.....................
I see, that must be the problem i had with the voltage not being enough for powering a regulator (since the output voltage will be the set voltage of the opamp, in this case 0 - 3v. What do you recommend i do to make it not depend on the load resistance? Or so that i get enough voltage to power a regulator. Regards.

That's what the circuit I showed in post #15 did.

Below is your circuit of post #35 inverted to drive a grounded load.
It requires a rail-rail type op amp to work at the positive rail.
It uses a TL431a 2.5V reference to supply the limit control voltage (0 to minus 0.3V @ op amp input referenced to the positive rail).
You can substitute a PNP darlington stage for the P-MOSFET if you like.

 

That's what the circuit I showed in post #15 did.

Below is your circuit of post #35 inverted to drive a grounded load.
It requires a rail-rail type op amp to work at the positive rail.
It uses a TL431a 2.5V reference to supply the limit control voltage (0 to minus 0.3V @ op amp input referenced to the positive rail).
You can substitute a PNP darlington stage for the P-MOSFET if you like.

View attachment 109681

Thank you once again! I've decided to use the circuit you posted in #15!

Maybe I've missed something, Tried to simulate the #15 circuit. But i still don't get enough voltage for the regulator. It does work but i do not get the "full" voltage out.

I replaced the zener with a TL431, never used one before so i might have it wired wrong. Excuse the wiring chaos..

 

Yes the TL431 is wired incorrectly (did you look at the data sheet?).
Connect the control pin to the cathode (top) of the device, not the pot output, which will cause it to drop a constant 2.5V (like a zener).

Also reduce R2 to about 67kΩ.

RE the wiring chaos -- Please take the time to do a clean schematic and spread it out so the designations don't overlap.
It is well worth the effort so one can more easily see how the circuit is connected and help detect any wiring errors.

 

Yes the TL431 is wired incorrectly (did you look at the data sheet?).
Connect the control pin to the cathode (top) of the device, not the pot output, which will cause it to drop a constant 2.5V (like a zener).

Also reduce R2 to about 67kΩ.

RE the wiring chaos -- Please take the time to do a clean schematic and spread it out so the designations don't overlap.
It is well worth the effort so one can more easily see how the circuit is connected and help detect any wiring errors.

Yes i have spread out all the labels etc so you can see it better now. I think i've wired it correctly. But the voltage is the same. Atm it won't go above 1.3 volts..

Update: when i change the pot R8 from 1k to around 200k the voltage go up to about 10.7 volts (like it should), if i change it to 100k the maximum voltage goes down do about 6 volts.. any idea why it needs to be such a high value? Thanks.

Update once again: Changed regulator from LM317 to a LM723 and now everything works fine!

 

You have U3 connected incorrectly is the reason it won't work properly.
You need a 120Ω resistor (you show 240Ω in your previous schematic) from the output to the Adj pin to be able to control the output voltage.

Incidentally, to clean up the schematic you should also eliminate excess wire lengths and jogs, such as those around the op amp. It still looks rather ratty.

 

You have U3 connected incorrectly is the reason it won't work properly.
You need a 120Ω resistor (you show 240Ω in your previous schematic) from the output to the Adj pin to be able to control the output voltage.

Incidentally, to clean up the schematic you should also eliminate excess wire lengths and jogs, such as those around the op amp. It still looks rather ratty.

Thanks i will do that too, oooh 120Ω solved it! I might as well ask instead of making a new thread. Everything that i wanted to know about current limiting is now solved thanks to you and a few others!

The last thing i wanna learn is how you calculate what value the pot/resistor should be for a good adjustment. This LM723 that iam now using to adjust the voltage from 2.8 to 16v. No matter how i change the resistor/pot to different values i can't seem to get the voltage to adjust directly (i need to move the pot maybe 30% before the voltage actually changes). If you look at this picture:

I have the same setup with the voltage adjustment, i wanna know how to figure out the R9 and R10 values.

If you have time could you explain how to calculate this? Not only for this regulator but for others too (if it is the same process for every regulator) Regards!

 

I don't like the 723 which is an old device, doesn't readily give a wide adjustment voltage range, and is somewhat messy to use.

Try increasing the value of R9 to 1kΩ or more to see if that helps reduce the dead-band of pot R10.

As far a calculating the needed value for R9, that's a complex relation between the desired voltage range and the value of R7 and R8, which I can't readily explain in a reasonable number of words.

The voltage adjustment scheme for each regulator type is different so you (gasp) need to look at the data sheet for the particular regulator you are using.

The basic scheme for most regulators is a feedback scheme where a reference voltage is compared with the output voltage (often through an adjustable voltage divider to allow output voltage adjustment) using a differential (op amp type) circuit.
The internal series pass element is adjusted by the differential output to keep the output voltage (as determined the divider) equal to the reference voltage.
Below is a basic block of such a circuit.

