REF5025: REF5025

Hi Robert,

Can you share the schematic?

Have you taken scope shots of the output?

AC coupled, no oscillation from regulator

Let's try this again, Doesn't look like I can copy and paste pictures here, so I'll attach a fileRespond to TI.pdf

Hi Robert,

25mV is very unusual.

I would recommend to swap the REF5025 to see if it is that device that is causing the issues.

If it is not then we have to isolate the device by removing the loads. Is it possible to remove FB2 and remeasure?

Are all the caps low ESR caps?

Yes with FB2 removed I still get 18 to 20 mv high on the sample board test dept sent me. It does change with heat If I heat the chip up the voltage drops , I can bring it down to the correct 2.500V by heating the case with my soldering iron. certainly seems to be greater the 3 PPM/°C

The caps are TDK,  C3216X7R1C106M160AC , CAP CER 10UF 16V X7R 20% 1206 smt

We had a batch of 100 boards made, test dept rejected about 5 for being over our test spec of 10 mv, , they really should be < 2.5mv.

Because we also use the ref in the A/D , it will be a wash for our measurements, but we are concerned because we use the same in chip in two other products.

Hi Robert,

That is highly unusual. The total error at 25C should be less than 20mV.

Is it possible to try a different capacitor just in case? I just want to cover all the bases.

I am working with my quality team to see if we can get any information on the lot.

20mv ?    with a 0.1% tolerance , the error should be down around 2.5mv, . or am I missing something?

Pin 5 measures  1.234 V, if that is a clue.

Hi Robert,

Have you performed the swap of the faulty REF50 with a known good REF50?

I want to understand if this is a device issue or an application issue.

OK, I replaced the `chip with another one (Lot code TI98) I now get 2.4988 Volts , within the ± 2.5mv spec.

OK, but I'm going to resolder the part onto a scrap board first. Make sure it wasn't just some unseen flux issue.  I did clean the board several time and looked for flux after removing chip. The board was clean and cleaning didn't change anything.

If it still 18mv high,  I'll send the chip on the board. I'm also going to have test dept replace the chips on other boards and send me the bad chips. 

OK, I resoldered chip on a perf board, with some input and output caps. I still get 2.518V . The return form is a bit difficult, We had these boards made for us in China and I don't know all of the details about their purchase. We did specify thay buy the parts from Digi-key.

I would like to send the chip soldered on the perf board with the three wires input, ground and output. I included a referance to Marcoo Z in comment section.

Hi Robert,

Start the process and someone should contact you where you can explain the issue.

It seems after an AB swap the REF50 on the board is unusually high so we need to see what happened or possibly FA.