TPS25947: schematic check

Hi Gareth,

1. Did you try replacing IC?

2. How much is load current?

3. Is fault pin pull up supply stable?

4. Can you probe other pins like VOUT,EN,OVLO,dvdt,also current through device?

5. Also at what input voltage level this problem starts?

Regards

Kunal Goel

Hi kunal

1. Have not replaced the IC

2. Almost no load

3. Pull-up voltage is stable

4. If 0.5A is applied to the VOUT terminal, the fault problem disappears. Do you have any comments?

Hi Gareth,

1. This problem is with U35 only? and not with other 259470 devices on board?

2. FLT can go low in these scenarios , can you see if any of these may be happening?

,

3. Do you have other boards also? We can do swap testing to see whether device issue or board issue?

Regards

Kunal Goel

Hi Kunal

May be caused by reverse current

Customers have measured vout> vin,

but the gap is not big

Is there any solution to this problem?

Hi Gareth,

They can use TPS259461. It is pin to pin to TPS259470 but does not block reverse current so FLT will not go low. 

Regards

Kunal Goel

Hi Kunal

May I ask EVM Schematic Component D11 can reduce Vout>vin?

D11 clamps VOUT to VIN voltage at safe level for device. It may not help in here as VOUT is nearly equal to VIN. Why is VBUS going above VIN, is it application requirement? Is some voltage source applied at VBUS? How much is VIN during this test?

Hi Kunal

Control method is as follows,

First PD1_CSP_P2 will offer 5V (U33 ON & U35/U34 OFF) as default waiting for Type C VBus configuration.

When power contract of variable PDO is matched and created then will turn U33(5V) off and turn U35(20V) on, turn U34(9~12.6V) off with AUXOFF of U35.

The FLT# of U35 is always trigger when U35 is ON.

Hi Gareth,

Do you have a waveform showing VIN,VOUT, current through eFuse? I am not able to understand why VOUT is going higher than VIN. How much is VOUT above VIN? 

Regards

Kunal Goel

Hi Kunal

Please refer to the waveform provided by customers

Thank you

CH2: ADP_VIN (eFuse IN)

CH4: VOUT (eFuse_OUT)

M1= CH2 – CH4

Hi Gareth,

Can you explain waveform? Looks like VIN and VOUT are both negative, below the 0 zero line.  Also FLT is not going low in this waveform, which means it is working fine?

Regards

Kunal Goel 

Hi Kunal

Sorry for typo on M1 definition

Corrected M1 definition is CH4 – CH2 (Vout -Vin)

CH2: ADP_VIN (eFuse IN)

CH4: VOUT (eFuse_OUT)

M1= CH4 – CH2 =  Vout – Vin

V1=Cursor 1 = CH2 (VIN)

V2=Cursor 2 = CH4 (VOUT)

△V= V2 – V1= CH4 – CH2; if △V is positive value  Vout > Vin; △V is negative value  Vout < Vin

Because the time is  2ns, FLT may not display LOW instantly

Hi Gareth,

It looks that because of some reason VIN is dropping but Vout is staying high due to cap at output. Any idea why VIN is dropping? We can try test again by removing cap at output. Also do you have a EVM in hand to test on it also?

Regards

Kunal Goel

Hi Kunal

After the customer disables the BQ25703A on the board, TPS259470 FLT# is normal (layout is as follows) It should be a charger couple.

Are there any sensitive areas on the TPS259470 layout?

Hi Gareth,

Layout is not much clear. But it looks fine at top level. Looks like BQ25703A was causing VIN to drop. May be we can try increasing R359 to see if helps.

Regards

Kunal Goel

Also one thing how is pullup supply to FLT pin coming? 

Hi Kuanl

TPS25947 and bq25703 are the same vin bq25703 Enable, VIN will drop due to high side mosfet switch as shown below

1.Also one thing how is pullup supply to FLT pin coming? check ok

2.2. Whether TPS25947 is too sensitive on the VIN end based on the customer's application, is it recommended to use LM73100 or have other comments

Hi Gareth,

Sorry I could not understand why VIN drops? Which High side switch you mean?

Gareth TPS25947 has reverse current blocking feature. Since VIN is dropping beyond detection threshold so device detects it and make FLT low to show that it is blocking reverse current. device is working in  expected way. 

I will recommend using TPS259461. It does not have reverse current blocking feature , so FLT will not go low.

They can try adding high cap on VIN to prevent it from dropping much.

Regards

Kunal Goel

Hi Kuanl

Can TPS259461 also use TPS259470 auto control as shown in the figure below?

Hi Gareth,

I just realized your application is power muxing , TPS259461 does not block reverse current always  and for power muxing application always RCB is needed. You can use TPS25461 like above figure ,SPLYGD is same as AUXOFF . TPS259461 block reverse current when device is off but when on then it can allow. So suppose your 20v rail is on,12v rail is off , then if external supply is applied at output >20V then reverse current can go into device . Does your system need always RCB?

Regards

Kunal Goel 

Hi Kunal

TPS259461 no stock samples TPS2597 is recommended?

Hi Gareth,

TPS2597 does not have back to back FET. It will allow reverse current so not suitable for power MUXing application. 

You can contact here for stock related query on TPS259461:https://ticsc.service-now.com/csm/?_ticdt=MTYxMTc0MDIxMnwwMTczMDU5NDU1ZmUwMDAwM2Q4OWQ2YzRhYjUzMDMwNzMwMDEzMDZiMDA3ZTh8R0ExLjIuMTEzOTAzMTgyMi4xNTkzNTI2NzExfDA

Did customer try with adding  more cap at  VIN pin near to the VIN pin  of TPS25947 device? 

Regards

Kunal Goel

Hi

Got it thanks

Cap have been added to the vin side Vin still drops

Hi Gareth,

Can you explain me why VIN is dropping? Reason is not clear to me. Also if FLT goes low is it a major issue for customer system?

Regards

Kunal Goel

Hi Kunal

1. This Vin voltage drop is found on the charger IC next to it

2. FLT will be used to judge the current state

Thank you

Hi Gareth,

Best I can think is shifting to TPS259461. Meanwhile customer can keep trying ways to prevent VIN drop. I felt some cap like 5-10uF very close to VIN pin of eFuse would have helped. Let me know if I can close the thread?

Regards

Kunal Goel