 

I don't like the 723 which is an old device, doesn't readily give a wide adjustment voltage range, and is somewhat messy to use.

Try increasing the value of R9 to 1kΩ or more to see if that helps reduce the dead-band of pot R10.

As far a calculating the needed value for R9, that's a complex relation between the desired voltage range and the value of R7 and R8, which I can't readily explain in a reasonable number of words.

The voltage adjustment scheme for each regulator type is different so you (gasp) need to look at the data sheet for the particular regulator you are using.

The basic scheme for most regulators is a feedback scheme where a reference voltage is compared with the output voltage (often through an adjustable voltage divider to allow output voltage adjustment) using a differential (op amp type) circuit.
The internal series pass element is adjusted by the differential output to keep the output voltage (as determined the divider) equal to the reference voltage.
Below is a basic block of such a circuit.

I see thank you! Well, which voltage regulator do you recommend? I only know of the LMxxx and the LT3080, and of course the LM2678 etc for switching supplies but i was going for a linear one (not a requirement tho). The LT3080 looks awesome but it is pretty expensive.. do you think it's worth the money?

 

The LT3080 is great if you want a low dropout regulator that can be adjusted down to 0V output.
Don't know if there's anything cheaper available that can do that.

But if you don't need the low dropout and adjustment down to 0V, then the cheap old standby LM317 is commonly used.
It can be adjusted down to 1.2V output with two resistors, is nearly bulletproof (current and over temperature protected) but requires a minimum of about 3V across it for proper operation.

So which you use depends upon your requirements and budget.

Below is an interesting variation on the LT3080 circuit to minimize the dropout voltage when it is operated from a bridge rectifier.
It uses an added diode and capacitor to supply voltage with low ripple to the Vcntrl input.
This allows the minimum difference (dropout voltage) between the low point of the main ripple voltage value and the output voltage to be 0.5V minimum, compared to 1.5V if Vcntrl were powered directly from the bridge output.
That gives you about one extra volt of headroom between the minimum ripple voltage and the output voltage.

This can be seen in the LTspice simulation below for an output voltage close to the bridge output voltage.
Out1, which has the separate Vcntrl input voltage, has a smooth no ripple output, whereas Out2, which has the Vcntrl pin connected to the bridge output, has ripple in its output, indicating the regulator is dropping out.

The advantage of this is that it can allow you to use a 1V lower transformer voltage for a given output voltage, or it allows you 1V more ripple, which means a smaller required filter capacitor.

 

The LT3080 is great if you want a low dropout regulator that can be adjusted down to 0V output.
Don't know if there's anything cheaper available that can do that.

But if you don't need the low dropout and adjustment down to 0V, then the cheap old standby LM317 is commonly used.
It can be adjusted down to 1.2V output with two resistors, is nearly bulletproof (current and over temperature protected) but requires a minimum of about 3V across it for proper operation.

So which you use depends upon your requirements and budget.

Below is an interesting variation on the LT3080 circuit to minimize the dropout voltage when it is operated from a bridge rectifier.
It uses an added diode and capacitor to supply voltage with low ripple to the Vcntrl input.
This allows the minimum difference (dropout voltage) between the low point of the main ripple voltage value and the output voltage to be 0.5V minimum, compared to 1.5V if Vcntrl were powered directly from the bridge output.
That gives you about one extra volt of headroom between the minimum ripple voltage and the output voltage.

This can be seen in the LTspice simulation below for an output voltage close to the bridge output voltage.
Out1, which has the separate Vcntrl input voltage, has a smooth no ripple output, whereas Out2, which has the Vcntrl pin connected to the bridge output, has ripple in its output, indicating the regulator is dropping out.

The advantage of this is that it can allow you to use a 1V lower transformer voltage for a given output voltage, or it allows you 1V more ripple, which means a smaller required filter capacitor.

View attachment 109721

Thank you once again for all the detailed answers. Might go for the LT3080 then, this is my first "real" power supply build. With current limiting etc. before i have just built a simple LM317 psu with only voltage adjust. That's why I'm having a hard time choosing the right regulator hehe

 

this is my first "real" power supply build.

Audio amps aside, the most recurrent for hobbyists. Count me in the list, for many.

 

So upon building the circuit in #44. i noticed that i do not own a single P-channel mosfet. I only have a lot of N-channel ones. Is it possible to use an N-channel instead? if so what would i need to change? Thanks!

 

Sorry, nope.
You need a P-MOSFET.

 

Sorry, nope.
You need a P-MOSFET.

Got it, i will order a few then! Thank you for the answer.