JkSpinner Posted April 2, 2021 Share Posted April 2, 2021 (edited) Hi all can someone explain speaker impedance to me please, especially how it relates to an amp. I have a Denon 3312 that has includes two impedance specs in the specifications 8ohm, and 6 Ohm does this mean the amp will not power 4ohm speakers? I have a mixed bag of speakers, bought for a purpose, ie, dipole speakers for the side wall, and thin speakers for the rear wall, not matching. My front three are matching though I also have several speakers some are 8 ohms normal others suggest 8 ohm normal, and 2.7 ohms minimal what does this mean in relation to my amp i am considering a Rotel 1555 5 x 120w amp anyway, but trying to get an understanding of what all this means, as I am looking to make several upgrades thanks Edited April 2, 2021 by JkSpinner Link to comment Share on other sites More sharing options...
Quark Posted April 3, 2021 Share Posted April 3, 2021 You can think of impedance as the AC equivalent of resistance. Speakers with a lower impedance will draw more current. At higher volumes an unsuitable amp will run out of current capacity and clip - essentially send DC to the drivers - this is likely to fry them. 2.7 Ohms is quite low (and likely to be problematic with your 3312), but often low impedances are in bass frequencies and if your AVR's crossover is set to the typical 80Hz AND your speakers are set to Small in the AVR, much of the problematic low impedance won't be loaded on your AVR. What make/model are your speakers? Often there will be reviews that measure the impedance curve at different frequencies. 1 Link to comment Share on other sites More sharing options...
stereo coffee Posted April 3, 2021 Share Posted April 3, 2021 On 02/04/2021 at 1:24 PM, JkSpinner said: Hi all can someone explain speaker impedance to me please, especially how it relates to an amp. I have a Denon 3312 that has includes two impedance specs in the specifications 8ohm, and 6 Ohm does this mean the amp will not power 4ohm speakers? I have a mixed bag of speakers, bought for a purpose, ie, dipole speakers for the side wall, and thin speakers for the rear wall, not matching. My front three are matching though I also have several speakers some are 8 ohms normal others suggest 8 ohm normal, and 2.7 ohms minimal what does this mean in relation to my amp i am considering a Rotel 1555 5 x 120w amp anyway, but trying to get an understanding of what all this means, as I am looking to make several upgrades thanks When amplifier manufacturers rate power amplifiers they do so into a variety of resistance loads, loudspeakers as @Quark informs are different, in that they present to a amplifier a reactance which stores some of that energy and later releases it. Impedance differs to resistance as it is the contribution of both resistance and reactance. Here is a graph showing a solid state amplifiers ability into a number of resistance loads. it can provide some indication if we fall for thinking loudspeakers represent resistance, but that is not the case, as few of us enjoy the heat of a resistor, rather there is expectation a loudspeaker will be used. Also within this is the need to look at the contribution of the crossover network and two way vs three way which has its own thread You could choose to use planar speakers which typically have very high impedance at lower frequencies and very low impedance at high frequencies. See the Quad ESL 57 impedance graph https://medialux.blog/2018/04/25/erster-blogbeitrag/ At this point its good to keep in mind that there are fundamental differences between solid state and valve amplifiers, stated here http://education.lenardaudio.com/en/14_valve_amps_7.html Valve amp power is directly proportional to speaker Impedance therefore power increases as the speaker Impedance rises. Solid-state power is inversely proportional to speaker Impedance therefore power decreases as the speaker Impedance rises. The reason for amplifier manufacturers not taking the further step to test with loudspeakers is that they are presently voltage amplification devices which only indirectly affects the current flowing in a loudspeakers voice coil. The future for amplifiers is to be current driven, and have sensing of the loudspeaker as part of its feedback path, but until then, I think Martin Colloms famous words that " all amplifiers sound different " needs to be reflected upon vs Peter Walker of Quad, words, "all amplifiers sound the same" Auditioning of methods of attenuation, power amplifiers and the speaker you intend to use, is absolutely essential. Link to comment Share on other sites More sharing options...
