Still having fun :)
However, I've been allowing this to slop around inside my head for a while, and I haven't really found an answer. I see 0 Ohm resistors (yes, zero) all over the place. For instance there are 11 zero ohm resistors on the AVR Butterfly.. Why??
My best guess is a Poor mans fuse?
Also, Why would they tie the "aref" pin 63 to the ADC port pin PF3, couple both to ground with with some bypass caps, and call it VCP? I don't know, just seems odd..
My best guess is a Poor mans fuse?
I'd have to look at the schematics to comment on the particulars of each, but using a 0 ohm jumper is a very cost-effective "tool".
It can be used for jumpers for optional features, or optional circutiry. It can substitute for an optional non-0 ohm resistor, to have a place to put the series resistor if needed.
It will be much cheaper than a wire jumper, and perhaps easier to add/remove than a solder jumper.
Lee
I can't answer the AREF/VCP question but, I'll give my opinion and use of Zero-Ohm resistors.
1. I only make single sided PCBs for my projects. Zero-Ohm resistors make excellent jumpers but, rather then be on the non-foil side (in the form of a wire) of the PCB, they are just another component mixed in with all of the other valued components. You can typically run two 12 thousanths traces under a 1206 SMD resistor. The fact is, many PCB designers use Zero-Ohm resistors for the same reason.
2. There are times where different options on a PCB come into play, depending on the variations within a given product. Lets say, a device uses a component that drives an input to a microcontroller. Well, maybe the scaled down version doesn't require that particular component. You can't (shoulden't) leave an input to a device floating. The Zero-Ohm resistor is a convienient way to pull the un-used input to VCC or GND.
I can think of a couple of other sinerios for Zero-Ohm SMD resistors, as well but, I think this will open a field of view that will get you thinking about other practical uses of Zero-Ohm resistors.
Huh.. Well that makes sense. Best part, I've been working on my first smd board layout and was running into space issues. I thought about wire jumpers, but that meant drilling a holes. Never considered 0hm smd resistors!!
(actually I didn't think they made them until looking at the schematic.. LOL)
Very cool, you are are great, thanks!
Jumpers, and they can be placed automatically with pick and place machines.
Harvey
About VCP (from the butterfly application code):
// To save power, the voltage over the LDR and the NTC is turned off when not used
// This is done by controlling the voltage from a I/O-pin (PORTF3)
About VCP (from the butterfly application code):
Quote:// To save power, the voltage over the LDR and the NTC is turned off when not used
// This is done by controlling the voltage from a I/O-pin (PORTF3)
/Lars
However VCP also feeds the Aref pin, which means PF3 needs to be on, AND the Aref voltage suddenly depends on both the ambient lighting AND temperature. The Gcc port of the Butterfly code even talks about it. But is sounds bizarre from my novice point of view. I figured they had to be a "REASON" they went this route.
So far the only reason I can come up with is they need something to feed the Temp and Light circuits, so they used the ARef power line. Once they realized their ooopsie, the did the classic "let the guys writing the code fix it" :)
Being I would like to do some real ADC, would I be best suited to hardware hack in a new Aref signal, and power the temp / light sensors off another circuit? (power management is low on the to-do list) :)
Thanks for your time.
theusch: Well, I've also seen them (atleast I think it was that) used as a fuse in something. It was in the PSU part of the device and there was not much to jump, so I guess that was a fuse.
But not a poor mans fuse, more like a cheapskate mans fuse.
A better fuse than 0R would be something like 1R. Limits the current, and the heating will destroy the 1R resistor eventually. A 0R might not have enough resistance to heat it up before something else like a PCB trace blows up.
But mostly I use 0Rs for jumpers and optional features, and even for bypassing some circuitry that can be omitted if not needed.
- Jani
A 0R might not have enough resistance to heat it up before something else like a PCB trace blows up.
Hey, it's 0R +/- 5% ;)
Heh, I've sometimes smiled at the spec on resistor bags, indeed, there are 0R +/- 5% resistors :)
But seriously, when there is a chance a lot of current will go through 0R, better pick a 1206 than 0408 package if the voltage matters a lot. Just measure the voltage over the 0R when 100mA goes through it. There must be a maximum safe current specified in datasheets as well.
