How to control SRC4192 in hardware mode? Do i need a pull up or pull down resisiter?

[benjaminre]

hello, everyone.
i am new here and i want to know about how to control the src4192 in hardware mode?

The control signal of 0 or 1 means low and high voltage. but do i just connect the pins to Vcc and GND or do i need a pull up or pull down resistor? if a res is needed, how much value do it need? 10k or 47k or what?

thank you

[SoNic_real_one]

"1" means direct connection to the 3.3V. "0" means direct connection to ground.

[benjaminre]

Originally Posted by SoNic_real_one

"1" means direct connection to the 3.3V. "0" means direct connection to ground.

thank you a lot for answering me. but I also see that the wm8741 use a 10k pull up or pull down res to control 1 or 0. but I could find no information in the 4192 pdf. do you ever try it ? I am afraid that I might bridge the ic.

thank you .

[5th element]

You can generally connect any logic input pin straight to the positive supply, or to ground, but there are a couple of reasons one could use a pull up resistor.

The first reason is that placing a 10k resistor between the positive supply and the pin limits the current that can flow through the pin. If any fault condition were to be present that would cause a large amount of current to flow, the pull up resistor could potentially stop the chip from frying. Similarly using a pull down resistor will also do a similar thing, that is if a large amount of current, for whatever reason, wants to flow from the input pin to ground.

The second is that the resistor will isolate either the 3.3v supply from the input pin, or isolate the pin from the supply. This will help prevent noise from either one from coupling into the other.

Of course this overlooks the main reason for one using a pull-up resistor and it's in case you want to change the state of the pin. As the 3.3 volts goes through the 10k resistor before being connected to the input pin, you are now free to connect the input pin to ground without anything bad happening. This changes the state on the input pin from 1 to 0 and in doing so simply causes 330u amps to flow through the resistor to ground. If you hadn't built the thing with the pull up resistor in you'd have to slice up the PCB trace etc, which gets ugly real fast. A pull down resistor accomplishes the same thing.

[SoNic_real_one]

You forgot that a resistor makes that input prone to outside induced noise due to higher impedance.
it's an input, if anything "goes wrong" with that to the point that needs limiting of the current, then that cip is fryed already.

[5th element]

Originally Posted by SoNic_real_one

You forgot that a resistor makes that input prone to outside induced noise due to higher impedance.

Hmm, I hadn't considered that. Usually though I don't think the state would ever 'flip' because of this, but that's not perhaps your point. I do know that states can flip on micro controller inputs if they are left floating, another good reason to use a pull down resistor if you want the micro to do something an a positive input signal.

it's an input, if anything "goes wrong" with that to the point that needs limiting of the current, then that cip is fryed already.

I agree with you completely about this, it's just nice to know sometimes. The micro controller thing does bring up something else however, where the general purpose pins can be configured as either inputs or outputs and then tied to various different internal peripherals. Now due to some internal system error, perhaps the chip gets configured in the wrong way and it could cause a pin to sink too much current, blowing something up, the pull up resistor would prevent this from occurring, but it would be very rare. Some audio ICs do have multi purpose pins depending on the chip configuration.

[SoNic_real_one]

That's different, the multifunction pins. In this care there is no need, I saw schematics using SRC4192 and they don't have resistors.

[5th element]

Indeed, neither does mine using the SRC4192, I was just pointing out where it may be applicable.

[kevinkr]

Resistors are generally considered good design practice if for no other reason than it permits easy ICT testing of circuit boards, they aren't strictly necessary but in the case of GPIO which may be configured (inadvertently in some cases) as either input or output the resistor is a good idea to limit current. Values as small as 330 ohms (or less) may be used effectively and have no practical impact on noise immunity.

Just because an application note for the SRC4192 doesn't show them doesn't mean that they should not be applied if the designer so chooses or corporate design guidelines require them. (I've worked in a number of places where they were required without exception.)

I would be likely to use them particularly if I thought I might want to change configurations later - if this is really likely a dip switch might be a good idea.