andyr Posted April 3, 2021 Share Posted April 3, 2021 16 hours ago, stereo coffee said: You could choose to use planar speakers which typically have very high impedance at lower frequencies and very low impedance at high frequencies. See the Quad ESL 57 impedance graph https://medialux.blog/2018/04/25/erster-blogbeitrag/ Perhaps you should've used the term "electrostatic speakers" instead of "planar speakers", Chris. Here is the impedance curve for a Maggie 1.6 which has a series passive XO instead of the stock parallel XO: Pretty flat all the way through! Andy 1 Link to comment Share on other sites More sharing options...
stereo coffee Posted April 4, 2021 Share Posted April 4, 2021 1 hour ago, andyr said: Perhaps you should've used the term "electrostatic speakers" instead of "planar speakers", Chris. Here is the impedance curve for a Maggie 1.6 which has a series passive XO instead of the stock parallel XO: Pretty flat all the way through! Andy Hi Andy You are right , yes I was associating planars , thinking of them just as electrostatics, when there are many varieties of planar speakers. The Maggies as you show being exceptional with their impedance, that part of impedance related to resistance, being very flat indeed. Link to comment Share on other sites More sharing options...
Ittaku Posted April 4, 2021 Share Posted April 4, 2021 19 hours ago, stereo coffee said: The future for amplifiers is to be current driven, and have sensing of the loudspeaker as part of its feedback path That's just how you see it and wish for it to happen. Unfortunately no matter how much you wish for it, I'm quite sure that apart from perhaps some exceptional examples, won't be happening any time soon, if ever. 2 Link to comment Share on other sites More sharing options...
Addicted to music Posted April 4, 2021 Share Posted April 4, 2021 13 minutes ago, Ittaku said: That's just how you see it and wish for it to happen. Unfortunately no matter how much you wish for it, I'm quite sure that apart from perhaps some exceptional examples, won't be happening any time soon, if ever. Most if not ALL conventional A and B amplifiers apart from Class D etc driving speakers are Voltage source amplification. Constant voltage across the 20-20k when a input is applied for the said freq Unfortunately a speaker is a reactive RCL device that varies with freq. The best out there such as Dan D’Agostino etc, his designs are no different..... So your best bet for SS amplification is to have as many BJT devices to handle the toughest of loads, that’s loads where current is demanded. Note I said BJT, as there are positive temp coefficient, Link to comment Share on other sites More sharing options...
stereo coffee Posted April 4, 2021 Share Posted April 4, 2021 16 minutes ago, Addicted to music said: Most if not ALL conventional A and B amplifiers apart from Class D etc driving speakers are Voltage source amplification. Constant voltage across the 20-20k when a input is applied for the said freq Unfortunately a speaker is a reactive RCL device that varies with freq. The best out there such as Dan D’Agostino etc, his designs are no different..... So your best bet for SS amplification is to have as many BJT devices to handle the toughest of loads, that’s loads where current is demanded. Note I said BJT, as there are positive temp coefficient, However current drive amps, with advantage of eliminating major distortion, can be obtained https://www.integeraudio.com/ or DIY built https://www.current-drive.info/9 Link to comment Share on other sites More sharing options...
Addicted to music Posted April 4, 2021 Share Posted April 4, 2021 3 minutes ago, stereo coffee said: However current drive amps, with advantage of eliminating major distortion, can be obtained https://www.integeraudio.com/ or DIY built https://www.current-drive.info/9 I hate to say it and I’ll say it once; “stop linking marketing BS”. Do some research and look at how some of the most the respected designers and you’ll see how to design something that can be a “spot welder” 2 2 Link to comment Share on other sites More sharing options...
ray4410 Posted April 4, 2021 Share Posted April 4, 2021 22 minutes ago, Addicted to music said: I hate to say it and I’ll say it once; “stop linking marketing BS”. Do some research and look at how some of the most the respected designers and you’ll see how to design something that can be a “spot welder” you mean something like this 1 Link to comment Share on other sites More sharing options...
