Pondering Power - resurrected
Thread Starter
Honorary Member
Joined: May 2001
Posts: 2,121
Likes: 3
From: Indian Land SC
jzr - You're concern for others is refreshing. You are welcome in my neighborhood anytime. 
And thank you for not mis-reading my comments as a personal bash. I was concerned after I posted it with the 
., thought you might take it the wrong way.
And thank you for not mis-reading my comments as a personal bash. I was concerned after I posted it with the ., thought you might take it the wrong way.
Community Organizer
Joined: Oct 2000
Posts: 8,642
Likes: 143
From: Chapel Hill, NC
Originally posted by modifry
[B]Yes, a sub is a motor. Yes, a sub is not a constant 4 ohms, as you would have seen I was aware of if you read my previous posts in this thread. That is EXACTLY why you should use a resistor to test with rather than a sub, if you're trying to establish any kind of general rule of thumb or establish any standards. Speakers vary way too much to use a speaker as a reference. Anybody else on this board can conduct a test with their amp and a 4 ohm resistor and compare the results to my test. We then know that the difference is due to the amp, not the load it's driving.
[B]Yes, a sub is a motor. Yes, a sub is not a constant 4 ohms, as you would have seen I was aware of if you read my previous posts in this thread. That is EXACTLY why you should use a resistor to test with rather than a sub, if you're trying to establish any kind of general rule of thumb or establish any standards. Speakers vary way too much to use a speaker as a reference. Anybody else on this board can conduct a test with their amp and a 4 ohm resistor and compare the results to my test. We then know that the difference is due to the amp, not the load it's driving.
Do you mean the load is lower or the ohms are lower? These are opposing factors. If you mean the ohms are lower, this is not 100% correct. My concern is that people sometimes make one observation (speaker moves more) and make an assumption about something else (lower ohm load). You have to look at the impedance curve of the speaker/enclosure to know, or measure excursion and impedance directly. In general, it's actually just the opposite - lower frequency = higher impedance.
If you meant that lower frequency = higher impedance (less load on the amp), that is something I can agree with.
If you meant that lower frequency = higher impedance (less load on the amp), that is something I can agree with.
Good low-bass drivers have a low Fs, (resonance frequency), usually in the 20 - 30 Hz range. This means the impedance (ohms) will be HIGHEST at the frequency of resonance (look at the impedance curve supplied with the driver). Resonance is also where the speaker is most efficient - most output (dB) for input (watts). So as the frequency drops through the audible range, the speaker cones moves farther because it is approaching resonance and efficiency is increasing. BUT - the impedance is rising, so the speaker is drawing less current as the frequency drops and you approach resonance. Below resonance, cone excursion will rapidly decrease as the frequency drops.
All this changes with different cabinet volume, porting, stuffing, even shape of the box. Which is why you can't establish any reference point or standard by using a speaker. I would be very surprised if any amplifier manufacturer rates power output and distortion figures with the amp driving a speaker. Even though using a resistive load is not the same as a speaker, it's the only way to compare one amp with another.
[quote]
I'm sure you've never seen an amp rated "500 watts with G4LNB battery and PRG57 alternator". They are rated at a specific voltage input (13.8 volts), with the assumption that the power source can supply unlimited current at that voltage. This is not the real world, as you know. Input voltage will be constantly changing as the alternator rpm changes. But it is repeatable by any manufacturer or test lab, and gives the end user an idea of how the amp will perform. It appears you readily accept manufacturers ratings based on "standardized non-real-world units" (like 13.8 volts input or 100 watts into 4 ohms), but don't like my use of those same standards.
In reference to "instaneous bass hit"
We might be on different tracks here. When you mentioned "instaneous bass hit" I thought you were talking about rise time of the waveform. Your question about "how long does the average bass note last" makes me think you're talking about a sustained high output requirement. That's what I tested, only to an extreme. My test was at a constant 40 Hz bass note of unlimited duration, which I assummed would put the most strain on the amp. The power source would drop and stay down because the bass note is constant. But then the capacitor stays down too, because there is no pause between notes to allow the cap to recover.
[quote]
This is getting way out of hand and over my head again with the EE crap that everybody throws out that very few understand.
