EN6337QA Datasheet, feedback pin voltage variations.

[MBaqu]

In EN6337QA Datasheet, we have two informations about feedback pin voltage variations.

First range is between 0.7425 and 0.7575 at 0 load 5V input and 25°C.

Is it process variations?

Second range is between 0.735V and 0.765V including temperature, Vin and load variations.

Is this range include also the first variations or we have to add it?

Thanks

Regards

[Mostafa_Intel_AE]

Hello Matthieu,

The first range is the voltage accuracy when there is no circuit variation which you can consider as a process variation.

The second range consider the voltage accuracy when the whole expected circuit changes happen and that range second range include the first range so no need to added it again.

So sorry if that confuse you.

Thanks,

Mostafa

[MBaqu]

Hello Mostafa,

Is it possible to have some curves about the variation of the feedback pin voltage with temperature and aging?

I have the same request for EN63A0QA

Thanks

[Mostafa_Intel_AE · Accepted solution]

Hello Matthieu,

Below is the required charts, by the way it is for Vout normalized which will be equivalent to the VFB accuracy .

For EN6337

For EN63A0

Thanks,

Mostafa

[MBaqu]

Hello Mostafa,

Thanks a lot to share these datas. Also, do you have same data (feedback variation) with aging?

Regards

[MBaqu]

Hello Mostafa,

I'm a little confused with the curves. It seems that temperature variation is about 1% and load variation is also 1%.

But if I search the point at 25°C without load and Vin=5V to check the +/-6mV (1%) given in the datasheet, I find more than 1% of variation in the curves.

So it seems that we need to add 1% (temp), 1% (load) and more than 1% (process?) to have the total tolerance?

What I'm missing?

Thanks

Regards

[Mostafa_Intel_AE]

Hello Matthieu,

I am really sorry if i confused you.

But the datasheet has two tolerances for the VFB as shown below:

1. VFB tolerance at only one single point without the affect of the temp and it is +/-1%, and it think this is the one confused you.
2. The other VFB tolerance with all the possible variation and the tolerance is +/-2% or we can say the total tolerance is 4%

And based on the above curves we can get a conclusion that the total tolerance is 3% ( equal to the highest tolerance {1.02} - lowest value {0.99}))

Thanks,

Mostafa

[MBaqu]

Hello Mostafa,

Some points I need to clarify. If I extrapole the curves from the point 25°C, 5V Vin, 1V Vout in order to find the same point 1 as in the DS, I find more than 1% of variation.

Is this variation due to the feedback leakage current through the feedback resistors? I assume that curves result from a test with a feedback resistor divider?

Also, is it possible to have the curves at 2.5V, they aren't displayed?

And what is the variation due to load. By looking the Datasheet, I can say it is 12mV (total tolerance) - 6mV (process variation) - 4mV (temperature variation from curve temperature of datasheet p12) so result is 2mV.

But with the curves you provide, it seems it is more : 1% for 1V so 10mV on Vout or 6mV on Vref.

Thanks

[Mostafa_Intel_AE]

Hello Matthieu,

The curves that i showed before have the worst case and it is expected to be lower accuracy than the datasheet curves, but it will be within the datasheet electrical characteristic table VFB range as i explained in my previous response.

Also please note that the Vout will be shifted up or down based on how accurate Ra and Rb ratio which will effect the normalized values in the previous shared curves.

if you ignore the voltage shifting at 25C you can notice that the total accuracy within 1% over the Vout and the load which match the datasheet spec.

Also the below curve is for Vout=1V with many samples and you can notice the total accuracy within 1% over Vin and the load

Also below Vout=2.5V accuracy data at 25C, but unfortunately can't find the data over temp.

 

Thanks,

Mostafa

[MBaqu]

Hello Mostafa,

Thanks again for your answers.

I joined the worst case curves with 3 virtual points added. I add a first point with -1% from ideal 1V output, a second point with temperature variation and the last point with load variation added.

In this case, the total tolerance would be more than +/-2% (in my example around -2.5%)

I realize that it's a wost case and you nerver measure that in your tests. But I would like to be sure that I well understand the data.

So, could you comment my interpretation?

Thanks

Matthieu

[Mostafa_Intel_AE]

Hello Matthieu,

Yes, you understand the data correctly.

BTW, the worst case will be +/-2% as shown in the electrical characteristic table.

Thanks,

Mostafa

[MBaqu]

Hello Mostafa,

Thanks a lot for your clarification.

Regards

Matthieu B

[Mostafa_Intel_AE]

Hello Matthieu,

Unfortunately we don't have the measurement through the VFB directly to avoid the noise injection to the controller loop and this could lead for unstable device.

So the normalized Vout measurement should be the same as VFB accuracy (VFB accuracy will be better where the measurement done with the feedback resistor accuracy effect included).

Thanks,

Mostafa

[MBaqu]

Hello Mostafa,

My last request is related to the aging of the component. Your measurement of Vout is fine for me but I just want to know if you have the same datas but not in function of temprature but aging (ie: what is the value after 2000Hrs or more).

Thanks

Matthieu

[MBaqu]

Hello Mostafa,

Do you have same news about aging?

Thanks

Regards

[Mostafa_Intel_AE]

Hello Matthieu,

So sorry for late response, i was checking internally about that kind of data, but the feedback is that test done during the reliability tests like Vref drifting test and the test results will be pass or fail based on the datasheet's characteristic table. so unfortunately we don't have a waveforms for the Vout accuracy during that reliability test to share it.

Thanks,

Mostafa

[MBaqu]

Hello Mostafa,

Thanks for your feedback.

I have again a question about Ivfb. Is the -10nA / +10nA specification is guaranted including temperature variations?

Thanks

Matthieu

[Mostafa_Intel_AE]

Hello Matthieu,

The feedback pin input leakage current is guaranteed by design and it is not coming from production tests.

That numbers based on the simulation analysis and cover the changes over temp variation.

Thanks,

Mostafa