Addicted to music Posted April 4, 2021 Share Posted April 4, 2021 18 minutes ago, ray4410 said: you mean something like this Yep Link to comment Share on other sites More sharing options...
stereo coffee Posted April 4, 2021 Share Posted April 4, 2021 3 minutes ago, Addicted to music said: I hate to say it and I’ll say it once; “stop linking marketing BS”. Do some research and look at how some of the most the respected designers and you’ll see how to design something that can be a “spot welder” There is very little marketing in a well written 342 page book, rather more so an enormous passion by its author to solve why voltage drive is far from complete, even at welding type currents. Would you like me to send it to you ? Having read it it you can reflect on the all too common welding approach, and perhaps better see why it is incomplete, vs current drive. A simple example is if we consider the resistance portion of a high impedance loudspeaker as being a simple form of current drive, as its resistance portion is partially allowing current driving of a normal voltage amplifier. The infamous L/S 3,5a provides 15 ohms, and is far from being unpopular. There is no doubting voltage amps are good at what they do when measured with a resistor, but the next step of making a current drive amp, is not too much of a departure, from what is already being provided, if we aspire to achieve it. Perhaps we can both find some graphs and measurement of actual amplifiers when attached to a loudspeaker, we would then see what was BS. Link to comment Share on other sites More sharing options...
stereo coffee Posted April 4, 2021 Share Posted April 4, 2021 3 hours ago, Addicted to music said: I hate to say it and I’ll say it once; “stop linking marketing BS”. Do some research and look at how some of the most the respected designers and you’ll see how to design something that can be a “spot welder” There is very little marketing in a well written 342 page book, rather more so an enormous passion by its author to solve why voltage drive is far from complete, even at welding type currents. Would you like me to send it to you ? Having read it it you can reflect on the all too common welding approach, and perhaps better see why it is incomplete, vs current drive. A simple example is if we consider the resistance portion of a high impedance loudspeaker as being a simple form of current drive, as its resistance portion is partially allowing current driving of a normal voltage amplifier. The infamous L/S 3,5a provides 15 ohms, and is far from being unpopular. There is no doubting voltage amps are good at what they do when measured with a resistor, but the next step of making a current drive amp, is not too much of a departure, from what is already being provided, if we aspire to achieve it. Perhaps we can both find some graphs and measurement of actual amplifiers when attached to a loudspeaker, we would then see what was BS. I found Stereophile admitting something is missing with amplifier tests, when they are conducted when not using actual loudspeakers https://www.stereophile.com/reference/60/index.html and reference to a Audio Engineering Society article, which is also at research gate, which I have requested a copy of, which may be subject to copyright. Failing that I might have to set up my spare oscilloscope, and run a sine wave signal with and without speaker so we can see the effect on a scope of a few speakers. As a collective of persons interested in audio IMO we should always expect actual loudspeaker loads to be part of amplifier testing, and reject as having as much less meaning resistor loads, as being nice, but not too meaningful. Link to comment Share on other sites More sharing options...
Volunteer sir sanders zingmore Posted April 4, 2021 Volunteer Share Posted April 4, 2021 2 minutes ago, stereo coffee said: As a collective of persons interested in audio IMO we should always expect actual loudspeaker loads to be part of amplifier testing, and reject as having as much less meaning resistor loads, as being nice, but not too meaningful. Which loudspeaker should we select to replace resistor loads for amplifier testing? Link to comment Share on other sites More sharing options...
stereo coffee Posted April 4, 2021 Share Posted April 4, 2021 2 minutes ago, sir sanders zingmore said: Which loudspeaker should we select to replace resistor loads for amplifier testing? Good question, I think a three way design and a two way design, and electrostatic where available should form part of all amplifier tests. The tables would begin to turn to show preference I think for simple crossover networks. A standard reference with two way being the LS /3,5a, i don't think any speaker other than maybe the ESL57 has received so much research. With 3 way lets plumb for a Yamaha NS1000 , and electrostatic the venerable ESL57 As long as testers were testing with each type ie 2 way, 3 way and a electrostatic is the emphasis, rather than resorting to resistor measurements, as they presently get too easily away with. Link to comment Share on other sites More sharing options...
