TPS63900: In real test I found this device actually need around 400uA quiescent current

Dear User:

How do you test the Iq? It should be measured under no switching status.

Hi Minqiu,

Thanks for your reply.

Here I use the schematic as the attachment.

VBAT is 4.2V. Vout is 3V3 and doesn't connect to any load.

EN and SEL connect to VBAT. 

In this state I've got a quiescent current from VBAT to TPS63900 about 400uA.

I have tested this circuit with PTPS63900(which is the version that before mass production of TPS63900), and the current value is just 500nA.

If the chip is XPS63900DSKT(which has a known silicon issue mentioned in it's datasheet) then the quiescent current is about 430uA.

Seems pretty, the final version TPS63900 has the same performance as XPS63900DSKT?

Dear User:

I afraid you are testing the "No load current", no quiescent current. The quiescent current is measured under no switching status. (Generally, we will test in special TEST mode in special bench. You can add another DC supply at Vout pin with a little higher voltage)

Here is an application note shows the details.  www.ti.com/.../slyt412.pdf.

Hi Minqiu,

Sorry for my misunderstanding.

For our application we don't have this special condition that force a little higher voltage on VOUT pin. So seems no need to verify this quiescent current.

So my question should be about the minimum normal working current this chip required.

Seems it needs at least a 400uA current.

That means if we only need about 15uA output from 3V3, then this chip's efficiency seems pretty.....not good.

Even the pre-production version PTPS63900 is better than this mass production version. Am I right?

Dear User:

I have no idea about the different version, as I only get the data for the standard version. 

The main loss is the switching loss and external components like the inductance in such condition. If the Vout is higher than target, it won't switch, so that it only need a little current as the Iq. Once it switch, the power loss will increase soon, comparing with such low loading.

Actually, even  the efficiency is not pretty good with such ultra low loading, but I think it's still higher than a lot of other devices. 

Hi Minqiu,

Thanks for your advice. Anyway we will continue the product with this device. After all in it's datasheet it indicate around 90% efficiency at 1uA load current. Hopefully we got some material mistake. I'll get some of this device from TI store and test them. 

I'll update this then.

Thank you very much for your help :)

Best Regards

Kevin

Hi User:

I'm so sorry that I mistaking the unit. It shouldn't be 400uA with no load. The efficiency is not 90% but 80%.   

Would you check the soldering quality? How about replace another new IC?