions the Emitter-base voltage as 5V.

That is its max allowed reverse voltage. The emitter-base will never have a reverse voltage in your circuit.

In Q2 PNP, the base is 11.2V and Emitter is 12V.

No.
A single power transistor will have its base at about 11V. This darlington transistor will have its base at about 10.5V.

Does it mean that the max difference between emitter and base can be 5V, so in my case I'm fine? Or as it's emitter-base voltage and I'm using it for collect-emitter, this 5V doesn't matter?

No.
The darlington transistor begins to conduct a small current when its base is 11V and it conducts 4A when its base is as low as 8V.

Also the Vce, it says maximum is 3V, is this the maximum loss of voltage between collector and emitter that occurs when at 4A (so my 12V would be 9V) or is it something else?

The max Vce is 3V when its collector current is 4A and its base current is 40mA. Every transistor is different, some are better than others. The graph shows that a "typical" transistor has a Vce of 2V.

 

[alexalex]

Thank you for clearing that up.

The darlington transistor begins to conduct a small current when its base is 11V and it conducts 4A when its base is as low as 8V.

Could you advise how you figured this out? Does this apply in my case?

If it does apply in my case, I have the 12v going to the collector but I don't know how to control the base to 8V? I thought the voltage did really matter as long as it's more than 5V? What if the base is at 10.5V but it has 40mA going through it, will it still conduct 4A through the collector and emitter?

Edit: Also is it true that if you are just using the transistor as a switch, you don't need to worry about voltage unless you are using a supply voltage more than 15V?

And when you say

The graph shows that a "typical" transistor has a Vce of 2V.

, this does mean the the 12V going to the emitter will drop to 10V on the collector of the PNP, so really the motor will only get 10V instead of 12V?

 

[audioguru]

Originally Posted by audioguru:
"The darlington transistor begins to conduct a small current when its base is 11V and it conducts 4A when its base is as low as 8V."
Could you advise how you figured this out? Does this apply in my case?

The datasheet has Electrical Characteristics that says the base-emitter saturation voltage is a max of 4V when the collector current is 4A.
Your motor has a stall current higher than 4A so the darlington might have a base-emitter saturation voltage that is higher than 4V.

If it does apply in my case, I have the 12v going to the collector but I don't know how to control the base to 8V? I thought the voltage did really matter as long as it's more than 5V? What if the base is at 10.5V but it has 40mA going through it, will it still conduct 4A through the collector and emitter?

You do not have 12V at the collector. You have 12V at the emitter. The collector voltage is less than the emitter voltage due to the collector-emitter saturation voltage which is a max of 3V when the collector current is 4A and the base current is 40mA.
The transistor controls its base voltage. Some transistors have a lower base-emitter voltage and some transistors have a higher base-emitter voltage.

Edit: Also is it true that if you are just using the transistor as a switch, you don't need to worry about voltage unless you are using a supply voltage more than 15V?

The supply voltage doesn't matter when a transistor is used as a switch and the transistor has a max allowed voltage more than the supply voltage.
The darlington has a max allowed voltage of 80V, not 15V.

And when you say , this does mean the the 12V going to the emitter will drop to 10V on the collector of the PNP, so really the motor will only get 10V instead of 12V?

The PNP darlington has a max collector-emitter saturation voltage of 3V. You also need an NPN darlington that will also have a max collector-emitter saturation voltage of 3V. So the motor might get only 6V. If your darlington transistors are "typical" then the motor will get 8V.
That is why we use Mosfets as switches. An IRF3711Z logic-level N-channel Mosfet has a max saturation voltage loss of only 0.04V when its current is 4A.

 

[alexalex]

Thanks once again for that, it's starting to make sense . Yes I was planning on using a NPN darlington too because I couldn't find a regular transistor that could accept such low base current, I think I will look into Mosfets as you advise.

I finally got around to measuring the amps that the motor takes when spinning and at stall, it's 0.1A and 0.5A. So I believe a 1A transistor should be fine? So lets say I do grab a 1A transistor, would I want to allow the maximum current to pass through it or only what's needed (say 600mA)?

 

[audioguru]

Why does the datasheet for the motor show a stalled current of more than 5A?