andyr Posted April 4, 2021 Share Posted April 4, 2021 On 02/04/2021 at 11:24 AM, JkSpinner said: Hi all can someone explain speaker impedance to me please, especially how it relates to an amp. I have a Denon 3312 that has includes two impedance specs in the specifications 8ohm, and 6 Ohm does this mean the amp will not power 4ohm speakers? I have a mixed bag of speakers, bought for a purpose, ie, dipole speakers for the side wall, and thin speakers for the rear wall, not matching. My front three are matching though I also have several speakers some are 8 ohms normal others suggest 8 ohm normal, and 2.7 ohms minimal what does this mean in relation to my amp i am considering a Rotel 1555 5 x 120w amp anyway, but trying to get an understanding of what all this means, as I am looking to make several upgrades thanks I'm not sure you have so far received the direct answer you were hoping for, JkS ... so I thought I might try. The specs for your Denon 3312 don't include a 4ohm rating. That suggests to me ... the amp doesn't perform well into 4 ohms. (Coz if it did - they would trumpet it! ) An "ideal" amp doubles output power as you halve the load - most amps don't achieve this. With a good amp ... you might see figures like "500w/8 ohms, 900w/4 ohms, 1500w/2 ohms". What are the figures quoted for the Denon 3312, at 8 & 6 ohms? Is the 6 ohm power rating 1/3rd more than the 8 ohm power rating? As you've found from your own collection of spkrs - just because they say it's an 8 ohm spkr doesn't mean it doesn't drop to a much lower value somewhere during its frequency range. So the one that says "8 ohms nominal, 2.7 ohms minimum " is going to be a harder load for an amp to cope with than one that said "8 ohms nominal, 4 ohms minimum ". In general AVRs and multichannel amps don't handle difficult spkr loads as well as good stereo amps - simply because in a multichannel situation, the one power supply is feeding all channels (whereas in a good stereo amp, each channel has its own power supply). The specs I saw for your Denon 3312 say 7 channels at 125w per channel; so in what way is the Rotel 1555 better (given you say it is 5x 120w)? Andy 2 Link to comment Share on other sites More sharing options...
JkSpinner Posted April 4, 2021 Author Share Posted April 4, 2021 41 minutes ago, andyr said: I'm not sure you have so far received the direct answer you were hoping for, JkS ... so I thought I might try. The specs for your Denon 3312 don't include a 4ohm rating. That suggests to me ... the amp doesn't perform well into 4 ohms. (Coz if it did - they would trumpet it! ) An "ideal" amp doubles output power as you halve the load - most amps don't achieve this. With a good amp ... you might see figures like "500w/8 ohms, 900w/4 ohms, 1500w/2 ohms". What are the figures quoted for the Denon 3312, at 8 & 6 ohms? Is the 6 ohm power rating 1/3rd more than the 8 ohm power rating? As you've found from your own collection of spkrs - just because they say it's an 8 ohm spkr doesn't mean it doesn't drop to a much lower value somewhere during its frequency range. So the one that says "8 ohms nominal, 2.7 ohms minimum " is going to be a harder load for an amp to cope with than one that said "8 ohms nominal, 4 ohms minimum ". In general AVRs and multichannel amps don't handle difficult spkr loads as well as good stereo amps - simply because in a multichannel situation, the one power supply is feeding all channels (whereas in a good stereo amp, each channel has its own power supply). The specs I saw for your Denon 3312 say 7 channels at 125w per channel; so in what way is the Rotel 1555 better (given you say it is 5x 120w)? Andy Thanks Andy, well said. I have been doing heaps of reading, and posted a few specific questions. I am looking to upgrade several pieces, but want to make sure I understand most things first. It’s not just the amps, many speakers only quote nominal, and do not quote minimal. but in the end, I think if I get a 5 channel power amp, all should be fine, just need to worry about my 2 channel stuff then your comment about the 8/4 ohm. If the power doubles, I am lead to believe that it is a higher current Amp. Without trying to get into the mathematics of things, I am lead to believe that if an amp is able to produce twice the power at half the resistance, it could be considered higher current amp, and that could be useful. thanks, you were right, I don’t think the question was answered Link to comment Share on other sites More sharing options...
