I've been auditioning speakers and the ones that have really interested me happen to be 6 ohms "nominal" impedance. My receiver, however, recommends speakers between 8 and 16 ohms. How significant of a factor will this be in affecting the sound? Could using 6 ohm speakers with my receiver damage the speakers? And finally, is there any way you can adapt either the speakers or the amplifier to each other? Thanks for any advice.
You can try whatever speakers you like with your receiver to find out if any problems exist. Other than that, there aren't many ways to guarantee a good match when you begin dropping the impedance of the speaker. Speakers are not constant loads and nominal can mean almost anything the manufacturer desires. A nominal 6 Ohm speaker can still dip low enough or have a difficult phase angle to cause too much current draw from the amplifier's power supply, or it might stay close enough to 6 Ohms and be sufficiently benign in its electrical phase to be usable with some amplifiers. Try the speakers and let the shop know your situation. They should have some idea what speakers you should obviously avoid. Make certain they have a return policy that fits your needs. If you don't listen a high volumes to music with heavy, low bass content, you might be fine with most 6 Ohm speakers. Listen at lower than normal volumes when auditioning at home with your receiver and check for unusual heat from the amplifier. Increase the volume and difficulty of music selection until you feel confident the amplifier and speaker can handle what you are likely to feed them. If at any time the amplifier runs extremely hot, shut everything down immediately. You run the risk of damaging the amplifier when you ask it to produce more current than the power supply can deliver. Unfortunately, since "nominal impedance" is a very vague term, even some 8 Ohm speakers can cause problems with an unstable amplifier.
Thanks for the advice. Seems like I should just try to find some 8 ohm speakers I like. However, could one just add a 2 ohm resister somewhere in the circuit, or is this a bad idea?
You can always pad the output of the speaker level. Most of the less expensive speaker selector boxes use nothing more than a 10 watt ceramic resistor in series with the outputs to provide some form of "protection" against an impedance load that would drop too low. This isn't the ideal solution if you are looking at new speakers for better sound. The receiver has to work through the resistor and you loose power. Then the resistor changes everything about the way the speaker works, not just the overall resistance (which is not the same as impedance). You also have no guarantee that other aspects of the speaker's crossover won't still present a difficult load to the amplifier. Resistors won't compensate for a difficult phase angle that still demands current from the amplifier or a crossover network that starts the amp into oscillation. Most mid-priced speakers designed to run off receivers have no real problems in either area, but I wouldn't take anything for granted here.
If the receiver uses internal rail fuses for the power supply protection, you could swap them to a fast blow type, but that would probably lead to constantly changing fuses if the mismatch of amp and speaker is severe or you like your music loud.
I would suggest the best route is to either do the audition process until you are satisfied the match is compatible or do some research into the actual impedance and eletrical phase load of the speakers you are considering. If the shop can't provide some information, possibly there is a review which gives the necessary information. If not, the speaker manufacturer or manufacturer's rep should be able to provide the right answer.
Many receiver manufactureres warn against low impedance loads as a protection for their own use. It gave them an out if the user continually blew up the amp. On the other hand, low impedance loads are not friendly to most receivers. If you have a difficult to drive speaker load, no matter what amplifier you are using and no matter the "nominal impedance", you should consider the consequences of the mismatch before you proceed.
thanks again Jan, I'm not sure I'll know what to make of that information once I get it. I may have to ask you about it once I get it. if this tells you anything: The speakers I'm considering are NHT absolute zeros. They're very small, surprisingly loud, however, they're low end response only goes down to 65 or 75 hz. so, no matter what, there won't be much bass. But you were referring to the source material being bassy, not what the speaker reproduces, right? I really appreciate all this help
eugene, My opinion, and it is ONLY that, is that you are on the right path looking for 8 ohm speakers that you really enjoy---it makes the power source/amplification issue much less important and you are less apt to have a bad result due to amplifier/integrated/receiver issues.
Speakers that are easier to drive are easier to match.
thanks again for all the input. Jan, what did you mean by "pad the output of the speaker level?"
