TINA/Spice/LMP8603-Q1: Gain Reduction doesn't work

NOBUO FUJIHARA

Part Number: LMP8603-Q1
Other Parts Discussed in Thread: TINA-TI, LMP8603, LMP8601,

Tool/software: TINA-TI or Spice Models

Hi Team,

I simulated LMP8603 TINA-TI model by putting 150kohm resister to reduce Gain as attached file.

This method is described in the datasheet, I expects Gain would be reduced from 100 to 60V/V, but output signal gets 1/24 instead.

Could you check attached TINA-TI model and fix LMP8603 model?

Best regards,

LMP8603 Test Circuit.TSC

over 4 years ago

0 Patrick Simmons over 4 years ago

TI__Mastermind 21825 points

Hello Nobuo,

Thanks for reaching out on the forum. I am investigating your issue right now. I suspect your issue is similar to the issue found for the lmp8601 in this e2e post. As such, I do not think you will have the option to adjust the gain when you are attempting to do bidirectional sensing. I will try to confirm this within the next day or so.

0 Patrick Simmons over 4 years ago in reply to Patrick Simmons

TI__Mastermind 21825 points

Hello Nobuo,

As the lmp8603-q1 uses the same architecture as the lmp8601 I tested in that previous post, you will not be able to adjust the gain. I verified on the bench with LMP8603 that you will not be able to get a meaningful measurement with the offset set for bidirectional measurements and and resistor placed at A0 and A1 to adjust the gain. As in the other e2e post, you might be able to compensate for the fixed gain by changing your shunt resistance. If you go that route, you will need to short offset to VS if you want to do a bidirectional measurement.

0 NOBUO FUJIHARA over 4 years ago in reply to Patrick Simmons

TI__Expert 8265 points

Hello Simmons,

In actual customer's project, changing shunt resistance is not feasible.

From spec ,reliability and adjustable Gain point of view, I have promoted LPM8603-Q1.

Is there no plan to modify the model?

Best regards,

0 Patrick Simmons over 4 years ago in reply to NOBUO FUJIHARA

TI__Mastermind 21825 points

Hello Nobuo,

Your request extends beyond the model. The silicon is designed such that the gain can be adjusted for unidirectional forward current sensing. However, the adjustable gain does not work for bidirectional sensing. For bidirectional sensing, the customer must use the default gain.

The model operation mimics the actual device operation. Attached below are Tina circuits that show the model works for adjusted gain unidirectional sensing and fixed gain bidirectional sensing.

/cfs-file/__key/communityserver-discussions-components-files/14/LMP8603_5F00_Operation.TSC

0 NOBUO FUJIHARA over 4 years ago in reply to Patrick Simmons

TI__Expert 8265 points

Hello Simmons,

By providing bias voltage to previous simulation model as attached, it looks operating nice with bidirectional usage.

Best regards,

LMP8603 Test Circuit_2.TSC

0 Patrick Simmons over 4 years ago in reply to NOBUO FUJIHARA

TI__Mastermind 21825 points

Hello Nobuo,

That is an interesting find. You might have found a workaround. The LMP860x is a part we inherited from the acquisition of National Semiconductor, as such some details on the inner workings of this part are not as readily available as the parts our design engineers made. I will check on the bench, to verify that the model behavior actually matches the device behavior for this operating condition.

0 Patrick Simmons over 4 years ago in reply to Patrick Simmons

TI__Mastermind 21825 points

Hello Nobuo,

Thanks for your diligence. It does look as though the device can do what you simulated. Here are my bench results:

0 NOBUO FUJIHARA over 4 years ago in reply to Patrick Simmons

TI__Expert 8265 points

Simmons-san,

Thanks for the bench test for confirmation.

Best regards,

Amplifiers

Amplifiers forum

TINA/Spice/LMP8603-Q1: Gain Reduction doesn't work