TPS7A8101 output voltage can't be adjusted

Alan,

You are correct in you assumption that the FB pin should equal the NR pin voltage, but it sounds like the device may have been damaged around the FB node, which has an absolute maximum rating of 3.6V.

The damage can come from a number of things, so I would suggest trying a new part as your first step to debugging.

Regards,

David

David,

I have tried several parts, the issue is the same.

I also checked EN pin, it works. And I also checked NR pin, it is about 0.78V, not the exact 0.8V.

The problem is that the output voltage is almost same as the input value.  When input changes, the output changes too.

FB node voltage is not fixed and will change with the output voltage, its value is just the ratio of the output divider resisters.

From these clues, do you have any idea?

Thanks.

Alan

Alan,

The NR pin will typically be a little lower than specification if your measurement of the NR pin loads down the pin.

I would need to see your schematic for further debugging, as these hints point to something else going wrong outside of the device.

Regards,

David

David,

 Please find below schematic, I just designed according to the datasheet.

Alan,
There are some cases where having a feedforward capacitor (C2351) can lead to problems with the device. We advise the use of them because they do improve the performance of the device, but in abnormal circumstances problems show up, including possible damage to the device. Please refer to the following application note for more information:

http://www.ti.com/lit/an/sbva042/sbva042.pdf

I would advise removing the feedforward capacitor and testing your circuit again with another device. If that fails, can I get copy of your board layout around the LDO? 

Regards,

David

Hi David,

Please find below Schematic and Layout:

When I tested the LDO, I took off L511 and L512, which means there will be no other load that can impact the LDO.

I have removed the C2351 according to your suggestion, but the issue still exists.

The key issue is that why the FB node doesn't work and the output voltage is same as input one. And also the output voltage changes with the input one. If as you said, the FB node broken, is that possible that all chips (at least five) was broken at the FB pin while other pins are OK? It is so strange. I am very confused now.

Have you guys ever met this phenomenon before? I have no idea at all, please help.

Thanks.

Alan.

Alan,

It is extremely unlikely that you have had 5 damaged units, and I still suspect this problem lies elsewhere.

What is the FB voltages that you are seeing? Are they below 0.8V?

Another thought from a coworker - can you remove R861 and short R851(the bottom resistor in the divider)?

This will put the device into unity gain feedback so Vout = Vref. The thought was that if one of your actual resistor values is drastically different than what it should be than this could cause the issues, mainly if R1 was much larger than expected FB would never reach Vref, and Vout = Vin.

Regards,

David

Hi David,

Today I did the testing according to your suggestion that made the FB pin short to Out pin.

Now Vout is not equal to Vin, but still not fixed as what we expected should be = 0.8V. It still changes with Vin but looks like by a certain ratio.  Below is the list of the test result:

Vin      Vout

2.5       1.19

2.75     1.25

3           1.32

3.25       1.38

3.5          1.45

4            1.57

it shows that when Vin increase 0.25 V, the Vout will increase about 0.06V.  

And I also did another test, I removed all divider resistors, that meant the FB pin is floating.

The result is that Vout = Vin and Vfb = 0.

Could you get some ideas from these hints?

BTW, for this project, out goal is that we need a 3.3V power, current needs about 1A. Our input source is 5V, so I think maybe I can use a fixed output LDO which can provide 3.3V.   As we now don't have enough time to change the layout, we need a same package and similar pin format that I can reuse the layout.  Can you guys recommend any IC for me? 

Thanks.

Alan

Alan,

That seems like you have something downstream that is pulling up the output voltage. Once the output voltage increases above what you designed for, the LDO stops regulating. Can you check on what Vout is without the IC present, but with Vin still being swept through?

For a fixed voltage, the TPS73733 is nearly pin-to-pin compatible and should work for your requirements.

Regards,

David

Hi David,

Issue has been solved, LDO works well now.

It looks I have made a silly mistake.

The problem is that when I tested the LDO, I disconnected all downstream circuits that meant there are no any loads for the LDO, so the LDO worked abnormal.

Now I connected the downstream circuits, everything is ok.

Thanks for your help.

Alan Zhang

Alan,

Good to hear and I hope this was the only problem in your system - definitely a hard one to track down.

Regards,

David