• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Rebuilding My Power Supply

Status
Not open for further replies.

MrAl

Well-Known Member
Most Helpful Member
Hello again,


ronv:
Yes that's the problem i found too when going to build a hefty power supply of maybe 30 amps output.
The inductor becomes expensive. Maybe the only way around this is to build several smaller units
and use a paralleling scheme to get a higher output.
An air core inductor sounds interesting, but i have to wonder what kind of efficiency we would get
out of it. Lucky the LM2576 works at around 50kHz, so it's not too high. I guess we'd really
have to try it, and that means winding a large coil with heavy wire for say 20 amps out. Number
12 AWG wire for example. Might be lots of turns however. With a core we dont need that many turns
so it works out a little better, but then again we'd have to get a core too then.
Perhaps you have some other ideas too you can add as we try to think of ways to do this without
spending a fortune.

kinarfi:
I am posting my first thoughts on this project of using the LM2576 to drive a PMOS and thus obtain more
output current and still have the nice control of the LM2576 to look forward to. Keep in mind this
is like a preliminary drawing that i am using to illustrate the basic idea, although from your text
you seem to have a good grasp of this already.
Are you saying you tried something like this already? Note that i used an inverter too but the resistors
that load the internal drive transistor would be small enough to draw some significant current, maybe
at least 100ma or more, even 250ma perhaps, im not sure yet. The LM2576 has to think it is working
as usual. If that means connecting a small inductor too then so be it, if that's what it takes.
Alternately we could look for a controller IC and go from there.

I've also looked at inductors in parallel, but for some decent inductance like 100uH it takes quite
a few inductors. To double the current rating of a single 100uH inductor, it takes a total of
four inductors of 100uH to get the same inductance. So if we had four 100uH 5 amp inductors, we could
make ONE single 100uH inductor that had a current rating of 10 amps (that's twice one inductor rating).
That still might get a little expensive. Of course if we had two 200uH inductors at 5 amps we could do
it with two inductors, but they would be more expensive than the 100uH inductors in some cases.
So maybe a parallel setup would be better where we build several smaller units and parallel them. We'd
have to make sure the complete circuit is stable then and solve any problems that might come up with that.

Take a look at this schematic and see if that is what you tried or not. Let me know so we can figure out what to do next.
 

Attachments

ronv

Well-Known Member
Most Helpful Member
Hmmm. No new ideas yet. The problem gets worse (I think) when you add the 4 corners of a variable supply.
 

kinarfi

Well-Known Member
So far, my experience is that the higher the current, the lower the henreys, Xl= 2PiFL and at a 100uF, I doubt you'd get 5 amps,
I'm just trying for 20, 10 stable and smooth & steady and 15 volts would be great, I have seen 20 on my meter, but I'm thinking it has lots of ripple, I'll try 100 ohms for both base resistors and 1k for RC and 33uH and 4905 and let you know, may try an inductor in the base, we'll see.

I was wrong, using a ~300uh and it does work better than a 33uh or 47uh
 
Last edited:

MrAl

Well-Known Member
Most Helpful Member
Hi,

ronv:
What do you mean by "4 corners"?

kinarfi:
So you dont want to try the LM2576 circuit i posted then?
 

kinarfi

Well-Known Member
Hi,
kinarfi:
So you dont want to try the LM2576 circuit i posted then?
I haven't tried that design yet, I will today, but you didn't say what values to say what values to use, that what I was saying, Rb=100 Rc=1K
 

Attachments

kinarfi

Well-Known Member
couldn't find a model for LM2576, so I searched Spice and found the following and modified it and I appears your design will work
I'll let you know this evening, 6-11 hours,
 

Attachments

MrAl

Well-Known Member
Most Helpful Member
Hi,

Do you have the models for the MC device and the MOSFET? I cant view the file without them.
From what i can see though, the lower resistor on the NPN is missing. That's very important too.
 

kinarfi

Well-Known Member
Hi,

Do you have the models for the MC device and the MOSFET? I cant view the file without them.
From what i can see though, the lower resistor on the NPN is missing. That's very important too.
The model came from a search of spice for "buck" C:\A1 Spice\examples\LtSpicePlus\Smps\OnSemi\
change .txt to .lib on mc340683 .electro-tech-online.com won't let me load it a .lib.
I ran into the mc34063a while writing this, it's pretty good and sims too. but it doesn't show any switching.
the two screen shots are labelled
 

Attachments

MrAl

Well-Known Member
Most Helpful Member
Hi,

Ok thanks i'll check that out.

I did look at the schematic but i see it is still not quite right for the LM2576 just so you konw. Might be right for the MC chip though.
 

kinarfi

Well-Known Member
Hi,

Ok thanks i'll check that out.