Guest DrSK Posted April 4, 2021 Share Posted April 4, 2021 (edited) 1 hour ago, JkSpinner said: Thanks Andy, well said. I have been doing heaps of reading, and posted a few specific questions. I am looking to upgrade several pieces, but want to make sure I understand most things first. It’s not just the amps, many speakers only quote nominal, and do not quote minimal. but in the end, I think if I get a 5 channel power amp, all should be fine, just need to worry about my 2 channel stuff then your comment about the 8/4 ohm. If the power doubles, I am lead to believe that it is a higher current Amp. Without trying to get into the mathematics of things, I am lead to believe that if an amp is able to produce twice the power at half the resistance, it could be considered higher current amp, and that could be useful. thanks, you were right, I don’t think the question was answered That covers some basic pointers, but high current often refers to peak current as well. I have these, and the max current is in arc welder territory but they don't follow the rule of thumb about doubling power for halving resistance. https://www.theaudiotailor.com.au/products/electrocompaniet-aw180-mono-block-power-amplifier-pair Based on their 250W at 8Ohm this would give just under 6A. Or just under 18A at 2Ohms. But this rating is for continuous power and in part relates to thermal overload. However the peak output per channel is over 100A. This gives plenty of head room for drum hits etc so they don't get clipped. Most amplifiers at lifestyle consumer stores etc effectively clip. Eg my 130W per channel $2k-$3k AVR (among many other things, and even at moderate listening levels) could not reproduce vocal dynamics, drum hits, double base string plucks, and low register piano strings as they were struck by the hammers, yet my 100W Metaxas stereo power amp could. Edited April 4, 2021 by DrSK Link to comment Share on other sites More sharing options...
JkSpinner Posted April 4, 2021 Author Share Posted April 4, 2021 9 minutes ago, DrSK said: That covers some basic pointers, but high current often refers to peak current as well. I have these, and the max current is in arc welder territory but they don't follow the rule of thumb about doubling power for halving resistance. https://www.theaudiotailor.com.au/products/electrocompaniet-aw180-mono-block-power-amplifier-pair Based on their 250W at 8Ohm this would give just under 6A. Or just under 18A at 2Ohms. But this rating is for continuous power and in part relates to thermal overload. However the peak output per channel is over 100A. This gives plenty of head room for drum hits etc so they don't get clipped. Most amplifiers at JB Hifi etc effectively clip. Eg my 130W per channel AVR could not reproduce vocal dynamics, drum hits, double base string plucks, and low register piano strings as they were struck by the hammers, yet my 100W Metaxas stereo power amp could. I get it, there’s watts, then there are watts, then there are watts, but how does a novice work this out. Then many people talk about high current amplifiers, which appears even harder to work out Link to comment Share on other sites More sharing options...
Guest DrSK Posted April 4, 2021 Share Posted April 4, 2021 1 minute ago, JkSpinner said: I get it, there’s watts, then there are watts, then there are watts, but how does a novice work this out. Then many people talk about high current amplifiers, which appears even harder to work out It shouldn't be this hard. It is in part because specs are often hard to come by or not well defined. I'm rather cynical in that it is to many in the industries benefit that this is so. No good if we could simply just read the specs and buy exactly what we needed to match gear perfectly to our needs first time. Link to comment Share on other sites More sharing options...
Volunteer sir sanders zingmore Posted April 4, 2021 Volunteer Share Posted April 4, 2021 3 hours ago, stereo coffee said: Good question, I think a three way design and a two way design, and electrostatic where available should form part of all amplifier tests. The tables would begin to turn to show preference I think for simple crossover networks. A standard reference with two way being the LS /3,5a, i don't think any speaker other than maybe the ESL57 has received so much research. With 3 way lets plumb for a Yamaha NS1000 , and electrostatic the venerable ESL57 As long as testers were testing with each type ie 2 way, 3 way and a electrostatic is the emphasis, rather than resorting to resistor measurements, as they presently get too easily away with. It seems to me that there is huge variance even within your suggested subset of speaker types. Enough of a difference that the measurements for one set of two/ways (for example) would not be nearly as useful for any other. So unless you are considering that particular combination, you wouldn’t gain anything more out of this style of measurement. Link to comment Share on other sites More sharing options...
stereo coffee Posted April 4, 2021 Share Posted April 4, 2021 53 minutes ago, sir sanders zingmore said: It seems to me that there is huge variance even within your suggested subset of speaker types. Enough of a difference that the measurements for one set of two/ways (for example) would not be nearly as useful for any other. So unless you are considering that particular combination, you wouldn’t gain anything more out of this style of measurement. It does encompass quite a few owners of speakers, are yours 4 way ? what would you suggest ? Link to comment Share on other sites More sharing options...
