And what's the basic for your sweeping statement?

He needs to make the circuit in post#10, only Dick added current limit.

 

Still at around 200mA.?
Any suggestions that will make this circuit work

But I can't trouble shoot with out knowing the voltages.

What are the voltages?

What is the input voltage?

I have asked repeatedly for information that would tell me what you did wrong. It works here. Does not work for you. The design works, the making of it, does not.
Ok I am out of this one sided conversation.
Good luck.

 

I have asked repeatedly for information that would tell me what you did wrong. It works here. Does not work for you. The design works, the making of it, does not.
Ok I am out of this one sided conversation.
Good luck.

Sorry, I've not been getting notifications that you were even replying, and I didn't know this thread went onto another page until now. The input voltage is 12.5V. Getting 5V out! Around 250mA.

 

Sorry, I've not been getting notifications that you were even replying, and I didn't know this thread went onto another page until now. The input voltage is 12.5V. Getting 5V out! Around 250mA.

5V @ 250mA implies a load resistor of 20 ohms. What happens when you connect, say, a 10 ohm load? What voltage do you get then?

 

5V @ 250mA implies a load resistor of 20 ohms. What happens when you connect, say, a 10 ohm load? What voltage do you get then?

About 4.5V

 

About 4.5V

The one I built yesterday is still at 5V with 10 ohm load.

 

The one I built yesterday is still at 5V with 10 ohm load.

So what voltage do you get with yesterday's one and a 5 ohm load?

 

He needs to make the circuit in post#10, only Dick added current limit.

I have no problem trying that circuit. Though I have to build it with what I have. The only npn's I have are tiny surface mount L6's, and I have assorted n-channel fet's. All have similar characteristics as the one's I've been using so far. That will have to be used for the pass Q. I may need assistance coming up with the resistive values if you would be so kind? Here are the part numbers I will use.
NPn: 2SC1623-L6
MOSFET: K15A60U

 

So what voltage do you get with yesterday's one and a 5 ohm load?

Made a mistake on the last measurement, I'm using a homemade volt meter that is not so great. I should make my own. it is 4 volts at 10 ohms, way less at 5 ohm. I'm in a precarious environment. I don't have access to regular instrumentation. If you must know, I'll give you an encrypted e-mail address. I can't have this info in public.

 

You really should consider some current limiting for your circuit -there is no "intentional" current limiting in the circuit, so you have no control of how much current this circuit will try to shove into a battery if you connect a dead battery to it.

View attachment 189575
This circuit may look familiar. When the current through R2 develops enough voltage across the base-emitter of Q2 to make it draw current, the transistor will begin conducting, starving the base drive of pass transistor Q1 thus limiting the doesn't to the load. This works with MOSFETs too.

How do I calculate the R2? I also see the 100 ohm that would be sinking 50mA on top of it.
So here's a question. These lithium ion batteries already have chips in them for limiting current. The one in question has a limiter of 500mA via resistor calculated from 5V. It does not however regulate voltage. This much I do know. But the active electronics in the small package does a bit more. Like cut the output power altogether when it gets low to save batter life. Allowed a trickle charge when first plugged in for a charge to prevent an explosion. I was surprised to find all this out.
I'm guessing that without the current suppression, I would eliminate R2 and R3? Or is R3 still serving another purpose?
I still want to know how to use this circuit as it is because I can think of 100 uses off the top of my head for a regulator that features this. I like it. But is it going to allow my fet to deliver an amp? So far the other differential regulator has not gotten past 250mA with fet's.

 

To start with, I only recommend the current limiting technique, not necessarily this circuit. It has the slightly unpleasant aspect not having and voltage gain in the output stage, the same as with your circuit with the N-Channel MOSFET. I think you would be having fewer problems if you used a P-channel MOSFET or a PNP transistor. The current limit circuit would be the same idea for transistors of the opposite polarity.

Back to your questions:

The value of R2 is approximately 0.6V/current for the current limit point. It will vary with temperature but since the function is only there to protect things you should not see it operate under normal operation.

The 100 ohm resistor, R3 represents the load and you don't need it.

"But is it going to allow my fet to deliver an amp?"
If you use a 0.62 ohm resistor you can get an amp.

You can make a low value resistor by winding magnet wire over a form. The temperature coefficient of the copper will very closely compensate for the base-emitter drift of Q2 (interesting but not important) For example, #32 copper wire is 0.532 ohms per meter so you would need 0.62/0.532 = 1.17 meters of #32. Other wire sizes may be used.

