I'm working on designing a voltage controlled filter, and in order to control some other components, I designed a little transistor circuit like this:



and so far, I'm actually pretty impressed with how it works. The control over the output current seems pretty linear, and the transistor is biased somewhat okay. From what I've played around with, I've found that changing the resistor just above the collector will change the overall sensitivity, changing the emitter resistor changes the minimum/maximum output current, and changing the base resistor does pretty much the same thing as changing the emitter resistor. You can play around with it on Falstad's circuit simulator.

I could use some criticism on it from more experienced engineers. Also, if this circuit's been made before by someone else, that would be cool to know.

 

I did a few calculations on the circuit, and using the old \( I_E = I_B + I_C \) relationship, and changing it to \( I_B = I_E - I_C \), I got

$$ \frac{V_{cc}}{R_1 + R_3} = \frac{V_C}{R_4} - \frac{V_{cc}}{R_1 + R_2} \,,$$
or just
$$ I_B = \frac{V_C}{R_4} - \frac{V_{cc}}{R_1 + R_2} \,.$$

For large signals, and with a transistor in the active region, this should generally hold true.

 

I'm working on designing a voltage controlled filter, and in order to control some other components, I designed a little transistor circuit like this:

View attachment 199691

and so far, I'm actually pretty impressed with how it works. The control over the output current seems pretty linear, and the transistor is biased somewhat okay. From what I've played around with, I've found that changing the resistor just above the collector will change the overall sensitivity, changing the emitter resistor changes the minimum/maximum output current, and changing the base resistor does pretty much the same thing as changing the emitter resistor. You can play around with it on Falstad's circuit simulator.

I could use some criticism on it from more experienced engineers. Also, if this circuit's been made before by someone else, that would be cool to know.

It's not clear to me what this circuit is meant to accomplish, which parts vary in the real world application, what the desired output is, etc. Hard to critique or comment without more context.

In the real world application, you're wanting to start with a variable voltage signal and output a variable current in proportion to the voltage - is that right? And in this case, the 15V represents the input voltage? Or is that the power supply, which remains constant, while something else is varied?

If you could explain the intent, the input and output ranges, whether they're meant to be linear, logarithmic, etc, and which nodes are meant to be inputs and outputs, we might be able to comment better. Maybe I'm just dense, but so far I have no idea what you've got here!

 

Your circuit doesn't do anything.

 

It's not clear to me what this circuit is meant to accomplish, which parts vary in the real world application, what the desired output is, etc. Hard to critique or comment without more context.

In the real world application, you're wanting to start with a variable voltage signal and output a variable current in proportion to the voltage - is that right? And in this case, the 15V represents the input voltage? Or is that the power supply, which remains constant, while something else is varied?

If you could explain the intent, the input and output ranges, whether they're meant to be linear, logarithmic, etc, and which nodes are meant to be inputs and outputs, we might be able to comment better. Maybe I'm just dense, but so far I have no idea what you've got here!

It's made to be a sort of voltage controlled current supply in order to control some other transistors that will act as current controlled resistors (like in the Korg MS-20's filter, which uses a Korg35 IC). The 15 volt power supply is a constant voltage, and the only voltage that's supposed to change is the lower one (it's 0 V in the picture, and it's labeled Vc in the simulator). The output current should be linearly controlled by the control voltage to allow for easier control by musicians (i.e. myself or some poor schmuck that uses it).

I don't have any specific output ranges in mind, since they'll need to be determined to when I get to working on the transistor circuit they're controlling. The input ranges, however, are basically just for standard Eurorack control voltage ranges (which don't have a true standard, but I would say in the range of about ±5 V).

 

The standard voltage-controlled current source (as below) uses fewer components. Does your circuit have any particular advantage?

 

The output of your circuit connects to ground so it always has an output of ZERO!

 

Count me as another one who also does not understand what the circuit is supposed to accomplish.
Is the 22.9 mA the control input or the desired output?

 

It's made to be a sort of voltage controlled current supply in order to control some other transistors that will act as current controlled resistors (like in the Korg MS-20's filter, which uses a Korg35 IC). The 15 volt power supply is a constant voltage, and the only voltage that's supposed to change is the lower one (it's 0 V in the picture, and it's labeled Vc in the simulator). The output current should be linearly controlled by the control voltage to allow for easier control by musicians (i.e. myself or some poor schmuck that uses it).

I don't have any specific output ranges in mind, since they'll need to be determined to when I get to working on the transistor circuit they're controlling. The input ranges, however, are basically just for standard Eurorack control voltage ranges (which don't have a true standard, but I would say in the range of about ±5 V).

Ok, I ran some sims based on your description to see if the circuit accomplished more than I thought, and here's my take on it.

1. Your transistor isn't really doing anything useful.
2. Your output current is highly dependent on the input impedance of the receiving circuit.
3. This is essentially a voltage divider with a variable voltage as the lower input voltage.

My first sim took your circuit as-is, and only added variable input and output impedances to see how consistent your circuit would be with unknown I/O parameters. It turns out it's less affected by the source impedance of the voltage signal than I expected (at least within what I assumed would be reasonable ranges for your modular analog synth environment.) However, the load impedance of whatever device you're sending this current signal to has a large impact on the actual current delivered. Your circuit isn't really delivering a controlled current - it's delivering a scaled voltage, and the resulting current depends on the relationship between your 100ohm output resistor and the resistance/impedance on the receiving end.

My second sim demonstrates how little the transistor is doing in your circuit. It doesn't change the slopes of the I-V graphs, nor does it have any effect on the importance of load impedance. It just shifts the lines slightly and adds an upper limit to the output (although that limit is also dependent on load impedance.)

There are surely simpler and more effective ways to accomplish this, including @Alec_t 's post above. I believe his solution is dependent on transistor beta and would probably require manual trimming, possibly also with some temperature dependence. I may be wrong about that. More robust solutions that require less calibration might rely on op amps.

In the images below, the multiple slopes represent different input and output impedance parameters:

 

I believe his solution is dependent on transistor beta and would probably require manual trimming, possibly also with some temperature dependence. I may be wrong about that.

You're not wrong on either count. If a stable controlled current is required then an op-amp and an accurate voltage reference would probably figure in the solution. If the current source is to be used in a music synth then a very stable current will presumably be needed.

 

The standard voltage-controlled current source (as below) uses fewer components. Does your circuit have any particular advantage?
View attachment 199760

To be quite honest, I don't think there is any advantage to my circuit. If the voltage controlled current source you're showing has fewer components, and has a fairly linear output, and has already been proven to work, chances are it's better than my circuit.