A cap is used to keep voltage constant and not drop on a bass hit. Sustaining a note like you said will totally drain a cap and not let it recover. Again flaw in the testing which its basis is flawed.
I should test this again, if I can get some idea of how long a real bass note lasts. I can run the amp at say 30% full power, have a bass note of sufficient voltage to overpower the power source, and see how well the cap holds up. I also need to know how long between bass hits, as that will affect how long the cap gets charged between hits.
Please don't take this the wrong way but you have way to much time on your hands to be worrying about this. Just take it from those that have been doing this for a while. Trying to say if your running XXX watts you have to have XF Cap. exactly just dosen't work. Though a general rule of thumb can be applied. I have stated it earlier.
As for different amps NOT needing caps......that is not true! Though some need more help than others.(thats for sure) Though if you look at the size of the caps we use.........do you really think they can fit that size cap inside an amp to meet every demand of that amp and keep it from NOT being a surf board?! Plus they make you buy more stuff.
I am unfortunatly totally burnt out on people trying to totally prove something. I really need to just remove myself from the topic. So I will try to do just that. Have your fun with the testing but I hope that people do not read this and think of it as proven fact when the basis for the study is flawed and not a real world setup. What you need to do is get an alternator hook it up with a car batterie and spin it at different speeds idle/driving. Then you can do your bench testing at home!
Community Organizer
Joined: Oct 2000
Posts: 8,642
Likes: 143
From: Chapel Hill, NC
Phew! I would suggest that you do the experiment again though use a sub say an 8 or 10" speaker.
The reason I say sub rather than a resistor or even a tweeter is that the load shown to move such a piece is very little.......why I say caps are needed for (almost) only systems with subs.
The reason I say sub rather than a resistor or even a tweeter is that the load shown to move such a piece is very little.......why I say caps are needed for (almost) only systems with subs.
Registered User
Joined: Oct 2000
Posts: 7,134
Likes: 3
From: Columbia, MD
I'm on vacation (sort of...Indiana...no comment...and this keyboard SUCKS!), so I don't have the luxury of sitting down for an extended period of time and going over the details with a fine-tooth comb, but I'll do my best. This may extend over several emails, depending upon time contraints. As a forewarning, I have NOT read the prior thread lately...
Pinky's right (GASP!
)...if you want to make this a fair test of caps' abilities, you need to stress the system. A 50W max amp isn't going to do that, especially for an aleternator that can crank out, what, 135A was it? You're not going to see any useful test results until that system is starting to see brownout conditions. But as Jerry pointed out, the caps are there strictly for "maintaining headroom, not increasing it" [paraphrased]. Ideally, the amp would be able to comfortably suck down more juice than the alternator can supply for a good test...just crank the amp up to the lmit of what CAN be supplied and watch the results at that point. Upon rereading some later comments, I note that you scaled down all parts of the system, including available supply power...that should be valid, as long as it stay within the conditions I mentioned above.
As far as the 4 ohm resistor vs. speaker debate goes, I would like to think on that one some more. While using a resistor almost guarantees the amplifier will be pumping power out in the most efficient manner through the frequency range (you could always get more power from an amp loaded by a real resistance over a complex resistance), it doesn't necessarily help us, in and of itself. A real resistance (4 ohm resistor) does not provide the variable load that a complex resistance (a voice coil) has, but the real question for this test is "Does it matter?". Since we're only looking at how a cap helps amps in general, and not how several amps compare to each other with/without a cap, the answer is most likely "No, it doesn't matter.". We should be able to see the effects (or lack thereof) of a cap regardless of the load on the amp, it's just that those effects may appear at different points in the test (but that's OK).
The effects (if any) a cap will have on the system will generally only be evident as the system (alternator/battery combo) nears its limits. If the amplifier and other accessories are only requiring 60A from the alternator, then the caps effects will be minimal. However, as we approach that limit, the cap would prove more and more useful (theoretically). Under standard driving conditions, those limits will be roughly the same as if the car was standing still...crank out some high RPMs where we need some serious spark, and now we're lowering the limits from the amplifier's viewpoint.
