Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
Morning CBB
Had good day out yesterday but radio check wasn't that great, line of sight at best, 1 mile with 4w radio over brow of hill or just about 2 miles along top of hill
back to what we were talking about, R in series is adding more R
R in parallel is divided, I see this bit clearly now
Just working on KISS's calculation in post #411, I'm seeing of you had equal R & V they would cancel each other out so you would get no I, so I'm trying to work out what you are trying to show me
So if I had 10V & 10ohm but the 10ohm were 2x 5ohm R in parallel, I would get less R even though values are the same because the R is halved by division, hence, allowing power to flow
On the frequency front, I can see where I was misunderstanding, I was thinking if 1A = 1Hz how do they get 1MHz, where do they get all that power from, but it works backwards doesn't it, as R increases so does frequency, so we don't need more power for a higher frequency, we actually need less but at the same time as we lift frequency we are also increasing R against I making it take longer, so this has something to do with charge times, where 't' comes into play
So now I'm working on the overall picture still & somehow I need to understand the link between it all, I can see the logic for the LOG graph, due to the small amounts & as we are working backwards, the graph is somehow the oppisite side of 0 but the right side of 0 too?
No, it must be above & below for that comparision to work, so somehow frequency is a hidden part of I
I can almost visualize it in front of me, so the longer it takes, the less time the cap has to charge, so the higher the frequency the lower I due to time taken
That's it, I can cleary see whats happening now if thats right so far
How we work it out, well thats another story that I may need your help with **broken link removed** least I can see whats going on now
Hope you enjoyed your day off from me, now back to all these dumb Q's I keep asking **broken link removed**
The -3 db rule is sort of an average for AC = rms/equivilent to DC
I think I'm starting to get the picture now, a cap resists instant change, therefore it takes time to build the charge, so there are lots of different uses
I can see the filtering function, I'm starting to see the frequency function but I need to nail this down really to understand exactly what we are looking at, so frequency affects the output V via time which would alter the response time to anything on the end of it, i.e. phase, so if you wanted one part of the circuit to start before the other, you could use it like this or it could give you a different waveband etc etc as KISS was saying a variety of jobs are achievable, I just need to understand it now
So by being able to work out frequency gives us the conclusion for work available (VP) on the line should it be required but in a miniscule form
(Question for later while I remember it, does that mean an inductor would do the same thing for 'I' giving us delayed Amps available, only inductors collapse when powered off unlike caps)
Just working on KISS's calculation in post #411, I'm seeing of you had equal R & V they would cancel each other out so you would get no I, so I'm trying to work out what you are trying to show me
The formulas in #411 are not relevant to the task at hand (parallel resistors). They are relevant to circuits with "reactance" so disregard them for the time being.
So if I had 10V & 10ohm but the 10ohm were 2x 5ohm R in parallel, I would get less R even though values are the same because the R is halved by division, hence, allowing power to flow
To solve for the equivalent resistance for this circuit, we use the following formula (LateX isn't working):
R1 X R2 divided by R1 + R2. This can also be written: (R1*R2)/(R1+R2).
So, in this case: (10*40)/(10+40), or 400/50, which equals 8.
So in words, the equivalent resistance of a parallel, 2 resistor circuit consisting of a 10Ω and 40Ω resistors is 8Ω.
Try this in TINA with different value resistors, do the math and then confirm your answer with the multimeter.
It isn't so important to know why the formula is used as it is to realize that this is how it's done. And it works.
You might ask: "Why not use an 8Ω resistor, instead"? Well, when you can't find a certain value resistor (and an 8Ω resistor IS NOT a normal value [and neither is a 40, for that matter]), it's OK to create an "equivalent" value "network" circuit of multiple resistors to achieve that.
And since we've been investigating RC timing circuits, for a very specific RC time constant we might have to do just that.
One last example. Let's say you wanted an 18Ω resistor (also an odd value, as is 40Ω, but I wanted to keep the math simple). This is how to get there:
In other words, an equivalent 8 ohm parallel circuit (composed of 10 and 40 ohm resistors) in series with an 10 ohm resistor.
On the frequency front, I can see where I was misunderstanding, I was thinking if 1A = 1Hz how do they get 1MHz, where do they get all that power from, but it works backwards doesn't it, as R increases so does frequency, so we don't need more power for a higher frequency, we actually need less but at the same time as we lift frequency we are also increasing R against I making it take longer, so this has something to do with charge times, where 't' comes into play
To solve for the equivalent resistance for this circuit, we use the following formula (LateX isn't working):
R1 X R2 divided by R1 + R2. This can also be written: (R1*R2)/(R1+R2).
So, in this case: (10*40)/(10+40), or 400/50, which equals 8.
So in words, the equivalent resistance of a parallel, 2 resistor circuit consisting of a 10Ω and 40Ω resistors is 8Ω.
It isn't so important to know why the formula is used as it is to realize that this is how it's done.
You might ask: "Why not use an 8Ω resistor, instead"? Well, when you can't find a certain value resistor (and an 8Ω resistor IS NOT a normal value [and neither is a 40, for that matter]), it's OK to create an "equivalent" value "network" circuit of multiple resistors to achieve that.
And since we've been investigating RC timing circuits, for a very specific RC time constant we might have to do just that.
One last example. Let's say you wanted an 18Ω resistor (also an odd value, as is 40Ω, but I wanted to keep the math simple). This is how to get there:
Now RC circuits. I'm going to build (edit) this post on the fly. When I'm through, I'll indicate that at the end of the post.
