LM20144-Q1: LM20144-Q1 DC/DC 1,0Vout

GIUSEPPE FERRARO

Part Number: LM20144-Q1
Other Parts Discussed in Thread: LM20144

Hi at all people.

I have a strange problem with LM20144 device. I attached here the project and simulation that i realized on Texas WEBDESIGN .

I have realized a board where I have 5 DCDC with LM20144, different voltages. The problem is ONLY with LM20144 1Vout 2A.

All circuits have the same layout , same circuit, different inductance and FB resistors, obviously, BUT all other DCDC run correctly at 1.8Vout, 3.3Vout, 2.5Vout, 1,2Vout. For all DCDC the same Vin: 5V

Just only 1,0Vout doesn't run correctly: in output I measure 1,2V, Vfb = 0,91V (error) and PGood, obvious, LOW level.

I don't understand if this is a LM problem.

Inside the datasheet i can see that is characterized until 1.2Vout.... that is the limit ?

Thanks at all for help that You would to do to me in this little problem.

Regards

PS: I don't find attach possibility .... how can I send You my simulation file ?

over 6 years ago

0 GIUSEPPE FERRARO over 6 years ago

Prodigy 140 points

0 JohnTucker over 6 years ago in reply to GIUSEPPE FERRARO

TI__Guru**** 191445 points

The LM20114-Q1 has an internal reference of 0.8 V. It should be able to support output voltage of 1.0 V. To attach files you should see a link at the bottom right t switch formats. Then you will see a tool bar at the top of your posting window. The clicking the paper clip icon will allow you to attach files.

0 GIUSEPPE FERRARO over 6 years ago in reply to JohnTucker

Prodigy 140 points

webench_esim_1043377_132_2_108293959.pdf

I use paper click to attach file here. I hope it's oK.

This design diagram it's the same for all my DCDC converter with LM20144: different inductance and some res values... other DCDC run , this doesn't run. Same board, same layout. If I connect output directly to FB pin, I can see in output 1.2V and not 0.8. really I don't understand...

0 JohnTucker over 6 years ago in reply to GIUSEPPE FERRARO

TI__Guru**** 191445 points

Did you copy the schematic and bill of materials exactly for this design as shown in the Webench report? if so then we will need to see some waveforms. Start with Vin, Vout and SW waveforms. I'l need a timescale that shows individual switching pulses at SW.

0 GIUSEPPE FERRARO over 6 years ago in reply to JohnTucker

Prodigy 140 points

Hi,

sorry for delay: I was out of office.

I attach here a PDF with acquisition. It seems that LM doensn't run. Never SW switch.

I'don't understand what happen.

Have You some ideas ?

Thanks

0180321 – LM20144-Q1 – 1Vout 2A – PROBLEMS.pdf

0 GIUSEPPE FERRARO over 6 years ago in reply to GIUSEPPE FERRARO

Prodigy 140 points

Sorry, I sent an errata PDF file.

Here there are correct acquisition, on start up.

Regards

0180321 – LM20144-Q1 – 1Vout 2A – PROBLEMS_errata_corrige.pdf

0 GIUSEPPE FERRARO over 6 years ago in reply to GIUSEPPE FERRARO

Prodigy 140 points

From datasheet seems that Lmin should be more than 1.1uH : 2,6uH, near.

I tried to change L with 3.3uH MCSDC1006-3R3MU : nothing change. I calculated this L value with DELTA IL 30%.

Also I changed Cout = 100uF 6.3V - same result: Vout = 1.174V without Rload, 1.089V with Rload = 0.5ohm

0 JohnTucker over 6 years ago in reply to GIUSEPPE FERRARO

TI__Guru**** 191445 points

You show two different start up sequences. The first never switches at all. The second switches during start up then stops. Is this the same board, just different start up attempts? Or is it two different samples? How many total boards behave this way?

