LM2575 variable 7A power supply

Status
Not open for further replies.

Swend

Full Member level 4
Joined
May 14, 2019
Messages
237
Helped
10
Reputation
20
Reaction score
9
Trophy points
18
Location
Roskilde, Denmark
Activity points
2,304
Hi Friends.

I want to increase the current of a LM2575 and make a adjustable 7A/30V power supply. I want to keep the LM2575' current sense feature so I came up with this design which works sort of OK in LTspice, it splits the current in a 1:10 ratio i.e. 1 part through the LM2575 and 10 parts through Q2.

The problem is I can not figure out how to reduce the current sense resistor, I would like to use e.g. 0.1 ohm as current sense and not 3.3 ohm as it is now.

 

your schematic makes no sense. I have no idea how you're simulating this.
 

your schematic makes no sense. I have no idea how you're simulating this.

Hi barry. Sorry about that. The LM2575 is just there for illustrative purposes and give context, you can disregard it. What I'm simulating is this:


The total current that goes through LOAD is 7A, so I want 6A through R4 and 1A through R5, which it does now - but the SENSE resistor is too high.
 

Sorry, that still doesn't make any sense to me. Here's a thought: why don't you simulate your ACTUAL CIRCUIT?
 
Reactions: Swend

    Swend

    Points: 2
    Helpful Answer Positive Rating
The LM2575 is a switchmode regulator.... How can you predict that your simplistic transistor drive will dynamically share the current correctly, even if statically it "appears" to do so?
 
Reactions: Swend

    Swend

    Points: 2
    Helpful Answer Positive Rating
Sorry, that still doesn't make any sense to me. Here's a thought: why don't you simulate your ACTUAL CIRCUIT?

Yes that's a great idea, but a spice model for LM2575 is hard to come by if it exist.

- - - Updated - - -

The LM2575 is a switchmode regulator.... How can you predict that your simplistic transistor drive will dynamically share the current correctly, even if statically it "appears" to do so?

The correct answer is: I can't predict anything, that's why I'm asking more experienced people. I will interpret your question as an indicator that it will likely not or will be too complex, so I will discard the idea.

- - - Updated - - -

Restart your design, try a recent switch mode controller with MOSFET power stage.

Yes, that's what I tried yesterday, I'm sure they exist but I simply couldn't find a simple controller (as the LM2575) with external power stage among the thousands there is to choose from e.g. Arrow in 96 pin PGA or whatever. So I tried to solve the problem when I instead should have asked if anyone can recommend a suitable controller to me.
 

Try this:
1) Search (google/bing..)for LM2575 spice model
2) PRESTO!

Or:
1) Go to TI website
2) Click on Products/Power Management/Find a Part
3) Enter your parameters
4) VOILA! 15 parts that will do what you want
 

Try this:
1) Search (google/bing..)for LM2575 spice model
2) PRESTO!

Or:
1) Go to TI website
2) Click on Products/Power Management/Find a Part
3) Enter your parameters
4) VOILA! 15 parts that will do what you want

Hi barry

Thank you for your kind proposals.

On the LM2575, that's a lesson learned for me not to give up, I saw this post https://www.edaboard.com/showthread...ice-model-needed&highlight=LM2575+spice+model and I gave up searching before I began.

On the TI website. Thank you once again, but I figured out how to do it on the arrow.com website, you just have to sort by output current, and those with highest output are usually the ones with external power stage.

LT1339 https://www.analog.com/media/en/technical-documentation/data-sheets/1339fas.pdf - I like this one, but it's DIP so I could be in doubt if this is considered a modern chip?

LT1076 https://www.analog.com/media/en/technical-documentation/data-sheets/1074fds.pdf - I also like this one but what I don't get is why it is limited to 2.6A when the monolithic LM1074 can do 5A, wouldn't it be expected it was the other way around?

And this one LT3800 https://www.analog.com/media/en/technical-documentation/data-sheets/3800fc.pdf


And Ha Ha!
 
Last edited:

The LT1339 also comes in an SMT package. You can always check with the manufacturer if you're concerned about end-of-life.

The LT1076 and LT1074 have different output structures.
 

The LT1339 also comes in an SMT package. You can always check with the manufacturer if you're concerned about end-of-life.

The LT1076 and LT1074 have different output structures.

Hi barry

I'm more concerned that nobody would here be interested if I use an antiquated part.

By monolithic I mean that the LM1074 has on-chip output FET's and can output 5A, whereas the LM1076 you bring your own FET's but the output is only 2.6A, usually it's the other way around monolithic chips are limited because you can't change whats on the chip - therefore I'm puzzled.

I found this LT1339 reference design here and I'm wondering if it's possible to use an instrumentation amplifier across Rsense and feed the output of the amplifier into SENSE+ and SENSE- so you can get variable current limiting?

 

    V

    Points: 2
    Helpful Answer Positive Rating
Hi,

LT1076 also has internal power switch....

Klaus
 

Status
Not open for further replies.

Similar threads

Cookies are required to use this site. You must accept them to continue using the site. Learn more…