From a designer's standpoint, I can say that the quality of the amp will have very little to do with the usefulness of an external cap. It's simply a matter of economics and space. The external cap is providing a service...an external reservoir for quickly-needed power. A quality amp will need this every bit as much as a poorly designed amp, but there simply isn't enough room inside the cabinet to economically put that reserve tank in. If a high quality amp manufacturer DID put that in his amp, you would see a requisite size increase in the amp casing, plain and simple.
I've seen a few comments about battery drain issues. No real-world item is perfect, and even the highest quality caps have an internal resistance. Over time, this resistance will slowly bleed the power from any battery. Granted, a reasonable quality cap should have such a low resistance as to allow the battery to die itself faster than the cap could drain it, but it's something to be aware of.
OK, on to the voice coil/motor comparison. From a 10,000 foot standpoint, the comparison is valid, as long as no one tries to make the comparison in the other direction Some basic DC motors use a very simple coil design, like a voice coil, but beyond that, specifications go in tangential directions. Comparing the two seems akin to comparing a modem and a TV...they use some of the same components, but they use them in different manners...sorry, I'm being too picky. Anyway, coils don't change resistance, regardless of the signal frequencies passing through them, they change impedance. It may seem a minor difference, but this the heart of the resistor vs. coil argument. The coil and resistor BOTH show 4 ohms of resistance at all times, but the resistor will show zero impedance (only considering complex resistance, not real values) regardless of frequency, and the coil will show varying impedance with frequency. This varying impedance is one major factor in how an amplifier driven to it's limits will react.
Duration of bass notes is of less importance than how much power is sucked down between each wave peak (or zero crossing). A note could last for several seconds, but if the amplitude is small, it makes no difference.
On battery voltage...I cannot remember what the true voltage is of hand, but you can figure it out. They are comprised of cells similar in manner to placing several AA betteries in your TV remote. Each of those is 1.2V, so two in parallel gives you 2.4V. The car battery is some combination of series and/or parallel cells, so you only need to know the setup and individual cell voltage. I'll try to remember and look up these items when I get back into town...
Now that I've read through every message in the thread as I've tried to answer every point, it's nice to see we can all agree on some of the major points. This test shows real promise as to getting the facts straight, and I'd like to see more done just like it. Do I believe caps help? I believe there is some real theoretical basis for it from a 10,000 foot view. Can I say for sure? Certainly not, but some of the numbers and figures I've calculated for a few fellow board members show promise for the future of caps. One day I'll get a wild hair up my ass and calculate just how fast a cap needs to be for an average power delivery system, but for now I'm content with bench-level tests.
THESE are the discussions I like to see... a good mix of theory, calculations, logic, common-sense arguments, and real-world bench tests
Pinky's right (GASP!
)...if you want to make this a fair test of caps' abilities, you need to stress the system. A 50W max amp isn't going to do that, especially for an aleternator that can crank out, what, 135A was it? You're not going to see any useful test results until that system is starting to see brownout conditions. But as Jerry pointed out, the caps are there strictly for "maintaining headroom, not increasing it" [paraphrased]. Ideally, the amp would be able to comfortably suck down more juice than the alternator can supply for a good test...just crank the amp up to the lmit of what CAN be supplied and watch the results at that point. Upon rereading some later comments, I note that you scaled down all parts of the system, including available supply power...that should be valid, as long as it stay within the conditions I mentioned above.As far as the 4 ohm resistor vs. speaker debate goes, I would like to think on that one some more. While using a resistor almost guarantees the amplifier will be pumping power out in the most efficient manner through the frequency range (you could always get more power from an amp loaded by a real resistance over a complex resistance), it doesn't necessarily help us, in and of itself. A real resistance (4 ohm resistor) does not provide the variable load that a complex resistance (a voice coil) has, but the real question for this test is "Does it matter?". Since we're only looking at how a cap helps amps in general, and not how several amps compare to each other with/without a cap, the answer is most likely "No, it doesn't matter.". We should be able to see the effects (or lack thereof) of a cap regardless of the load on the amp, it's just that those effects may appear at different points in the test (but that's OK).
