I understand what resistors are, what they do and how to choose one(R=V/I) for say an led. Where I get confused when looking at circuits is with all these resistors is how exactly do you know which value to choose? I've noticed 220, 1k, 10k and 100k seem to be used alot. Are these like beginning values and you tweak them from there? How do you decide the right resistor when working with ICs and tying switches to them or putting this certain resistor in front of the base of a BJT or you need this certain resistor here to keep this output LOW then when current is added etc, etc, etc. This is where my head starts spinning.

Any help in explaining it so I can understand this better?

 

Remember that resistors function as current limiters and voltage dividers. The value depends therefore, on the current or voltage requirements of other circuit elements.

In a BJT switching circuit a base resistor will be chosen to limit the current to ~1/10 the collector current to assure saturation but in a BJT amplifier circuit with a voltage divider bias, resistor values will be selected to set the DC operating point.

Some useful resistor values become known from experience. For example, a 220Ω resistor will safely limit the current for a red LED running on 5V, a 10k resistor makes a weak pull up for TTL, a 1k resistor makes a strong pull up for TTL and timing resistors should usually be kept to less than 2M on a 555 timer.

 

There is always a math formula but, you have to be frankly awesome to know them all, and there are also uncertainty factors. I don't know if you are old enough to remember vertical and horizontal hold knobs on TV's. Now, those adjustments are not on the front of the tv. Why? Because the uncertainty has been taken out by better integrated chips with well defined behavior. Sometimes you are doing a new design and it's so complex that working out the math would take longer than trying a gut instinct value and adjusting the amount of resistance later.

So..there we have some examples. Uncertainty, unsophistocated design, and component tolerances cause adjustment knobs.

 

Every situation will require a specific value. You wil learn from knowledge and experience what value goes where.

The reason you see some common values is because many times the requirement is not critcal and we select what's available in our parts collection. Hence 100, 1k and 10k are quite popular.

 

Every situation will require a specific value.

Until the temperature or humidity changes, or the inductors age, or the customer wants the sound to be louder...
I say you can't always install a fixed resistor and there is only one "best" value and it never changes.

 

There is always a "Exact" resistor value that can be calculated mathematically, but sometimes this is a meaningless exercise.

Some applications will work ok with a WIDE RANGE of resistance; pullups, for example.

Other applications REQUIRE that the values are calculated exactly, as they MUST be correct for the desired function; voltage dividers, for example.

When you gain experience, you can make this call and avoid wasting time fretting over exact values when it's not critical.

This is why you find values like 10K ohms thrown around in circuit, it's better to have less parts in your bill of materials, even though the values are not "perfect" the circuit works fine and money and effort are conserved.

This is the real "optimization" that you must strive for.

 

Ahh I see, thanks for the help everyone. I assume while gaining more experience I should go higher rather lower to be safe?

 

Yes. Not enough resistance causes things to get too hot. Too much resistance just makes them not work. Then you have plenty of time to measure things and work out the right answer.

 

Ahh I see, thanks for the help everyone. I assume while gaining more experience I should go higher rather lower to be safe?

No. You cannot make a blanket statement like that about any component.

If modifying a resistance or any other parameter alters the voltage, threshold, frequency, bandwidth, pulse width, current, power, intensity, etc, etc, there is no guarantee that you are moving towards a preferable state.

 

No. You cannot make a blanket statement like that about any component.

Aren't you the person that made the blanket statement, "Every situation will require a specific value"?

Lots of respect here but, you don't seem to be your usual self today.

 

Maybe there is a breakdown in communication here.

I did not say "exact" value.

For every situation, one can determine via calculation or experimentation a preferred value.
You can also determine via calculation or experimentation the range of preferred values.

As Sensacell rightly points out, the range of acceptable values for a pullup resistor may be very WIDE. Indeed, the range of working values for a pullup resistor may span three or four orders of magnitude.

Putting it differently, you cannot use 220Ω resistors in every situation. Each specific situation calls for a specified value, 220Ω here, 2.2kΩ there etc.

and we use 2.2kΩ in that specific situation and not 2.247kΩ for practical reasons.

 

Many cases when I am designing I assume a value, and the rest follows. Experience helps a lot.

Many cases, as has been mentioned, it is not that important. You pick a value and see if it works on paper. It is all part of designing, and it is why the math is important.

 

I believe what #12 is referring to are situations where conditions are variable and hence requires one to install a variable or adjustable resistance.

In response to that, the engineering approach is to minimize the effect of all known variables and hence allow the circuit designer to use fixed resistances. If a variation is to be expected, such situations are best accommodated in software. Open up a smart phone and tell me how many trimpots you see.