Volunteer sir sanders zingmore Posted April 4, 2021 Volunteer Share Posted April 4, 2021 13 minutes ago, stereo coffee said: It does encompass quite a few owners of speakers, are yours 4 way ? what would you suggest ? How are my particular speakers relevant to this discussion?. Link to comment Share on other sites More sharing options...
Addicted to music Posted April 4, 2021 Share Posted April 4, 2021 (edited) 11 hours ago, stereo coffee said: There is very little marketing in a well written 342 page book, rather more so an enormous passion by its author to solve why voltage drive is far from complete, even at welding type currents. Would you like me to send it to you ?. you keep it for your own entertainment 11 hours ago, stereo coffee said: Having read it it you can reflect on the all too common welding approach, and perhaps better see why it is incomplete, vs current drive. A simple example is if we consider the resistance portion of a high impedance loudspeaker as being a simple form of current drive, as its resistance portion is partially allowing current driving of a normal voltage amplifier. The infamous L/S 3,5a provides 15 ohms, and is far from being unpopular. There is no doubting voltage amps are good at what they do when measured with a resistor, but the next step of making a current drive amp, is not too much of a departure, from what is already being provided, if we aspire to achieve it. Basic ohms law says for current to flow you MUST have a potential difference between 2 points in a close circuit... so you must provide a voltage that’s different between 2 points: Hence the term: Voltage Source How do ALL amplifiers work? They convert a small signal to a large one or a large signal from a small one! How do All amplifiers do this? Usually involves switching on/off a larger power supply with some 3 legged devices that’s attached to the PSU voltage rail in series with the transducer in series with ground . How do you have enough current capability to drive the hardest load that involves low impedance so the the 3 legged devices don’t self destruct and remain in SOA, you increase the number of devices on the end so it provides more current from the power supply voltage rails, a basic technique that’s referred as current sharing! note the output impedance of the device doesn’t change, its constant at all frequencies unless it’s neg temp coefficient mosfets .... Basic Electronics, all comes under the term: Voltage Source if you want to reduced the variation of reactance to get consistant volume at all frequencies, you need to correct the transducer reactance when designing/calculating the equivalent electrical circuit of the transducer to all the freq required which no loudspeaker manufacturers will do.. The different reactance in a LCR of a transducer is the job of the loudspeaker designer to correct not the electronics engineer for the electronics driving it! It’s like making a AC motor to run at 50hz 230VAC.... I bet you won’t be lodging a complaint to a power company to say that they have to do a power phase correction to the mains so your AC motor will work! You’ll be correcting it with an equivalent LCR circuit to the correct power factor for the motor so it runs on the mains....basic circuit theory! The mains has to drive many different components with different LCR! Why should it be different or unique to drive what you have. That’s basically what you’re asking as an analogy for speaker/amp design. Amplifier manufactures won’t make there products drive every speaker in the market, it wouldn’t be feasible, they make a range to cater for price and the more capable the more it costs. Usually the better capability involve a huge power supply with higher DC voltage rail and multiple 3 legged driver devices for current capabilities! See what I did there “high DC voltage rails “ Voltage source Edited April 4, 2021 by Addicted to music Link to comment Share on other sites More sharing options...
stereo coffee Posted April 4, 2021 Share Posted April 4, 2021 7 hours ago, Addicted to music said: " The different reactance in a LCR of a transducer is the job of the loudspeaker designer to correct not the electronics engineer for the electronics driving it! " Trouble is they don't usually meet to discuss each others requirements. Not discounting there may already be products seeing the current drive method as beneficial, particularly products from active speaker manufacturers. Rather generally presently physics wins, and few of us presently benefit, as EMF deriving from the voice coil motion, sums with the voltage applied to the driver, resulting in a mixture of the original signal and spurious signal corrupted by the speakers own mechanical electrical and acoustic properties. It goes without saying a current driven amplifier should also cater to the loudspeaker, and vice a versa. Link to comment Share on other sites More sharing options...
Recommended Posts