Skip to main content
Robmar
Senior II
November 24, 2023
Solved

ADC DC-bias auto-adjusts above 0 volts on F4 and H7, is there any documentation?

  • November 24, 2023
  • 28 replies
  • 7033 views

Both F407 and H743 ADC input add DC-bias to AC connected inputs, very handy, but I would like to have the specifications for this feature, perhaps even a circuit diagram explaining how the MCU averages the AC input above ground as needed.

I have read the Intro to ADCs AN4861 and the datasheets but must have overlooked the relevant information.

This topic has been closed for replies.
Best answer by Robmar

No, you are the one totally misunderstanding what I wrote and  making a bunch of assumptions.

Lets stop wasting each other time on this matter, I think the situation is clear.

As the application note AN2834 for ST ADC's states, the devices have limitations, not all of which can be solved End.

28 replies

Piranha
Principal III
November 28, 2023

The necessity of a DMA at such rates is so obvious that I did not even bother mentioning it. It's not some special case, but just a normal thing to do. Only beginners can think of it as some advanced achievement. The same is true for double buffering, asynchronous processing and other common principles.

So the 47 pF capacitor is a part of an anti-aliasing filter - that's how it is called. And, of course, for the best performance, the filtered signal must not contain any frequency above the Nyquist frequency. If that is not the case, then the aliased frequencies will end up in a captured data as a... noise! Couple this analog bandwidth requirement with the impedance requirement and it becomes clear that for a high performance capturing a decent input driver is a necessity.

You still haven't explained your input driver circuit and what exactly stops you from implementing this. It's you, who is asking for a solution, but cannot even explain the problem.

Robmar
RobmarAuthor
Senior II
November 28, 2023

Sit back and meditate, I think your ego is inventing it's own story here!

Nowhere did I say "advanced case" etc., I simply said that at higher rates polling could not be used, etc.

I've been designing electronics and software since well before you were born, so why you always assume you know more seems to indicate insecurity.

Frequencies above the Nyquist limit, by the way, show has repeated energy levels in the FFT bin, not as noise".

Now kid, I have work to do, so no more banter please!