0 GIUSEPPE FERRARO over 6 years ago in reply to JohnTucker

Prodigy 140 points

Hi, thanks for Your answer. I replies this situation on 4 boards where the only DCDC that doesn't run is 1.0V 1A dcdc. The others, 1.2V - 1.8V - 3.3V - 2.5V, run correctly . The layout is the same for all DCDC. 5 DCDC in the same boards, same Vin 5V. In my PDF first pictures are a NO LOAD. The other when the load is 0.5ohms. The start up is the same. you have to consider case when the system is loaded. I changed L , from 1.1uH to 3.3uH because calculating Lmin at 1MHz, I note that 30% DELTA IL needs an inductance value more than 2.6uH . I don't understand what's happen: my circuit is the same as simulation with Your web tool. Is it a problem for this chip , to work at 1.0V or 0.8V (at 0.8V I have the same results !!) ? Thanks for Your time

0 JohnTucker over 6 years ago in reply to GIUSEPPE FERRARO

TI__Guru**** 191445 points

Your first waveforms sho an output voltage with no switching. That means either your high side FET is damaged or you are getting a pre-bias voltage from another rail. Can you completely disconnect the output from everything else on your board to see if it will start? I will order an evaluation board to test on my end. It will probably be late this week or early next week before it arrives.

0 JohnTucker over 6 years ago in reply to JohnTucker

TI__Guru**** 191445 points

Also, can you post your actual schematic?

0 GIUSEPPE FERRARO over 6 years ago in reply to JohnTucker

Prodigy 140 points

Hi, output is isolated completely in my first measure. And with load You have seen the results. I think that with 1Vout there are problems because I can see similar waveform situation in other 4 DCDC that run correctly. I use this device in a power chain for XILINX ZYNQ 7030 device. This is 1V core power supply and must be not more than 1V. This power supply is the first switch on in the chain so... I can't use ZYNQ because this PGOOD don't run the second DCDC and so on. i'M IN TROUBLE !! Regards

0 GIUSEPPE FERRARO over 6 years ago in reply to JohnTucker

Prodigy 140 points

Sorry I can't. But, believe me, is the same of simulation that I sent You in previous my post, no differences. I have 4 DCDC with the same schematic diagram, little different values, obvious, for different Vout. Thanks for Your time. Regards

0 JohnTucker over 6 years ago in reply to GIUSEPPE FERRARO

TI__Guru**** 191445 points

OK, when I get the eval board, I will modify it for 1.0 V operation and ensure that it works properly.

0 GIUSEPPE FERRARO over 6 years ago in reply to JohnTucker

Prodigy 140 points

I think so but my question is: what can I do to run this circuit correctly ? Because I haven't a solution: I changed a lot of components , in particular capacitor, that are unreadable as values, to be sure that wasn't a reel error in mount operations. Nothing change.
If You have some ideas, don't hesitate. Thanks a lot for Your interesting in this question.
Regards

0 JohnTucker over 6 years ago in reply to GIUSEPPE FERRARO

TI__Guru**** 191445 points

I was able to simulate this and it did converge to 1.0 V output. I would recommend repopulating your circuit with all the original Webench values (which should work). I have ordered an EVM, but it will be a couple days before it arrives for me to test. Let me know if repopulating your board solves your issue.

0 GIUSEPPE FERRARO over 6 years ago in reply to JohnTucker

Prodigy 140 points

Hi,

thanks a lot for Your time and answers.

My circuit was correct, simulation and circuit was the same thing.

The problem was that, for a PCB problem (!) other DCDC 1.2V output was connected to 1.0V output and this is the problem, as You told me in previous post.

To understand this question was not very simple, in a particular pcb situation.

LM20144 run correctly also at 1Vout 2A, very well , I can say now.

Thanks a lot for our help and sorry but wasn't a my problem.

Regards

Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

Power management

Power management forum

LM20144-Q1: LM20144-Q1 DC/DC 1,0Vout