TPS1H000-Q1: Current limit variation

Yoshikazu Kawasaki

Part Number: TPS1H000-Q1

Hello,

Please let me confirm if my understanding for the current limit is correct because I looked at similar threads, but I'm not sure if my understanding is correct.

My customer's conditions

Holding mode
Rcl=1kOhm -> The expected current limit is 480mA

My customer's findings

It didn't hit the current limit with continuous 630mA even for >10secs
It hit current limit if they increased the current gradually, but it was 631mA with nFAULT=low

My understandings

If the current limit is 480mA, it has variation of +/-10% as shown below, so it will be between 432mA and 528mA. The actual current limit is 1.5x the current limit setting which is between 648mA(=432x1.5) and 792mA(528x1.5). The customer's case is slightly lower than 648mA, but it would be measurement error. Is my understanding correct? If not, would you please tell me the reason why my customer saw the higher current limit than expected?

Best Regards,

Yoshikazu Kawasaki

2 months ago

0 Patrick Shalton 2 months ago

TI__Mastermind 32830 points

Hi Kawasaki-san,

That is correct. Once the current limit is tripped, the device will limit the current to the calculated value (480mA typ). But if the current increases slowly past the programmed current limit threshold, the current limit may only trip up to 1.5x above the programmed limit.

Does the customer see the device limit the current to 631mA steady-state? What components do they have the DELAY pin connected to?

Thanks,

Patrick

0 Yoshikazu Kawasaki 2 months ago in reply to Patrick Shalton

Intellectual 931 points

Hello Patrick-san,

Thank you very much for your quick reply.

I'm not sure if they see steady state of 631mA, but it wouldn't be. They connect DELAY to GND directly.

> Once the current limit is tripped, the device will limit the current to the calculated value (480mA typ).

Do you mean if the current changes from 100mA to 500mA, for example, it will hit the current limit? The following shows the plot I got from them, but it didn't hit the current limit.

Best Regards,

Yoshikazu Kawasaki

0 Patrick Shalton 2 months ago in reply to Yoshikazu Kawasaki

TI__Mastermind 32830 points

Hi Kawasaki-san,

One more point, can the customer provide a schematic? If they are using a diode/resistor ground network and the current limit resistor is connected to system ground instead of IC ground, the current limit will be shifted.

Thanks,

Patrick

0 Yoshikazu Kawasaki 2 months ago in reply to Patrick Shalton

Intellectual 931 points

Hello Patrick-san,

I sent you friendship request. Please accept it so that I can send you their schematic without showing in public.

Best Regards,

Yoshikazu Kawasaki

0 Patrick Shalton 2 months ago in reply to Yoshikazu Kawasaki

TI__Mastermind 32830 points

Hi Kawasaki-san,

I've accepted the request, let's continue via messages.

Thanks,

Patrick

0 Yoshikazu Kawasaki 2 months ago in reply to Patrick Shalton

Intellectual 931 points

Hello Patrick-san,

I sent you the schematic last Thursday. I expected you had noticed that, but it doesn't seem to be. Please review that and give me your comments.

Best regards,

Yoshikazu Kawasaki

0 Yoshikazu Kawasaki 2 months ago in reply to Yoshikazu Kawasaki

Intellectual 931 points

Hello Patrick-san,

I sent you the schematic to your inbox around a week ago. Are you aware of that? I'm sorry when you're busy, but would you please give me your commets?

Best regards,

Yoshikazu Kawasaki

0 Patrick Shalton 2 months ago in reply to Yoshikazu Kawasaki

TI__Mastermind 32830 points

Hi Kawasaki-san,

I've received the schematic, and apologies for the delay. I'll update this by the end of Monday, Dallas time.

Thanks,

Patrick

0 Yoshikazu Kawasaki 2 months ago in reply to Patrick Shalton

Intellectual 931 points

Hello Patrick-san,

Thank you very much for supporting me. I'll wait for your reply.

Best Regards,

Yoshikazu Kawasaki

0 Yoshikazu Kawasaki 2 months ago in reply to Yoshikazu Kawasaki

Intellectual 931 points

Hello Patrick-san,

I'm sorry to annoy you when you're busy. I know yesterday was Veterans day, so would you please give me your comments today for the schematic if it has a potential reason why their circuit didn't hit the current limti when tripped?

Best Regards,

Yoshikazu Kawasaki

0 Patrick Shalton 2 months ago in reply to Yoshikazu Kawasaki

TI__Mastermind 32830 points

Hi Kawasaki-san,

Thanks so much for your patience on this issue.

For background, it will be helpful to read Section 4.1 of the following app note: https://www.ti.com/lit/an/slva859b/slva859b.pdf?ts=1762914583139&ref_url=https%253A%252F%252Fwww.google.com%252F

I believe that when the customer sees the current limit trigger at 631mA, the FET is still in the linear region, and thus the device still does not have enough VDS drop to properly regulate. Can the customer try increasing the current past 631mA? I believe that when the current is increased more past when the current limit first trips, it will then regulate to the programmed value.

Thanks,

Patrick

0 Yoshikazu Kawasaki 2 months ago in reply to Patrick Shalton

Intellectual 931 points

Hello Patrick-san,

Thank you very much for your comments when it's off in America.

I got no comment for the schematic from you, so I understand it has no issue.

As you mentioned, if they increased the current by having the load of 5Ohm from 50Ohm, it hit the current limit at around 480mA as expected. Do you mean the current hitting the current limit needs to have Vds > Ron x Io?

Best regards,

Yoshikazu Kawasaki

+1 Patrick Shalton 2 months ago in reply to Yoshikazu Kawasaki

TI__Mastermind 32830 points

Hi Kawasaki-san,

Yes, the schematic looks good.

When the device detects an overcurrent condition, it will start to regulate the main FET gate voltage, but it will not limit the current accurately until the VDS increases and the FET transitions into the saturation region. So in one sense, yes, in order for the current limit to be accurately limiting, VDS > Ron (typ) * Io. This will happen naturally at/below 1.5x the current limit setting, due to the current limit feedback circuit.

Thanks,

Patrick

0 Yoshikazu Kawasaki 2 months ago in reply to Patrick Shalton

Intellectual 931 points

Hello Patrick-san,

Thank you so much for your detail explanation. Now I understand the reason well why my customer saw the behaviors at the current limit detection.

Best regards,

Yoshikazu Kawasaki

Power management

Power management forum

TPS1H000-Q1: Current limit variation