I did look at the schematic but i see it is still not quite right for the LM2576 just so you konw. Might be right for the MC chip though.
Please explain what's not correct and which drawing you're talking about.
 

kinarfi

Well-Known Member
we did it!!!!
more later
 

MrAl

Well-Known Member
Most Helpful Member
Please explain what's not correct and which drawing you're talking about.

Well actually i dont think i see any circuit yet other than the one i drew myself that might work with the LM2576. Is it that you dont want to try that IC with the external MOSFET? Just wondering that's all.
 

kinarfi

Well-Known Member
This unit was holding 18 volts at 9 amps, the load got hot, but the LM2578, IRF4905, Diode and Transistor stayed cool.
The 100 ohm resistors got hot too, they need to be 2 watt or better, look like a good design! Now to do the same for the negative side of my power supply
https://www.dropbox.com/sh/wfpkr5w0hcrz7a2/RoElm-fUVF
 

Attachments

MrAl

Well-Known Member
Most Helpful Member
Hello again,

Well that is good to know, that at least it works, so we have a shot at doing this then. Thanks for doing that and posting the results.

But i do have to ask why you decided to use a PNP inverter rather than an NPN inverter? My original design had an end goal that included going to an offline buck that can convert 170vdc down to 20vdc or less. So there the NPN would be better because it would automatically do the required level shifting as well. We'd then have to limit the output swing.

Did you try it with an NPN too or no?
 

kinarfi

Well-Known Member
Short answer is yes, I thnk so, long answer, If I remember correctly, & I think I do, Pin 2, output, starts out High and turns the NPN on so the FET is on also. With the PNP, the output turns it off and the FET is off until the output sends its signal out. Also, if you look at the two screen shots of my scope, I had to switch to AC because it couldn't be seen on DC because it was on top of 24 volts. Sorry for not documenting everything, but I get going and don't take time to write it down as I go from one possible solution to the next, and then, I have it working and have to try to remember what I did to get here. Added some photos https://www.dropbox.com/sh/wfpkr5w0hcrz7a2/RoElm-fUVF . The drawing in post 75 is accurate except for the output filter and as you can see, the 100 ohm, 1 watt resistor do get HOT. Got any specific tests you would like me to make?
 

MrAl

Well-Known Member
Most Helpful Member
Hi again,

Oh so you tried it with the NPN and it worked that way too? That would be nice.

Well it would be nice if we could (using the NPN version) get to a higher input voltage that way, like say maybe 50 volts to start with only maybe 10 volts output, then work up to maybe 170 volts input. This could get dangerous though so safety first!

Also, we have not yet addressed the issue of current limit. With the PMOS there is no current limit because nothing senses the output like it did before when the LM2576 was driving the output (in the normal circuit without PMOS). The IC can sense an overcurrent but the PMOS can not, so next we need to add current limit.

If you do the 50v input test of course you need a regulator or some lower value voltage to run the LM chip itself because the regular version only takes something like 36 volts or around that, and the 170v version if you do that test also requires this lower voltage level too just to turn the IC.

Above all be very careful because when the input voltage goes up things can explode violently and shoot very hot metal parts across the room at high speeds.
Actually though even at 24v you should be very careful, but at 170v the potential to do serious damage increases by a large factor.

I'll try to draw up a mod for the current limit, and if you have any ideas you can draw them up too and post so i can take a look.
 

kinarfi

Well-Known Member
I don't remember if NPN worked or not, I'll try to test it tomorrow, I'm building a negative regulator. It's interesting how easy it is to get the negative voltage, same circuit, just move the negative feed, (ground) to the point between the inductor and the capacitor. A problem I have is the lack of higher voltage power supplies. I do have a Variac and I guess I could put a bridge on that. But that has no isolation. I got to see (hear) a 22uF 100V cap fail today, haven't found the canister and there was paper everywhere. I kind of think it got hot because of it's proximity to the 100 ohm resistors.
I just simmed the circuit with a NPN and it does run, (last attachment)
Since I'm always hands on with my power supply usage and it does have definite limit, I don't think I'm going to worry to much about current limits, If I find a need for it, I'll just use a LM317
 

Attachments

KeepItSimpleStupid

Well-Known Member
Most Helpful Member
I'll take a stab at 4-corners:

The limits of some CC/CV power supplies are rectangular, e.g all voltages at full current. Some supplies have "foldback" current limiting and not "constant current" per say. If I short one of my (0-32 V), 10 A supply; Actually, it's a linear (selectable power supply), They want you to change the transformer tap and then a screwdriver adjust adjusts between a minimum and maximum V. So, they called it a Fixed voltage power supply, selectable from 0-32 Volts. Anyway, if you short the output the current goes to 10A, and the voltage goes to zero, so the operating area is not rectangular,

I doubt the reference was 4-quadrant where V+,I+, V-,I- , V+,I- and, V-I+ is permissible.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top