Power to drivers
12-10-2003, 06:51 AM
#1
1000 Watt CAFz'r
Thread Starter
Join Date: Oct 2003
Posts: 1,783
Here we go again [img]graemlins/freak.gif[/img]
I was looking at some impedence graphs from various drivers recently,(tweeters) and an odd thought struck me.
It's really pretty much common sense with Ohms law. A driver recieves different amounts of current at different frequecies. Lets say with a 4 ohm resistor on an amp, it makes 50 watts. Now put a 4 ohm tweeter on it, how much power is the tweeter actually seeing? (I'm assuming that there is no passive filter)
I"m going to use tweeters as an example, because most people are far more afraid of over powering tweeters.
I'd like to pose the question of how much power a tweeter really recieves. Most drivers have a substial dip in the impedence in the middle frequencies of it's range and then raises up towards it's upper frequencies.(Inductance at work) So how much power do we think a tweeter can really handle at lets say 10 Khz?
Just rambling for fun, I'll probably post a much more coherent review of my idea later.
Adam
I was looking at some impedence graphs from various drivers recently,(tweeters) and an odd thought struck me.
It's really pretty much common sense with Ohms law. A driver recieves different amounts of current at different frequecies. Lets say with a 4 ohm resistor on an amp, it makes 50 watts. Now put a 4 ohm tweeter on it, how much power is the tweeter actually seeing? (I'm assuming that there is no passive filter)
I"m going to use tweeters as an example, because most people are far more afraid of over powering tweeters.
I'd like to pose the question of how much power a tweeter really recieves. Most drivers have a substial dip in the impedence in the middle frequencies of it's range and then raises up towards it's upper frequencies.(Inductance at work) So how much power do we think a tweeter can really handle at lets say 10 Khz?
Just rambling for fun, I'll probably post a much more coherent review of my idea later.
Adam
12-10-2003, 08:53 AM
#2
500 Watt CAFz'r
Join Date: Jul 2000
Posts: 870
We are correct that a driver receives different amounts of power at different frequencies. You half to remember that in a crossover set up that some inductors have resistances as high as 1-2 ohm. That being said these can rob power from any wheres up to 30%. This is found mostly on air core inductors.
So depending on where the filtering is takeing place, Prior to amplification there is no insertion loss, that is when it would be in question of how much power you do want to feed your tweeters.
If you do this in a passive desgin after ampification has occured then it is in question of how much power they will take. 250 watts less %25 187 watts. 150 watts less 25% 112 watts becuase of the passive crossover having a 1-2 ohm inductance.
So the more power at high frequencies doesn't always mean you are getting that exact power out put and that would determin how much power the tweeter really gets.
So depending on where the filtering is takeing place, Prior to amplification there is no insertion loss, that is when it would be in question of how much power you do want to feed your tweeters.
If you do this in a passive desgin after ampification has occured then it is in question of how much power they will take. 250 watts less %25 187 watts. 150 watts less 25% 112 watts becuase of the passive crossover having a 1-2 ohm inductance.
So the more power at high frequencies doesn't always mean you are getting that exact power out put and that would determin how much power the tweeter really gets.
12-10-2003, 01:39 PM
#3
Guest
Posts: n/a
Duude, if your insertion loss is 20%+ on a passive there is a HUGE problem. 2ohms resistance on an inductor can occur on large values with small ga. wiring but very rarely would these be seen other than for a woofer.
Remember - passives are cool [img]graemlins/thumb.gif[/img]
Adam - set up correctly, your tweeters will receive all the power they need too..
ooooohhhmmmmmm
Remember - passives are cool [img]graemlins/thumb.gif[/img]
Adam - set up correctly, your tweeters will receive all the power they need too..
ooooohhhmmmmmm
12-11-2003, 01:37 AM
#5
1000 Watt CAFz'r
Thread Starter
Join Date: Oct 2003
Posts: 1,783
I am very familiar with how passives work. Stated above is the supposition that there is NO passive x.o. between the amp and driver.
I used a tweeter as an example, but the same theory would apply to a mid or sub. An impedence curve is a measure of more than just electrical resistance. That said, it has the same net result on the amplifier. Different frequencies recieve different amounts of current. If a graph of this was made, you would see widely fluctuating current absorbed by the driver across it's frequency range.
If drivers had flat 4 ohm impedences across the entire frequency range, passive x.o.s would be a breaze to design. Some of the higher end passive cross over networks have impedence shaping circuits to help "load" an amplifier more evenly across it's frequencey spectrum. It's very common on high end subs to put a notch type filter circuit in parallel with the driver to dampen the impedence spike at the driver's resononant frequency.