It's the same thing with PCB traces, if you know you must get 1Amp current from PCB edge to another, you don't use 8 mils track width, you use 42 mils..
- Jani
Hey, it's 0R +/- 5% ;)
0, +/-0, is still ZERO. Restated, Zero times ANYTHING is still ZERO...
But in reality, a Zero-Ohm resistor will have some small value of resistance, in the low to mid milii-Ohm range.
And in very sensitive measurements, even the voltage over a solder junction can ruin your day..
(it's like a thermocouple, two different metals attached together. PCB track is copper, it has a plating of some sort maybe, then there is solder, either lead-free or not, then there is the plating on the resistor, and then the resistor material itself. Lots of junctions between different materials.)
- Jani
Lerch wrote:0, +/-0, is still ZERO. Restated, Zero times ANYTHING is still ZERO..Hey, it's 0R +/- 5% ;)
[sarcastic humor flag]
I disagree! :roll: There's always quantum uncertainty and vacuum energy flux to be concerned about.
[/sarcastic humor flag] :)
Original post was meant to indicate a moment of levity or humor. (especially the minus 5% part)But, I must also clarify that while I find -5% of zero ohms funny, I conceed that negative resistance is real phenomenon. Ie tunneling diodes..)
[sarcastic humor flag]
I disagree! :roll: There's always quantum uncertainty and vacuum energy flux to be concerned about.
[/sarcastic humor flag] :)
:shock: :shock: :shock: :? :? :?
And, how does this apply to Zero-Ohm resistors?
Actually, looking at your avatar, I was expecting some sort of fantastic fish story here... :lol:
Original post was meant to indicate a moment of levity or humor. (especially the minus 5% part)But, I must also clarify that while I find -5% of zero ohms funny, I conceed that negative resistance is real phenomenon. Ie tunneling diodes..)
I do like humour, but I am also kinda nitpick, or whatever the appropriate word is :)
But the resistance it's not actually negative on tunneling diodes, or is it? The resistance is always positive, but the curve has a part where the slope is negative. So when the tunneling diode is used at certain DC bias point in the middle of the falling slope, and AC voltage is applied over it, it only seems to have negative resistance :)
- Jani
Lerch wrote:[sarcastic humor flag]
I disagree! :roll: There's always quantum uncertainty and vacuum energy flux to be concerned about.
[/sarcastic humor flag] :):shock: :shock: :shock: :? :? :?
And, how does this apply to Zero-Ohm resistors?
Actually, looking at your avatar, I was expecting some sort of fantastic fish story here... :lol:
/* written as an aside to the audience}
OK, boy have I got myself into a corner now. How to weave a tale involving quantum uncertainty, vacuum energy flux and 0ohm resistors :{we need a scratch head emoticon}
end aside */
Quantum uncertainty involves the startling idea that measurements repeated multiple times will produce different result each and every time (not because the measurement was in error, but because it is the nature of things). The only thing we know with certainty is the results will fall into a probability distribution. So saying a we have a 0ohm +/- 5% resistor clearly reduces to a single, non probabilistic result. Since the known physical universe is in its very nature indeterminant, claiming to have a 0ohm +/-5% resistor is highly improbable! :)
With regards to Vacuum energy flux, a branch of modern physics postulates that the spontaneous creation of a particle + anti-particle pair happens continually (and is claimed to be the reason black holes evaporate), also they claim it is possible to violate the law of energy conservation, but only for periods of time inversely proportional to the quanta of energy borrowed. Since this vacuum energy flux can happen anywhere, it there for happens inside our 0ohm +/- 5% resistor as well. Clearly, this is something to take into consideration when claiming to posses a 0ohm +/-5% resistor :)
And no, I'm not smart enough to take this debate any further, and hence forth concede victory to whom so ever is need of being a victor :)
/* all this over a 0ohm resistor */ 8)
But, I must also clarify that while I find -5% of zero ohms funny,
Quote:
But, I must also clarify that while I find -5% of zero ohms funny,
So, in a high-precision app, does one need to pay extra for a 0ohm +/-1% resistor versus a "common" +/-5% resistor?
