LM2597 inductor value

Rade Vukanovic

Other Parts Discussed in Thread: LM2597

Hey TI.

I try to configure the LM2597 with following parameters.

Vin 30-38V, Vout 5-7V, Iout 120-200mA max.

In datasheet i to see a 330uH value for inductor, but in simulation it's 1mH. What's correct? Also, what other values I can use to preserve the high efficiency number?

over 9 years ago

0 Frank De Stasi over 9 years ago

TI__Guru 57555 points

When you say "simulation" do you mean Webench ?

I do not know why you would get such a large value of 1mH; what settings did you use?

The value of 330uH sounds like the correct one.

0 Rade Vukanovic over 9 years ago in reply to Frank De Stasi

Prodigy 130 points

Hello Frank. Yes, "Webench" calculate 1mh . Settings were in Vin 30-38v, Iout 0.130A Regulated Vout 6.8v. But thanks, 330uh is value I have on stock. Regards, Rade

0 Rade Vukanovic over 9 years ago in reply to Rade Vukanovic

Prodigy 130 points

Dear Frank. one more problem with Webench. I tried to export the schematics and hoped to speed my prototyping time by using the Eagle/Cadsoft export (I am a 6.5 user)..

Unfortunately, this appears to not be working at all..

when I try to open the .sch file, I receive this error:

line 1, column 22269: redefinition of name 'IDC' in tag <attribute>

Additionally, the Library import reports some errors too..
Any chance this can be fixed?

Thanks!!

0 Frank De Stasi over 9 years ago in reply to Rade Vukanovic

TI__Guru 57555 points

I will forward this to the Webench guys.

0 Tommy Jewell - WEBENCH Design Center over 9 years ago in reply to Frank De Stasi

TI__Intellectual 2930 points

Hi Rade,

It looks like the Custom inductor in the design is causing some issues with Eagle. I will need to take a deeper look into this.

In the mean time I'm emailing you a manually corrected copy of the Eagle file and library.

Best regards,
Tommy

0 Rade Vukanovic over 9 years ago in reply to Tommy Jewell - WEBENCH Design Center

Prodigy 130 points

Hi guys, I want to thank you for support. My prototype works as a charm. Now I wonder what are the disadvantages of placing feedback resistor closer to the inductor. In my prototype board i placed them similar to suggested position (from datasheet), now I need to finalize my board design and moving fb resistors will save me some space. Just to mention, inductor is Coilcraft's shielded MSS1246 type.

Cheers, Rade

0 Arief Hernadi over 9 years ago in reply to Rade Vukanovic

TI__Mastermind 23815 points

Hi Rade,

I think the key point is to keep the FB pin away from the noisy stuff (eg. switch node). After the inductor (the VOUT side) on a buck design, the current is continuous therefore there is less noise around that area compared to the inductor side that is closer to SW node.

A good practice for the location of the feedback resistor is to try to keep it close to the FB pin (especially the bottom feedback resistor). The reason is because the FB pin is a high impedance node and it's quite noise sensitive.

Thanks

-Arief

0 Rade Vukanovic over 9 years ago in reply to Arief Hernadi

Prodigy 130 points

Hey Arief . All clear. Thanks,

Rade

0 Rade Vukanovic over 9 years ago in reply to Rade Vukanovic

Prodigy 130 points

Hi again. As I mentioned before circuit works as expected but I have one issue. Since my Vin is not regulated and it's very "soft" (R/C network) it happen that Vin voltage fall under the regulation voltage. In that situation efficiency is very poor input current is too high and the regulator stays in that regime. Only disconnecting and restarting helps. Is it possible to achieve the soft start "restarts" without restarting the power?

Thanks, Rade

0 Arief Hernadi over 9 years ago in reply to Rade Vukanovic

TI__Mastermind 23815 points

Hi Rade,

In order to soft-start the LM2597, you can pull the SHDN/SS pin low (using a MOSFET) in order to shutdown the IC (no longer switching) without disconnecting and restarting the power. Just be caution if the Vin voltage is still low, then the input current will again be high.

Thanks

-Arief

0 Rade Vukanovic over 9 years ago in reply to Arief Hernadi

Prodigy 130 points

Hey Arief, thanks, idea looks promising. Rade

0 Rade Vukanovic over 9 years ago in reply to Rade Vukanovic

Prodigy 130 points

Hey Arief. At first it sounded like a simply circuit to design but i figured out that I cant arrange mosfet to the desired function. is the configuration or circuit available somewhere so I can use it as the start point? I will really appreciate if you can post some examples. Thanks, Rade

0 Anston Lobo over 9 years ago in reply to Rade Vukanovic

TI__Genius 16025 points

Hi Rade,

Assuming you are referencing a circuit implementation, please take a look at this:

2084.[E2E] LM2597 Enable circuitry.pdf

This should be sufficient to retain soft start functionality and still provide ON/OFF.

Thanks,

Anston

0 Rade Vukanovic over 9 years ago in reply to Anston Lobo

Prodigy 130 points

Hey Anston. Yes, I wonder about the circuit. Also I was not sure can the SS pin be short grounded. Now I see that it can be done simple. I just need to add the voltage threshold circuit in front of he mosfet gate. Right?

Thanks, Rade

0 Anston Lobo over 9 years ago in reply to Rade Vukanovic

TI__Genius 16025 points

Hi Rade,

No need of a voltage reference circuit. The SS/SHDN pin has an internal CC source that provides a 5uA charging current. This current charges the capacitor and provides the requested soft start feature.

All you need is the MOSFET and a gate bias to control its operation.

Thanks,

Anston

0 Rade Vukanovic over 9 years ago in reply to Anston Lobo

Prodigy 130 points

Hello Anston. Sorry, it must be that I don't determine my question good enough. Please check the picture. Thanks again and sorry for the dumb question.

Rade

0 Anston Lobo over 9 years ago in reply to Rade Vukanovic

TI__Genius 16025 points

Hi Rade,

You can use a resistor divider, or an LDO or connect it to a uprocessor to get targeted switching action. The list is endless.

Vgs is only 2.1V.

Play around with it, you have limitless options at your disposal and worst case you blow the FET. :)

Thanks,

Anston

0 Rade Vukanovic over 9 years ago in reply to Anston Lobo

Prodigy 130 points

Thanks Anston. Now I finally have the complete picture of the circuit. Very simple and I like it ;)

Cheers,

Rade

0 Rade Vukanovic over 9 years ago in reply to Rade Vukanovic

Prodigy 130 points

Anston, do i miss something or this mosfet arrangement works opposite as desired? When the Vgs is greater than 2.1V drain is on source to and when the Vgs is less than 2.1v drain is open?

thanks, Rade

0 Anston Lobo over 9 years ago in reply to Rade Vukanovic

TI__Genius 16025 points

Hi Rade,

When Vgs is less than 2.1V, the Mosfet is off, and the regulator is on.

When Vgs is more than 2.1V, the Mosfet is on, the SHDN pin is referenced to ground and the regulator is turned off.

Thanks,

Anston

0 Rade Vukanovic over 9 years ago in reply to Anston Lobo

Prodigy 130 points

Hey Anston. right, but I need SS pin to be grounded at voltage less than desired not more than desired. Right?

thanks, Rade

0 Anston Lobo over 9 years ago in reply to Rade Vukanovic

TI__Genius 16025 points

Understood.

In that case use a p channel FET. The IC will operate as long as the PMOS gate is supplied with a charge.

The PMOS works opposite to NMOS. They are off when the voltage to the gate is +V, and on when the voltage is negative, or zero.

-Anston

Power management

Power management forum

LM2597 inductor value