I'm just interested in understanding more of how driver's interact with amps.
Anyway, just chewing the fat online.
Later,
Adam
I used a tweeter as an example, but the same theory would apply to a mid or sub. An impedence curve is a measure of more than just electrical resistance. That said, it has the same net result on the amplifier. Different frequencies recieve different amounts of current. If a graph of this was made, you would see widely fluctuating current absorbed by the driver across it's frequency range.
If drivers had flat 4 ohm impedences across the entire frequency range, passive x.o.s would be a breaze to design. Some of the higher end passive cross over networks have impedence shaping circuits to help "load" an amplifier more evenly across it's frequencey spectrum. It's very common on high end subs to put a notch type filter circuit in parallel with the driver to dampen the impedence spike at the driver's resononant frequency.
I'm just interested in understanding more of how driver's interact with amps.
Anyway, just chewing the fat online.
Later,
Adam
12-11-2003, 09:01 AM
#6
Guest
Posts: n/a
Impedence rise is a factor with any driver. That's how SPL'ers get away with bridging a pair of 1000 watt amps onto a single 1 ohm voil without blowing their amps. At the frequency the burp at actual impedence can be much higher, allowing the amps not to implode.
Power handling of a driver is directly related to the intelligence of the user.
Power handling of a driver is directly related to the intelligence of the user.
12-11-2003, 09:41 AM
#7
Guest
Posts: n/a
Impedance curves have to do with a lot more than than just the characteristics of the driver, they are greatly affected by the application. Bass Reflex, Acoustic Suspension and Aperiodic enclosures all have widely varying impedance curves.
As for the insertion loss of an inductor or capacity, I to would be worried about something as high as 1 or 2 ohms.
As for the relationship between frequency response and the reactive nature of speakers, crossover networks and cabinet designs and characteristics, this all plays a role in the resulting frequency response and power being supplied to a driver.
If you have read reviews of reactive speakers being drive buy SET or other tube amps, then you will know that the type of amp used will have an effect on frequency response as the ratio of load impedance to output impedance varies greatly.
Back to the original question, does a driver get different amounts of power at different frequencies based on the impedance curve. Yep.
What gets really ugly, and something I am just wrapping my brain around is apparent power. This is a little deep, but hold on...
Power, in a resistive load, is always positive, in that its the RMS magnitude of voltage multiplied by the current. When driving a reactive load, the current through the load can lag behind that voltage supplied to it. The power delivery can acutally become quite non-linear. Ugly and confusing if you don't completely understand phase angles, capacitance and inductance.
Designing a speaker or speaker system to minimize reactive effects has the benefit of providing a more coherent power delivery to the drivers, and as there are less loses, better overall system efficiency.
There is an excellent review of the Canadian made 3A MM de Capo i loudspeaker by Art Dudley. The speaker uses only a capacitor on the tweeter. This is one of those drivers that sounds better than it measures, and Dudley absolutely loved it. JA's technical review is quite enlightening.
I have a true power meter on its way, I should have it within a month or so. This unit simultaneously measures current and voltage to calculate power. Once I have it, I'll do a little test to clear this up a bit...
As for the insertion loss of an inductor or capacity, I to would be worried about something as high as 1 or 2 ohms.
As for the relationship between frequency response and the reactive nature of speakers, crossover networks and cabinet designs and characteristics, this all plays a role in the resulting frequency response and power being supplied to a driver.
If you have read reviews of reactive speakers being drive buy SET or other tube amps, then you will know that the type of amp used will have an effect on frequency response as the ratio of load impedance to output impedance varies greatly.
Back to the original question, does a driver get different amounts of power at different frequencies based on the impedance curve. Yep.
What gets really ugly, and something I am just wrapping my brain around is apparent power. This is a little deep, but hold on...
Power, in a resistive load, is always positive, in that its the RMS magnitude of voltage multiplied by the current. When driving a reactive load, the current through the load can lag behind that voltage supplied to it. The power delivery can acutally become quite non-linear. Ugly and confusing if you don't completely understand phase angles, capacitance and inductance.
Designing a speaker or speaker system to minimize reactive effects has the benefit of providing a more coherent power delivery to the drivers, and as there are less loses, better overall system efficiency.
There is an excellent review of the Canadian made 3A MM de Capo i loudspeaker by Art Dudley. The speaker uses only a capacitor on the tweeter. This is one of those drivers that sounds better than it measures, and Dudley absolutely loved it. JA's technical review is quite enlightening.