Basic 10VDC powered, 1uF cap plus 1MΩ resistor (RC time constant of 1 second), then a 1uF cap with a 500kΩ resistor (RC time constant of 500m seconds):
The RC time constant is defined as "that point in the rise of the voltage across the capacitor where it reaches two-thirds (2/3, or 0.633 times) of the value of the power source (battery).
Note that the V value on the graph is 6.33 VDC (or 2/3 of the battery's 10VDC value) at that point where the rise time value is 1 second for the 1MΩ resistor, and 1/2 second (500ms) for the 500kΩ resistor.
Now these examples. The only difference is that the battery level has been raised to 20VDC.
Notice how the RC time constant is identical to the 10VDC battery examples. The voltage level at which the RC time constant is reached is higher, to be sure, but the TIME that it took to get there is unchanged.
This is why it is called an RC time constant.
In other words, the voltage level of the power source (battery) has absolutely no effect on the RC time constant value.
Now for fun, what I'll say is that for capacitors in parallel the formula becomes: Cp = C1+C2+C3+...Cn. So a 10uf capacitor in parallel with a 10 uF capacitor is 20 uf
and, not surprising for capacitors in series: 1/Cs=1/C1+1/C2+1/C3+....1/Cn
The formulas for R in series and parallel are just the opposite of C in series and parallel.
ok, that took an age but you are basically dividing V by A for ohms
however, the time it took to figure out why two 25ohm R's 400/50 = 12ohms
wouldn't work was because you need to be able to times V by ohms to be equal to the two R's
so if you work out you want a 8 Ohm resistor via 2 in parallel how are you working it backwards which is what I've been trying to do
Edit: it doesn't matter how I try, I simply cannot work backwards, so me knowing I need a 8ohm R wouldn't be a lot of use even knowing V & I, I can work back as far as needing two R's split but then get lost
The only thing I can conclude from today is that if you already have two resistors in parallel that you can work out V & I but not actually work it back to work out what actual resistors you need & R reading is fixed whatever V you have, changing V only alters I
Rp means the resistance of the parallel combination.
The other gibberish is known as an infinite series, but you can solve it for any number of resistors (n) in parallel.
So if there are two, 10 ohm resistors.
1/Rp = 1/10+1/10
1/Rp = 2/10
Rp = 10/2 or 5
I solved it using fractions, you can solve it using decimals too.
1/Rp = 0.1+0.1
1/Rp = 0.2
Rp = 1/0.2
Rp =5
The formula would work for 100 resistors in parallel and 1,000,000 resistors in parallel. It doesn't matter. It will even work for 1.
You learned a while ago that R*C has the units of ohms * farads (Usually written as (ohms- Farads) and that reduced to the units of seconds (s). I mistakenly used t. Units that are multiplied together show up as what looks like subtraction. (ohm-cm) is one such unit.
That seems long winded, why wouldn't you just add them together & divide by the amount **broken link removed** or is it that you have kept it simple for me & having different value R's would give varying amounts each side depending on value
You learned a while ago that R*C has the units of ohms * farads (Usually written as (ohms- Farads) and that reduced to the units of seconds (s). I mistakenly used t. Units that are multiplied together show up as what looks like subtraction. (ohm-cm) is one such unit.
The length of one cycle is called the period and it's measured in time units, like 20 mS. 1/(period) is the frequency. So 1/20 mS is 50 Hz. See if you can get that answer on your calculator?
Capacitors change their behavior based on frequency. When capacitors are part of a timing circuit, the quantity R*C is usually involved. That quantity is called the TIME CONSTANT.
So ignoring V for a moment, I can see the time taken by R & cap is equal, so these are setting for the charge time & nothing else... so I was right in thinking R was the control of the cap, I just couldn't relate it to time
I think I'm starting to get the picture now, a cap resists instant change, therefore it takes time to build the charge, so there are lots of different uses
Yes. And it takes the time dictated by the value of the resistor times the value of the cap to arrive at the RC time constant.
And notice KISS's note on parallel capacitor determination formulae.
So, now, we'll go on to a circuit and show how the RC constant is used to determine the timing of an event.
In this example, I'm using a simple, triggered "pulse generating" circuit that demonstrates how an RC circuit can control an event (in this case, the length of the pulse).
Most of the circuit is hidden, but the essential portion (the square positive pulse making circuit) that is relevant to this discussion is shown:
Note that, though not exactly 1 second, the pulse is close (2.5s to 3.75s, or a1.25s pulse).
Why not exactly 1s, as the RC formula predicts?
Well, here the planning world hits smack up against the real "let's make a circuit that does something" world.
The OpAmp has internal resistances (and capacitances as well). These values interact with the RC circuit and alter it.
But all is not lost. Here is how you "adjust" for these anomalies:
**broken link removed**
Note that I reduced R1 to 900kΩ. That sufficiently reduced the RC value to come into line with the desired 1s pulse width, given the unseen values within the OpAmp that needed to be allowed for.
Messy? Yes.
But then, most analog (as opposed to digital) electronics is a bit messy. Often, in order to get a circuit to work as you wish, compromises are made.
I'm going to post a simple pulse generating circuit later that you can alter to demonstrate how an RC circuit affects timing.
It's taking me some time to do it because there are a lot more set-up parameters that are needed to make it work. Not a lot of parts, but those few parts must be correctly configured.
You need to be able to keep this rule firmly set in your mind: under normal circumstances, neither Voltage (V), nor Current (I) can or will alter resistance (R). NEVER.
But the reverse IS true: resistance WILL alter both current AND Voltage.
Hiya CBB
I totally get that bit, haven't read the rest yet, but rest assured I can see the bit you have shown me as plain as day **broken link removed**
I think it's the time thing & frequency I'm struggling with
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