The effects (if any) a cap will have on the system will generally only be evident as the system (alternator/battery combo) nears its limits. If the amplifier and other accessories are only requiring 60A from the alternator, then the caps effects will be minimal. However, as we approach that limit, the cap would prove more and more useful (theoretically). Under standard driving conditions, those limits will be roughly the same as if the car was standing still...crank out some high RPMs where we need some serious spark, and now we're lowering the limits from the amplifier's viewpoint.
From a designer's standpoint, I can say that the quality of the amp will have very little to do with the usefulness of an external cap. It's simply a matter of economics and space. The external cap is providing a service...an external reservoir for quickly-needed power. A quality amp will need this every bit as much as a poorly designed amp, but there simply isn't enough room inside the cabinet to economically put that reserve tank in. If a high quality amp manufacturer DID put that in his amp, you would see a requisite size increase in the amp casing, plain and simple.
I've seen a few comments about battery drain issues. No real-world item is perfect, and even the highest quality caps have an internal resistance. Over time, this resistance will slowly bleed the power from any battery. Granted, a reasonable quality cap should have such a low resistance as to allow the battery to die itself faster than the cap could drain it, but it's something to be aware of.
OK, on to the voice coil/motor comparison. From a 10,000 foot standpoint, the comparison is valid, as long as no one tries to make the comparison in the other direction
Some basic DC motors use a very simple coil design, like a voice coil, but beyond that, specifications go in tangential directions. Comparing the two seems akin to comparing a modem and a TV...they use some of the same components, but they use them in different manners...sorry, I'm being too picky. Anyway, coils don't change resistance, regardless of the signal frequencies passing through them, they change impedance. It may seem a minor difference, but this the heart of the resistor vs. coil argument. The coil and resistor BOTH show 4 ohms of resistance at all times, but the resistor will show zero impedance (only considering complex resistance, not real values) regardless of frequency, and the coil will show varying impedance with frequency. This varying impedance is one major factor in how an amplifier driven to it's limits will react.Duration of bass notes is of less importance than how much power is sucked down between each wave peak (or zero crossing). A note could last for several seconds, but if the amplitude is small, it makes no difference.
On battery voltage...I cannot remember what the true voltage is of hand, but you can figure it out. They are comprised of cells similar in manner to placing several AA betteries in your TV remote. Each of those is 1.2V, so two in parallel gives you 2.4V. The car battery is some combination of series and/or parallel cells, so you only need to know the setup and individual cell voltage. I'll try to remember and look up these items when I get back into town...
Now that I've read through every message in the thread as I've tried to answer every point, it's nice to see we can all agree on some of the major points. This test shows real promise as to getting the facts straight, and I'd like to see more done just like it. Do I believe caps help? I believe there is some real theoretical basis for it from a 10,000 foot view. Can I say for sure? Certainly not, but some of the numbers and figures I've calculated for a few fellow board members show promise for the future of caps. One day I'll get a wild hair up my ass and calculate just how fast a cap needs to be for an average power delivery system, but for now I'm content with bench-level tests.
THESE are the discussions I like to see... a good mix of theory, calculations, logic, common-sense arguments, and real-world bench tests
Joined: Oct 2000
Posts: 8,502
Likes: 0
From: RUNNING!:)
Originally posted by MacGyver
Pinky's right (GASP!)...
Pinky's right (GASP!
)...
OMG!!!
I almost fell out of my chair when I read this!!!!Just kidding with you guys
By the way.... Why are you in Indiana?!?!?
Umm... Just kidding again
Have fun!Sorry to jump in here you guys... the above post sort of shocked me
Thread Starter
Honorary Member
Joined: May 2001
Posts: 2,121
Likes: 3
From: Indian Land SC
Originally posted by MacGyver
from a 10,000 foot view.
from a 10,000 foot view.
[QUOTE]Originally posted by MacGyver
[B]Pinky's right (GASP!
)...if you want to make this a fair test of caps' abilities, you need to stress the system.
Registered User
Joined: Oct 2000
Posts: 7,134
Likes: 3
From: Columbia, MD
Modifry,
I think we're on the same page here, but I'm guilty of not being specific in my own language. I don't doubt my quick review of the LOOOOONG thread missed the stressed-system setup. Testing an 18A system with a 12A supply should be adequate...though I wouldn't try to project results coming from a 0.18A system and a 0.12A supply, since then micro/macro system problems start to appear.