 

[alexalex]

Oh, the datasheet isn't of the motor, it's just one that I found which looked similar but the motor I have isn't that powerful at all. I connected the 6v directly to the motor and a multimeter in between to measure the amps. I've just been looking around for transistors but seems like most of them only have a beta of 10 and if I choose higher beta ones then they have a higher Vce. If I were to use a mosfet instead, are there cheap ones where you only need to put in 10mA or so for the base which leads to 1A current to emitter?

 

[mneary]

If I were to use a mosfet instead, are there cheap ones where you only need to put in 10mA or so for the base which leads to 1A current to emitter?

You should look up the difference between MOSFET and BJT. MOSFETs do not have base, emitter, and collector. They use driver circuits that are very different.

 

[audioguru]

Oh, the datasheet isn't of the motor, it's just one that I found which looked similar but the motor I have isn't that powerful at all.

Then I wasted my time explaining the voltage losses of a 4A motor in my circuit.

 

[alexalex]

Then I wasted my time explaining the voltage losses of a 4A motor in my circuit.

Not really, if I was to go with a 1A transistor like https://www.electro-tech-online.com/custompdfs/2010/05/5224.pdf and needed 1A to go through it, it would also have a loss of voltage say 0.8V, so the concept you've taught me still applies. So I would need a one PNP and one NPN of these and a darlington for the bottom NPN https://www.st.com/stonline/products/literature/ds/4126.pdf which has a voltage loss of 1V @ 1A, so total voltage loss would be around 1.8V which isn't that bad but still something to consider.

 

[alexalex]

I've just been looking into transistors a bit more and am wondering about the following NPN transistor graph.

**broken link removed**

Say I needed to allow 500mA for Ic, the saturation needed for Ib is 50mA but what if I were to only supply 20mA to Ib? Would it still allow 500mA to pass through, the only thing is the Vce is a little higher?

 

[audioguru]

The graphs show a typical device. But you cannot buy a typical device, some are but others are much worse but are still passable and are sold.
You don't know if your transistors are typical, better than typical or worse than typical unless you measure them all.

The written spec shows that the worst transistor saturates fairly well when its base current is 1/10th its collector current.
That is how I design transistor circuits so that every passing transistor works perfectly, not just the typical and better ones.

 

[alexalex]

Ah ok, thanks for that.

I've now updated my design as below so I can use it with my Arduino. I won't be using darlingtons rather I'm using 2A rated transistors that only has a Vce of 0.3V max @ 1A which I think it alright.
I've shown two ground connections so you can see my concern, it is acceptable that Q1 effectively loses 89mA back to the batteries ground?
So we are letting through 972mA of current to the motor but also using 89mA for the top transistors and 8mA for the bottom transistors from the battery. Is this excessive to just let 1A through or is this just a fact of using transistors?

**broken link removed**

 

[Grossel]

You still got a problem. You cannot trust that Hfe of Q1 is just 10 times. If that Hfe is t.ex 200 times, the you'll force too much current into emitter on Q2.

And why complicate the circuit by having two connectors to ground from Q1?

 

[alexalex]

Hrm, I don't quite understand? Why can't I trust that the hfe will only be 10 times when I'm saturating it by using as a switch?
I read "Ic = hFE × IB (unless the transistor is full on and saturated)" from here: Transistor Circuits
So it seems to say we can ignore the hFE and only focus on the minimum hFE value which the datasheet says is 10.

The two connectors to ground won't be there in the final circuit, it was just to show the 89mA current going back to the battery to see if anyone thought that was a bit too much.

 

[mneary]

Saturated HFE is shown at 10 in the saturation graphs mostly because it's usually the minimum you'll see when Vce is less than Vbe.. It is not the upper limit. It does NOT apply when the transistor is not saturated.

In your circuit the Vbe of Q2 maintains the Q1 collector voltage at about 11.3V. In that case Q1 isn't saturated at all, and it's driving very heavily and the current in Q1 and Q2 is limited only by the small signal HFE which is often from 200 and upwards. With 10mA of base current in Q1 it will try to have a collector current of 2 or more amperes. And for this, Q1 and Q2 will die.