I have a true power meter on its way, I should have it within a month or so. This unit simultaneously measures current and voltage to calculate power. Once I have it, I'll do a little test to clear this up a bit...
12-11-2003, 10:32 AM
#8
1000 Watt CAFz'r
Thread Starter
Join Date: Oct 2003
Posts: 1,783
Thanks Dave, I know it takes a bit of time to write something like that out. Not everybody is grasping the idea behind this, but I can see where you are going with this. This is kind of related to the "how much power" subject I brought up a while ago. When is that amp test?
I can see your point about the voltage versus amperage, but both would ultimately be affected by the impedence curve. I've never considered the amperage lagging behind the voltage. Time to do some head scratching.
Even wire has a capacitive value to it, so those with 20 ft of cable running to the door speakers have something to consider.(although it's an absolutely tiny value)
You are absolutely right about the type of box a driver is in. I"ve done tests on one specific driver in different sized boxes. What most people don't realize, is that putting a midrange (for example) in a small enclosure versus IB will change how the passive X.O. filters. I've heard it, I've measured it,(256 point 3 octave graph) and I can tell you it's a bitch to tune.
I can't wait to see what you are going to do with that power meter. What's the sampling rate of it? (ie how many times per second does it measure assuming it's digital?)
Later,
Adam
I can see your point about the voltage versus amperage, but both would ultimately be affected by the impedence curve. I've never considered the amperage lagging behind the voltage. Time to do some head scratching.
Even wire has a capacitive value to it, so those with 20 ft of cable running to the door speakers have something to consider.(although it's an absolutely tiny value)
You are absolutely right about the type of box a driver is in. I"ve done tests on one specific driver in different sized boxes. What most people don't realize, is that putting a midrange (for example) in a small enclosure versus IB will change how the passive X.O. filters. I've heard it, I've measured it,(256 point 3 octave graph) and I can tell you it's a bitch to tune.
I can't wait to see what you are going to do with that power meter. What's the sampling rate of it? (ie how many times per second does it measure assuming it's digital?)
Later,
Adam
12-15-2003, 09:46 PM
#9
50 Watt CAFz'r
Join Date: Sep 2003
Posts: 192
Hey Dave,
do you have to worry about reflected power at low frequencies (audio) to get an accurate measurement of the true power? HF stuff like radio antennae and the like can have some of the power reflected back due to impedance mis-matches, is this true for audio applications as well?
I've been wondering about this for a while because as the impedance changes (both the driver/XO and amp output to a lesser degree) the amount of reflected power could swing greatly and in some cases return all of the power delivered by the amp. Or does any of this matter since the system is not impedance matched?
So many questions, so much typing. J
do you have to worry about reflected power at low frequencies (audio) to get an accurate measurement of the true power? HF stuff like radio antennae and the like can have some of the power reflected back due to impedance mis-matches, is this true for audio applications as well?
I've been wondering about this for a while because as the impedance changes (both the driver/XO and amp output to a lesser degree) the amount of reflected power could swing greatly and in some cases return all of the power delivered by the amp. Or does any of this matter since the system is not impedance matched?
So many questions, so much typing. J
12-16-2003, 10:29 AM
#10
Guest
Posts: n/a
There is so much 'weird ****' going on in an audio system, I pretty much can't worry about everything. There are reflections, Back-EMF, and more weird stuff.
Antenna operate are really, really high frequencies, where impedance matching is far more critical.
The output impedance of an amp, compared to the load is around 100:1 or greater (damping factor), so most of the power goes to the load.
For now, I just measure voltage. That keeps it simple. When I get the new meter, I'll be measuring voltage and current.
If I get some time over the holidays (yeah right, you should see the stuff I already have in-house to test), I'll do a frequency vs current vs output amplitude graph for y'all, that should clear it up.
Antenna operate are really, really high frequencies, where impedance matching is far more critical.
The output impedance of an amp, compared to the load is around 100:1 or greater (damping factor), so most of the power goes to the load.
For now, I just measure voltage. That keeps it simple. When I get the new meter, I'll be measuring voltage and current.
If I get some time over the holidays (yeah right, you should see the stuff I already have in-house to test), I'll do a frequency vs current vs output amplitude graph for y'all, that should clear it up.
Thread
Thread Starter
Forum
Replies
Last Post
ZachCHartwell
General Discussion
4
04-30-2004 09:10 PM
Hardcore Rock Superstar
General Discussion
17
07-09-2003 01:01 PM