As an aside, you can buy a 1 or 1.5 amp voltage regulator that has excellent regulation, over-current and over-temperature protection integrated integrated circuit very inexpensively. You won't learn as much about analog circuits if you use it as you are learning makign this one, but I want to make sure you are aware of these parts and how simply they work:
https://www.electronicshub.org/understanding-7805-ic-voltage-regulator/

 

To start with, I only recommend the current limiting technique, not necessarily this circuit. It has the slightly unpleasant aspect not having and voltage gain in the output stage, the same as with your circuit with the N-Channel MOSFET. I think you would be having fewer problems if you used a P-channel MOSFET or a PNP transistor. The current limit circuit would be the same idea for transistors of the opposite polarity.

Back to your questions:

The value of R2 is approximately 0.6V/current for the current limit point. It will vary with temperature but since the function is only there to protect things you should not see it operate under normal operation.

The 100 ohm resistor, R3 represents the load and you don't need it.

"But is it going to allow my fet to deliver an amp?"
If you use a 0.62 ohm resistor you can get an amp.
View attachment 189711
You can make a low value resistor by winding magnet wire over a form. The temperature coefficient of the copper will very closely compensate for the base-emitter drift of Q2 (interesting but not important) For example, #32 copper wire is 0.532 ohms per meter so you would need 0.62/0.532 = 1.17 meters of #32. Other wire sizes may be used.

As an aside, you can buy a 1 or 1.5 amp voltage regulator that has excellent regulation, over-current and over-temperature protection integrated integrated circuit very inexpensively. You won't learn as much about analog circuits if you use it as you are learning makign this one, but I want to make sure you are aware of these parts and how simply they work:
https://www.electronicshub.org/understanding-7805-ic-voltage-regulator/

Thank you for that! I do have a p-channel mosfet . It's the first one I tried, but I haven't tried it with the new values. Though I will today! Lol.
Thanks again.

 

To start with, I only recommend the current limiting technique, not necessarily this circuit. It has the slightly unpleasant aspect not having and voltage gain in the output stage, the same as with your circuit with the N-Channel MOSFET. I think you would be having fewer problems if you used a P-channel MOSFET or a PNP transistor. The current limit circuit would be the same idea for transistors of the opposite polarity.

Back to your questions:

The value of R2 is approximately 0.6V/current for the current limit point. It will vary with temperature but since the function is only there to protect things you should not see it operate under normal operation.

The 100 ohm resistor, R3 represents the load and you don't need it.

"But is it going to allow my fet to deliver an amp?"
If you use a 0.62 ohm resistor you can get an amp.
View attachment 189711
You can make a low value resistor by winding magnet wire over a form. The temperature coefficient of the copper will very closely compensate for the base-emitter drift of Q2 (interesting but not important) For example, #32 copper wire is 0.532 ohms per meter so you would need 0.62/0.532 = 1.17 meters of #32. Other wire sizes may be used.

As an aside, you can buy a 1 or 1.5 amp voltage regulator that has excellent regulation, over-current and over-temperature protection integrated integrated circuit very inexpensively. You won't learn as much about analog circuits if you use it as you are learning makign this one, but I want to make sure you are aware of these parts and how simply they work:
https://www.electronicshub.org/understanding-7805-ic-voltage-regulator/

Believe me. If I could go to the store, I would use an IC any day. I'm very familiar with them. Especially the one you mentioned, but I can't get there. I'm in a very unique situation. Picture me on a deserted island with nothing but a ton of n-channel fet's and an unlimited supply of L-6 npn transistors. I do have one p-channel. That one was the same result as the n-channel in the first attempt. The only difference was that I was using 25k at the emitters, and a 5V Zener instead of the voltage dividers.
Now knowing I can't get anywhere, do you know a way I can get 5V at 1 amp using what I have? My 12.5V power supply is very stable. I at least have that.

 

Been in that situation. A day's travel to the nearest parts store and only a dozen kinds of transistors and a few ICs.

How is your supply of higher current NPN transsitors?

 

Been in that situation. A day's travel to the nearest parts store and only a dozen kinds of transistors and a few ICs.

How is your supply of higher current NPN transsitors?

None. All L6's. All I have are fets for higher power. Mostly assorted n-channel. The funny thing is, I ran a sim and it works! Don't know why it's not for real? Do you have any better ideas? Your knowledge far exceeds mine.

 

If you compared all the node voltages of your real circuit with the simulation values, you might be able to determine the problem.

 

In many cases the gate-to-source voltage versus drain current varies considerably among MOSFETs with the same part number.

 

6=

If you compared all the node voltages of your real circuit with the simulation values, you might be able to determine the problem.

would this be any different?

 

In many cases the gate-to-source voltage versus drain current varies considerably among MOSFETs with the same part number.

Look at what I sent Crutschow. Would that help?

 

6=

would this be any different? View attachment 190002

How bout this also?