2 transistor current source

[Omicron]

I have a small question about the current source depicted in figure 4 below:







I've noticed that in the Douglas Self book (the Amplifier Design Handbook) this kind of current source is often used but with an additional 1K resistor in the base path of the lower transistor. What is the purpose of this extra resistor? I built the current source as shown above without this resistor and it seemed to work ok. What am I missing?

[Tube_Dude]

Originally posted by Omicron
I've noticed that in the Douglas Self book (the Amplifier Design Handbook) this kind of current source is often used but with an additional 1K resistor in the base path of the lower transistor. What is the purpose of this extra resistor?

Hi Omicron

The 1k resistor is only used to diminish the risk of oscilation.

[Omicron]

Hi Jorge,

Thanks for the quick reply. Since the 2 transistors form a local feedback loop it makes sense that this thing can oscillate.

Still there is one thing I don't quite understand: when more than one current source is needed an additional transistor is often connected to the node below R1. This new transistor will have it's own equivalent of R2 to set the current. I've seen that these too are given a base resistor of a slightly higher value (2K2). Is this also to prevent oscillation? It seems this extra transistor is not part of any feedback loop so I was wondering how that could oscillate.

Best regards,
Kurt

[Tube_Dude]

Originally posted by Omicron
Hi Jorge,

Still there is one thing I don't quite understand: when more than one current source is needed an additional transistor is often connected to the node below R1. This new transistor will have it's own equivalent of R2 to set the current. I've seen that these too are given a base resistor of a slightly higher value (2K2). Is this also to prevent oscillation? It seems this extra transistor is not part of any feedback loop so I was wondering how that could oscillate.

I don't think that I understand what you are trying to say.

But is a very usual design practice to use a resistor in series with the base of the transistors for enhanced stability.

[Omicron]

Do you know of any guidelines about how to determine the best value for these base resistors? Or are they mostly experimentally determined?

I've just noticed that there is a little snippet of information about this on Douglas Self 's site where he states (in relation to the emittor follower circuit):

"The standard way to stop this is to put a "base-stopper" resistor in series with the base. This should come after the bias resistor to minimise loss of gain. Depending on the circuit conditions, the resistor may be as low as 100 Ohms or as high as 2K."

[Tube_Dude]

Originally posted by Omicron
Do you know of any guidelines about how to determine the best value for these base resistors? Or are they mostly experimentally determined?

These base resistors create a low pass filter in conjunction with the input capacity of the transistor(that depends of the type of the transistor ). In some configurations these capacity is even multiplied by the voltage gain of the transistor (Miller effect ) . With so many variables there are no fixed rule for the value of the base stoppers resistors.

So as Douglas Self say , the must common values fall between 100 and 2.2 KOhms.

[davidsrsb]

At uhf frequencies resistors of almost any value look like about 100R

[peranders]

Omicron, do you ask because you are going to build a power amp?

My advice here is: If nothing is broken don't fix it.

Meaning, make a good pcb and have a decent amount of decoupling caps. If you check my QRV-05, QRV-06 and QRV-08 projects nowhere can these base stopper resistors be seen. Those are very rare if you look around.

Here you can see a very good power amp very similar to my QRV-06/08 and no base stoppers can be seen. It is much about good pcb design. I don't know how much Mr. Self emphasize this. Normally a good pcb is extremely important if you are working with fast circuits with high performance.

[Omicron]

Originally posted by peranders
Omicron, do you ask because you are going to build a power amp?

At the moment I'm just refreshing my electronics knowledge by playing around with some of the basic circuits. Generally just enjoying myself with a soldering iron and a scope. I know I could simulate all this stuff but it's more fun to just play around with some real tools and equipment for a change (I'm a software engineer and already spend too much time behind a computer as it is).

So, I was reading the book by Self and was just wondering where all these resistors came from in his final schematics. I hadn't used them in my experiments and didn't experience any oscillation. But I may just have been lucky. Generally I build my experiments on top of a blank piece of PCB material on which I glue smaller pieces of PCB material to act as "soldering islands". Connections are generally fairly short this way and I use the base PCB material as a ground plane. Anyway it works well for me so far for building small experimental circuits with discrete components.

Very nice projects on your website by the way. I don't yet feel skillfull enough with the soldering iron to try my hand at SMDs however.

[mikeks]

Originally posted by Tube_Dude


Hi Omicron

The 1k resistor is only used to diminish the risk of oscilation.

No...the resistor is used to prevent the control transistor from failing if the current source device fails short circuit...