Hi,

Sounds like you just need a voltage regulator circuit which would be simpler.

 

The beating around the bush is because you haven't shown me what's in the bush.

I don't know what's in the bush either, I just assumed I could throw a jfet in series with the circuit and control it with the gate

Hi,

Sounds like you just need a voltage regulator circuit which would be simpler.

What kind of voltage regulator?

 

I don't know what's in the bush either, I just assumed I could throw a jfet in series with the circuit and control it with the gate
.............................

Fine, but the devil is in the details.
So where in the circuit will the JFET go, how will you control the gate-source (not just gate) voltage, and where will you detect the output to linearize the circuit?
Perhaps this will give you some additional info.

 

Actually that source follower buffer idea of yours is working pretty well in circuits that have a resistive load at the source.
Circuits like this

don't work though. I've spent a few hours trying to find a workaround in that circuit.
It's a CCS that was recommended to me once but I struggle to understand how it works or how it is better than any other CCS.
The output current seems to be based off of the current through R1 which I haven't been able to replicate with transistors or fets.
I'm having a difficult time understanding the purpose of the darlington pair.

 

Here is an LTspice simulation of your constant current circuit.
It uses a Darlington to minimize the error current through the base (which adds to the constant current). If you aren't that concerned about a slight error in the current, then you can use just a single transistor.
The constant-current is equal to the base-emitter voltage of the Darlington or Vbe / R1.
In the simulation Vbe is about 0.96V.
As you can see, the current through the R_Ld resistor is constant at about 7.4mA for a load resistance varying from 1Ω to 1kΩ.

 

Ah okay, thanks for that useful bit of info. I'm looking for precise current control so I think I will stick with the darlington set-up.
Is there a way I can turn this circuit into a VCCS? I haven't had any luck.

 

Although the Darlington configuration minimises error due to base current it does have the disadvantage that Vbe drift with temperature is greater than that for a single transistor.
Replacing R1 with a rheostat would give you a crude VCCS.

 

Isn't a rheostat analog? How am I supposed to control it with voltage?
Also if you have any better ideas for a VCCS I'm open.

 

If you want a voltage controlled, accurate VCCS, then you can add an op amp.
Here's the LTspice simulation of such a circuit.
The current equals Vin / R3.
As you can see the current is constant for load from 0.1 ohm to 500 ohms.
The upper load resistance limit is determined by the supply voltage.

 

Hmmm, since I'm going to be displaying active current on my display anyway perhaps I can just use a current sense resistor on the output of the current source and then use digital feedback instead of the opamp? Are there any advantages of transisfor vs jfet vs mosfet for a CCS?

 

Hmmm, since I'm going to be displaying active current on my display anyway perhaps I can just use a current sense resistor on the output of the current source and then use digital feedback instead of the opamp? Are there any advantages of transisfor vs jfet vs mosfet for a CCS?

Digital feedback certainly is a possibility.

JFETs and MOSFETs have zero gate current so they don't affect the accuracy of the setting.
BJTs have a base current that can affect accuracy (perhaps giving a error of 1% or so) of the constant-current value.
MOSFETs are generally enhancement-mode and JFETs are depletion mode devices, so that would affect the bias voltages for normal circuit operation.
Generally MOSFETs have lower ON resistance and are more common than JFETs so I would go with a MOSFET.

 

Sweet, looks like I got it all figured out then, thanks