I could SWEAR I put in a disclaimer saying zero impedance meant (in this context, at least, for simplicity) considering only complex resistances (i.e., removing the 4 ohms of real resistance and only considering complex resistances of inductors and capacitors), but I may have deleted it in my editing. The resistor will always show the best power output for any particular amplifier, which is why it's used. However, amps will act differently if their loads consist of complex loads, such as those shown by an inductor (i.e., voice coil) or capacitor. For these tests, it won't make any difference, since we're not trying to see how well one amp can do compared to another, we want to know how one amp does under several conditions.
To be specific in answering your question, a resistor will provide a proper load for an amplifier. It also (generally) provides an optimal load, in terms of power efficiency. So, you could say that the resistor will sink more power than a speaker, but you cannot say a speaker an resistor have an equivalent impedance (since the resistor has an impedance strictly in the real plane, and the coil has an impedance with a significant portion in the complex plane). If the voice coil of that speaker is guaranteed to be 4 ohms (with NO reactance, which isn't possible, but it could be pretty CLOSE to zero) at 40 Hz, then there isn't a difference between the two tests.
The "quickly-needed" power from the cap is just as you mention...a fast rise time. Car batteries are not designed to provide quick spikes of power, but caps are. Again, the duration issue you mention is at the heart of all of this...can the cap provide a bit of energy at a fast enough rate, but only for a short time. I went through a similar discussion, I believe, in the other thread, but as long as the amount of power you suck from the cap isn't greater than what it can replenish between long pulls (4-5 second long bass notes), it should do it's job. The calculations I mention doing are to determine whether the cap can provide that small spike of energy fast enough, and if they can, how many spikes (and at what power level) can it provide before needing to be recharged (i.e., end of bass note). This may invalidate a "long-term" test, but would provide some useful info on whether a cap can/cannot "hold it's breath", so to speak, between looooong notes.
Oops, gotta go....I'll be home later this weekend, and check this thread out again...
I think we're on the same page here, but I'm guilty of not being specific in my own language. I don't doubt my quick review of the LOOOOONG thread missed the stressed-system setup. Testing an 18A system with a 12A supply should be adequate...though I wouldn't try to project results coming from a 0.18A system and a 0.12A supply, since then micro/macro system problems start to appear.
I could SWEAR I put in a disclaimer saying zero impedance meant (in this context, at least, for simplicity) considering only complex resistances (i.e., removing the 4 ohms of real resistance and only considering complex resistances of inductors and capacitors), but I may have deleted it in my editing. The resistor will always show the best power output for any particular amplifier, which is why it's used. However, amps will act differently if their loads consist of complex loads, such as those shown by an inductor (i.e., voice coil) or capacitor. For these tests, it won't make any difference, since we're not trying to see how well one amp can do compared to another, we want to know how one amp does under several conditions.
To be specific in answering your question, a resistor will provide a proper load for an amplifier. It also (generally) provides an optimal load, in terms of power efficiency. So, you could say that the resistor will sink more power than a speaker, but you cannot say a speaker an resistor have an equivalent impedance (since the resistor has an impedance strictly in the real plane, and the coil has an impedance with a significant portion in the complex plane). If the voice coil of that speaker is guaranteed to be 4 ohms (with NO reactance, which isn't possible, but it could be pretty CLOSE to zero) at 40 Hz, then there isn't a difference between the two tests.
The "quickly-needed" power from the cap is just as you mention...a fast rise time. Car batteries are not designed to provide quick spikes of power, but caps are. Again, the duration issue you mention is at the heart of all of this...can the cap provide a bit of energy at a fast enough rate, but only for a short time. I went through a similar discussion, I believe, in the other thread, but as long as the amount of power you suck from the cap isn't greater than what it can replenish between long pulls (4-5 second long bass notes), it should do it's job. The calculations I mention doing are to determine whether the cap can provide that small spike of energy fast enough, and if they can, how many spikes (and at what power level) can it provide before needing to be recharged (i.e., end of bass note). This may invalidate a "long-term" test, but would provide some useful info on whether a cap can/cannot "hold it's breath", so to speak, between looooong notes.
Oops, gotta go....I'll be home later this weekend, and check this thread out again...