I'm going to look for some 8 ohm speakers, but I'm also going to audition the NHTs and see how they gel with my receiver. I'll probably be getting back to you with more questions soon. thanks
"Padding" the speaker output merely places a resistor in series between the speaker outputs and the speaker itself. This has the effect of raising the mostly the resistance of the circuit, and to a lesser extent the actual impedance, but has some drawbacks that relate to what you are trying to do in searching for better sound.
First, as I said, resistance and impedance are not the same thing. Raising the resistance is just a plug in the dam that can eventually lead to cracks elsewhere in the system. One advantage of a solid state amplifier is the low output impedance of the (probably) direct coupled output transistors. The lower the output impedance of an amplifier, the more immune it is to the effects of the speaker's impedance swings as the amplifier tries to maintain flat frequency response while driving a speaker that fights back. Raising the resistance with a series resistor, and therefore to some extent the circuit impedance, of the output means the rising and falling curve of the speakers will have a more dramatic effect on the frequency response of the amplifier, resulting in deviations from flat response. How serious these variations are depends upon the speaker and the original output impedance of the amplifier. Generally these effects are small enough to be swamped by the room's contribution to the sound but you should still be aware that you are changing the sound to some extent.
Secondly, as I noted, you are now asking the amplifier to slug its way through yet another component in its quest to drive what is already a very inefficient system. Most dynamic speakers operate at only about 2% efficiency and placing a resistor in the line will take that level down slightly as more power is converted to heat and literally goes up in the air as wasted power.
The wattage of the resistor you would use means you will be using a ceramic resistor. Hardly what anyone would consider "high fidelity". Ceramic resistors became bad joo-joo in modern speakers quite some time ago.
Finally, if you place the resistor in series between the amplifier and speakers in an attempt to raise the impedance of the amplifier/speaker circuit, you will increase the resistance/impedance of the entire system (this includes the inductance of the ceramic resistor). This will sum with the crossover components and raise the value of all the crossover and driver components downstream of the ceramic resistor. The effect is to change the crossover values slightly and what the designer inserted as a low pass inductor to effect a crossover point at, say, 2,500 will shift slightly to some other point. The effect might be slight but could be critical to how the drivers operate as far as distortion produced in out-of-bandwidth frequency response.
How much all this affects what you hear is unpredictable as it figures into a global circuit comprising the amplifier's positive terminal out/resistor/speaker/amplifier's common ground terminal where the negative feedback of the amplifier is set. Negative fedback is used in an amplifier to, 1) maintain stability in the amplifier, 2) lower distortion, 3) set gain in the amplifier and 4) affect damping factor of the amplifier. By altering the overal circuit resistance/impedance you are altering all of these effects of the negative feedback circuit which sees something it doesn't "expect".
All this adds up to trying to avoid a resistor in line to "pad" the speaker output. Though, if you do, you won't be the first person to try this approach. It will work despite my comments regarding the effects of the "correction".
As I've noted in another thread, the amplifier is being fed a full range signal. Whether the speaker audibly reproduces this full range is some what irrelevant to the amplifier. Unless the signal is filtered before it reaches the amplifier, the amp can only reproduce what it is fed (at least, we hope so). So, it is using some of its power to do a 35Hz signal whether the speaker can go that low or not. Remember, speaker frequency response is merely an indication of what you are likely to hear from the speaker. The driver is fed the information and it does make an attempt at reproducing the signal, the signal merely falls outside the usable response bandwidth of the driver. An example of this is the pumping of the woofer in response to the subsonic frequency it is fed from a warped record (the reason for your subsonic filter on your receiver). The driver responds by pumping slowly to an 8Hz signal even though the frequency is well beyond its bandwidth of usable frequency response. So, you are using the power of the amplifier and you are asking the driver to move in response to an incoming voltage no matter whether you can hear the sound or not.
All this is somewhat esoteric and isn't of much concern to most listeners as they will never hear the difference all of this makes. But, you asked if it made a difference. Yep, it does.