TINA/Spice/LM3488: TINA/Spice/LM3488

Dean Munson

Part Number: LM3488
Other Parts Discussed in Thread: TINA-TI, , LM5155

Tool/software: TINA-TI or Spice Models

I am trying simulate the LM3488 in a boost configuration with VIN supplied from another source. The input voltage is 1.3 VDC with the output voltage set to 4VDC. The simulation works fine if I don't load the output. As soon as a load is applied the output drops to near 0 volts and the MOSFET gate drive stays high. There is also a large (and not unexpected) increase in the MOSFET current. It's at this point that the LM3488 stops driving the MOSFET. Schematic and simulation results attached.2480.SCHEMATIC1 _ PAGE1.pdf

over 4 years ago

0 Dean Munson over 4 years ago

Prodigy 150 points

2728.boost.pdf

0 Youhao Xi over 4 years ago in reply to Dean Munson

TI__Guru 68535 points

Hi Dean,

Thank you for considering the LM3488. Please increase your current sense resistor to 0.06. The circuit seems having inadequate current limit threshold, thus triggering the max duty cycle.

You may also consider the LM5155, our latest boost controller.

Thanks,
Youhao

0 Dean Munson over 4 years ago in reply to Youhao Xi

Prodigy 150 points

7776.boost.pdfMore or less same result except the MOSFET gate drive goes to 0 volts.

0 Dean Munson over 4 years ago in reply to Dean Munson

Prodigy 150 points

Tried increasing output cap to 1000uF but the LM3488 stops switching 2mS into the simulation.

0 Youhao Xi over 4 years ago in reply to Dean Munson

TI__Guru 68535 points

Did you increase the current sense resistor as we suggested?

0 Dean Munson over 4 years ago in reply to Youhao Xi

Prodigy 150 points

Sorry for the slow response. Was out of town.

I did increase the sense resistor as suggested but as I stated in my last post the result was the same except that the peak current was reduced and after the device stopped switching the gate drive pin went to 0 volts.

0 Youhao Xi over 4 years ago in reply to Dean Munson

TI__Guru 68535 points

Please attach your TINA native file so we can take a look.

0 Dean Munson over 4 years ago in reply to Youhao Xi

Prodigy 150 points

What is the TINA native file ? I am using the .lib file included with the .zip file. The .lib file is referenced via the Pspice profile editor under Configuration files/library.

0 Youhao Xi over 4 years ago in reply to Dean Munson

TI__Guru 68535 points

Hi Dean,

It is your original TINA model file, which should be the zip file.

Thanks,
Youhao

0 Dean Munson over 4 years ago in reply to Youhao Xi

Prodigy 150 points

snvmax0b.zip

Zip File attached

0 Dean Munson over 4 years ago in reply to Dean Munson

Prodigy 150 points

I also tried the encrypted model but got the same results.

0 Youhao Xi over 4 years ago in reply to Dean Munson

TI__Guru 68535 points

Hi Dean,

I will find time to check your model tomorrow.

Thanks,
Youhao

0 Dean Munson over 4 years ago in reply to Dean Munson

Prodigy 150 points

Hi Youhao,

Any luck with the model ?

0 Youhao Xi over 4 years ago in reply to Dean Munson

TI__Guru 68535 points

Hi Dean,

I have submitted the request and not yet get the reply yet. It may take a few more days.

Thanks,

Youhao

0 Saket over 4 years ago in reply to Dean Munson

TI__Intellectual 2680 points

Hi Dean,

I simulated your setup with Published Pspice model (on ti.com) and results are coming as expected. The main difference that I can see is with the MOSFET and DIODE model used in your schematic. I have used these models as per released Pspice model.

You can change them in your setup and see if results are coming fine.
I have attached the test bench,simulation results and Pspice project below. Please let me know if you have any issues.

snvmax1a_E2E_12JUN.zip

0 Dean Munson over 4 years ago in reply to Saket

Prodigy 150 points

Saket. I changed the diode and fet models but that's not the issue. In my simulation I have a Sw_tClose model that apllies the load at 12ms when the output has stabalized. It's at this point that the converter can't compensate and the output drops quickly to 0 vdc with the mosfet stuck on. I simulated a constant load and that works great but is not realistic. In the proposed system the load can vary from 4 watts to 20 watts and will not be applied until the converter output voltage is stable.

0 Dean Munson over 4 years ago in reply to Dean Munson

Prodigy 150 points

Hi Saket, I also tried making the output capacitor much larger to buffer the sudden change in load current but that also causes problems with simulation.

0 Dean Munson over 4 years ago in reply to Dean Munson

Prodigy 150 points

Hi Saket,

Any additional information available ?

0 Saket over 4 years ago in reply to Dean Munson

TI__Intellectual 2680 points

Hi Dean,

I was able to replicate the problem that you are facing with Pspice model. I think there is some issue with model. We will look into it and will try to get back by 06/24

0 Arpan Gupta over 4 years ago

TI__Expert 3140 points

Hi Dean,

I have looked into the issue and reduced the value of DCR of Boost inductor and ESR of the output capacitor to make it work. The values are kept as per the released test bench. Below is the test bench and simulation result.

Load transient Test setup(Vin=1.3V, Vout=4V, Iout= 200mA-->750mA):

Simulation result:

Thanks & Regards,

Arpan Gupta

0 Dean Munson over 4 years ago in reply to Arpan Gupta

Prodigy 150 points

So it still doesn't work. The output is no where near 4 volts.

0 Arpan Gupta over 4 years ago in reply to Dean Munson

TI__Expert 3140 points

Hi Dean,

Decreasing the resistance RFBB (Bottom feedback resistor) to 26.7kOhms we could achieve VOUT more closer to 4V. Below is the simulation result.

Thanks & Regards,

Arpan Gupta

Power management

Power management forum

TINA/Spice/LM3488: TINA/Spice/LM3488