High-Precision applicatons, not so much.
However, if your building Audiphool equipment, the 0ohm 1% devices will suffice, as long as they are connected with gold plated oxygen free cryogenic treated wires, and only driven by tube amps connected to the utility grid by $300 power cables. :roll:
But the resistance it's not actually negative on tunneling diodes, or is it? The resistance is always positive, but the curve has a part where the slope is negative. So when the tunneling diode is used at certain DC bias point in the middle of the falling slope, and AC voltage is applied over it, it only seems to have negative resistance :)
- Jani
I was under the (possible wrong) impresion that a properly biased tunneling diode would actually amplify an ac single couple to it..
For example:
http://www.tpub.com/neets/book11...
Above example it's self being a worthy topic of discussion :)
It's only about the slope of the negative resistance part, where the change of voltage vs change of current is negative (dV/dA). For example if the slope is -1, then if you feed 1 mVRMS AC over the DC bias point, then you get 1mVRMS out. If the slope is -0.5, then with 1mV input you get 0.5mV output. Likewise with -2 slope, with 1mV input you get 2mV output.
So yes it basically amplifies it (and you get negative gain), but it uses the external DC bias point as the "lever" so to speak, it gets power for the amplification there.
- Jani
However, if your building Audiphool equipment, the 0ohm 1% devices will suffice, as long as they are connected with gold plated oxygen free cryogenic treated wires, and only driven by tube amps connected to the utility grid by $300 power cables. Rolling Eyes
Ha, yeah those guys are insane. diyaudio.com is full of them. Some even buy $1200 power cables to connect to the stuff in the walls that costs $0.10/ft!
I hope they have good cables too. You know, the ones that are hand-braided by nude blonde virgins from pure unobtainium. There is a sucker, uh, I mean a customer born every minute.
- Jani
Lajon wrote:I read that, and they seem to turn it off by toggling pin PF3, which is fine and dandy.About VCP (from the butterfly application code):
Quote:// To save power, the voltage over the LDR and the NTC is turned off when not used
// This is done by controlling the voltage from a I/O-pin (PORTF3)
/LarsHowever VCP also feeds the Aref pin, which means PF3 needs to be on, AND the Aref voltage suddenly depends on both the ambient lighting AND temperature.
So you figure the changing load on the PF3 pin would change AREF significantly. I hadn't thought of that.
You don't need to worry about the light sensor circuit. The boards no longer have the light sensor installed. I guess Europe banned cadmium.
As far as the temperature circuit is concerned, I recommend you keep it in a constant temperature environment :lol: Otherwise it's easy to remove a resistor from that circuit to disable it.
As has been stated, 5% of zero is still zero. So any deviation from zero would be an infinite percent deviation.
As all 0 ohm resistors have some resistance unless they are superconductors, then surely the only correct specification for them is + or - infinite percent.
Or is my math faulty?
By the way, I thought copper traces were poor man's fuses.
As all 0 ohm resistors have some resistance unless they are superconductors, then surely the only correct specification for them is + or - infinite percent.Or is my math faulty?
D'Israeli goes on to say, "The reader desirous of being merry with Aquinas's angels may find them in Martinus Scriblerus, in Ch. VII who inquires if angels pass from one extreme to another without going through the middle? And if angels know things more clearly in a morning? How many angels can dance on the point of a very fine needle, without jostling one another?"
No faulty math. Divine resistors.
[edit] Or maybe the math >>is<< faulty. Either that or Ohm's law no longer holds.
I = V/R, right? If R is indeed 0, then an infinite amount of current can pass at any voltage. So much for "poor man's fuses". Either that, or THERE IS NO SUCH THING AS A 0OHM RESISTOR and this thread doesn't really exist.
Lee
Thanks for the quote. I, like many illiterates, thought it was "dance on the head of a pin".
