Simple Bench Power Supply 10A

[KeepItSimpleStupid]

You want to look at the voltage ACROSS the emitter resistors. This is directly related to the current each is passing. The schematic says 0.2 ohms (0R2).

 

[ortiko]

I used 0.22 ohms...
Maybe this causes the problem ?

 

[KeepItSimpleStupid]

No. If you don't have a transistor checker, you could make a jig to measure Hfe. You need to essentially measure the base and collector current. e.g. #2 here: https://www.technologystudent.com/elec1/tranfrm1.htm

The gains really need to be within 5-10% of each other.

For you MJ15* parts, Datasheet: https://www.google.com/url?sa=t&rct...sk9xerzDBN8s04w&bvm=bv.77880786,d.aWw&cad=rja

Hfe can vary from 25 to 150. or 6x. That is a pretty substantial range.

 

[ronsimpson]

You changed transistors and there was no change.
One emitter is 0.4 volts higher.
Please exchange the two emitter resistors and see if the other transistor gets hot. The problem might be in the resistors.

 

[MrAl]

I was just reading the datasheet for the LM350.... There is a cracking design for a 0 ~ 10 amp bench supply using 4 in parallel.

Watching with interest...

Hi Ian,

I am not sure if i trust that design. The main reason is that they have the power supply pin of the op amp connected to the input voltage minus a small voltage drop, and they have the same node connected to the non inverting input of the op amp. Is that a rail to rail input op amp? I dont think so, so the design might not work.
You could check on this if you like to see if it really is or not.

 

[ortiko]

The problem of one transistor draining all the current is resolved and as in most cases it was very simple...

Stupid me forgot to soldier the junction between the emitter and resistor

Now the temp is equall and both are running more cooler !

Transistor 1

Transistor2

And runing with lamp load...



Very happy that it is working as supposed to

 

[ortiko]

Changed planes for this PS and will use two transformers for non heating issues.

I have purchased a 12V 10A that will be used for low voltage and have kept the 24V 5A in the case for 15 Volts to 30 Volts
Tested the 12V transformer PS under 4A load and it dosent get hot at all.


The internal of the case will be like this:





The transformer selection can be done with a simple ON-ON switch but i want to use a 10A relay that i have for the transformer to be selected automatically by the voltage change from potentiometer

Problem is that i can't figure how to switch on the relay when 15V are selected from pot.
My relay is 24 Volts and it changes state from 12V...

Simple Switch connections:

Maybe if i use a transistor as a switch will do the job to activate relay at 15V dc ?
something like this :

 

[KeepItSimpleStupid]

!5 is probably a bad number. You might also make sure you have some hysteresis. e.g In at 18, out at16 V. or whatever, so you may want a window comparator.

 

[ronv]

I would stick with the high voltage low voltage switch. Sensing the voltage and switching the input to the regulator may be heartache.

 

[JimB]

I agree that automatic switching should have some hysterysis in the switching voltage.

Also, I think that you may find the auto switching a bit of a pain at times, I cannot fully justify this statement, just a gut feel that I have.

JimB

 

[ortiko]

I managed to solve the issue of Voltage adjustment by replacing the potentiometer with a
multi turn:

Now the voltafe adjustment is Very Very smooth
I like this pot alot...

 

[ortiko]

!5 is probably a bad number. You might also make sure you have some hysteresis. e.g In at 18, out at16 V. or whatever, so you may want a window comparator.

I found in my bench drawers the TL072 and planning to use it as voltage comparator...

will do a research for how it could be done.
Any ideas or suggestions are of course welcome since i have never used an opamp as a voltage comparator

 

[MrAl]

Hi,

Is that a series 3590 pot? I use the same one for my PS and i get a nice adjustment because it is ten turns.

If you look up the TL072 and similar IC's you can get a better idea what to expect. They might be faster than the typical LM324 for example, but they might also require stricter input/output/V+/V- relationships.

 

[Nigel Goodwin]

I agree that automatic switching should have some hysterysis in the switching voltage.

Also, I think that you may find the auto switching a bit of a pain at times, I cannot fully justify this statement, just a gut feel that I have.

On the contrary, I would imagine it should be simple and flawless in operation - as you say, it needs hysteresis - so apply positive feedback to the comparator that's doing the switching.

You want a fairly high degree of hysteresis, so (for example) that it switches to the high winding if over 25V, and switches to the low winding if it's under 20V.

Personally, I'd probably use a PIC to do - and use the PIC to display the output voltage and current on an LCD display - as you're monitoring the output voltage to display it it's simple to switch transformer windings (using relays), you could even switch multiple taps easily as well.

 

[ortiko]

Well i picked myself the LM 393 and will start the experiments this afternoon

For reference voltage i will use this switching 5V 2A little power supply that will be used also for the usb supply in front panel of the case

 

[ortiko]

Sorry i did not find the time to continue the project

I have in mind something like the below circuit.
Have never done this before with lm393.

 

[Nigel Goodwin]

You need a resistor from the output back to +ve, to give hysteresis.

 

[ortiko]

I have been trying to simulate the circuit to LTspice.
The Resistors values are for reference
.






I don't have any expirience with ltspice and althought i managed to upload LM393 from this link https://www.ti.com/product/lm393 and run the simulator of this circuit i don't know if this is ok

To upload LM393 i did the below from this very helpful post:

https://forum.allaboutcircuits.com/threads/importing-models-into-ltspice.36456/


1. On the link you posted, scroll down to the PSpice model, unzip the folder, and open LM339_5.1 with notepad. Save the file in C:/program files/LTC/LTspiceIV/lib/sub as LM339.sub. Change "save as type" to "All files".

2. If LTspice is already open, close and then reopen it.
3. Open a new schematic window (Leftmost icon on toolbar).
4.Click on the component icon (the AND gate on the toolbar).
5. Double-click on [Opamps].
6. Scroll all the way to the end and select opamp2.
7. Click OK.
8. Left-click to place opamp2 symbol on schematic.
9. Right-click on symbol to open Component Attribute Editor.
10. Left-click on Value.
11. In the edit window that says Value = opamp2, change opamp2 to LM339. In general, the value you enter here must be identical to the subcircuit name in the subcircuit file. In this case, that line reads
.SUBCKT LM339 1 2 3 4 5
12. Left-click on the .op icon (rightmost on the toolbar). This is the spice directive icon .
13. Type .lib LM339.sub in the window. Left-click on OK.
14. Place this spice directive on the schematic by dragging it to where you want to place it, and then left-clicking.
15. You are now ready to place and connect other components before simulating.
Can someone more expirenced help ?

 

[ortiko]

You need a resistor from the output back to +ve, to give hysteresis.

In my LTspice Circuit R4 is for hysteresis if i am not wrong

 

[audioguru]

Many LM393 comparators cannot be used in your circuit because its minimum output low current is only 6mA. Your circuit has an output current of 14.3mA.
It is a dual comparator. Do you know how to disable the unused comparator?