I used to think it was all about silliness until a mathematician friend corrected me. Apparently it was a serious mathematical study. About infinity and infinitesimals, I suppose.
Actually, I thought it was as well. ;) Here is the link I used; the guy >>seemed<< to know what he was talking about:
http://www.straightdope.com/clas...
Let's get a couple things straight. First, you're misquoting the saying in question. According to unimpeachable sources, it's not how many angels can dance on the head of a pin, it's how many can do it on the point of a needle--which, of course, makes more sense. Second, the earliest citation I can find is from a book by Ralph Cudworth in the 17th century, which is a suspiciously late in the day.
which, of course, makes more sense.
Lee
With a little effort I've managed to build my own 0 V batteries. Works pretty well, and they have very low internal resistance too.
One nice thing is that when you work with such low voltages, observing polarity doesn't seem to matter so much.
With a little effort I've managed to build my own 0 V batteries. Works pretty well, and they have very low internal resistance too.
If that device in the right in that battery pack is what I think it is, that battery pack will have a very high internal resistance...
Of course... :lol: Zero times anything is still zero.
The one that confuses most people is:
0
-- = undeterminable
0
It's a which came first - the chicken, or the egg???
With a little effort I've managed to build my own 0 V batteries. Works pretty well, and they have very low internal resistance too.One nice thing is that when you work with such low voltages, observing polarity doesn't seem to matter so much.
Lajon wrote:However VCP also feeds the Aref pin, which means PF3 needs to be on, AND the Aref voltage suddenly depends on both the ambient lighting AND temperature. The Gcc port of the Butterfly code even talks about it. But is sounds bizarre from my novice point of view. I figured they had to be a "REASON" they went this route.So far the only reason I can come up with is they need something to feed the Temp and Light circuits, so they used the ARef power line. Once they realized their ooopsie, the did the classic "let the guys writing the code fix it" :)
I hope I will be excused for getting back on topic for a moment :)
I think the circuit used by the Butterfly for measuring light and temperature is precisely correct. Those sensors are variable resistors that are in series with known resistors. The ADC measures the ratio of the voltage at the junction to the voltage across both resistors. It makes no difference what the "reference" voltage is. It could even vary over time as long as it doesn't change much while the conversion is happening. The only thing that matters is the "reference" voltage be the voltage applied across both resistors in series.
Of course if your sensors work differently then you would probably need to know what the reference voltage is. In that case you could use another digital output pin if you run from batteries or get it somewhere else if you don't need to switch it off when not in use.
So I had a quick peek at the schematic for the butterfly, and I actually think the design was intentional, even though it seems strange.
The temperature measurement and the light measurement are both analog, thus uses the ADC. If the Aref voltage should vary, the supply to the measurements would also vary. This would (ideally) mean that the error would cancel itself out!
All right, so I'll explain my thinking a bit more. The measurements are voltage dividers. For simplicity, let's say both the fixed resistor and the variable resistor have the same value when we're measuring. That would mean we're measuring half the voltage of Aref. The ADC uses Aref to do the conversion, and gives its result in parts (you can convert this to percent with a quick bit of math) of Aref. So, in the example above, we would get a reading of "50 % of Aref". Now, if Aref changes, the conversion would still net "50 % of Aref", since the resistors are also connected to Aref.
I might just have made things less clear than they were before. I'm sorry.
Also, looking at the resistance values used, the current drawn won't be very big. Maximum values are less than 1 mA for the light measurement, and just a few µA for the temperature measurement. That is probably not enough for the voltage to vary noticeably.
And on the 0 Ohm discussion:
No, there is no such thing as a 0 Ohm resistor. There are supraconducting materials which have no resistance, but none avalible at room temperature (last I checked at least). The 0 Ohm resistors will have a small resistance, just like the copper traces on a PCB and the connecting leads to the components. But this resistance is really, really low, probably around 1 milliOhm, and is almost always ignored. Ohms "law" is correct, at least for this case. There are some exceptions to it, but that are really special cases (very high frequency or really small dimensions, if I recall correctly).
