[DMurphy]

I'm often asked how sensitive various designs are, and I always cringe a little when I give what I think is an honest answer.  People are used to seeing ratings in the 90+ territory, and when they see ratings from 84.5 dB to a high of 88 db (SongTowers), I'm afraid they'll conclude that the Salk and MurphyBlaster designs are in a different universe.  I just came across this quote from Tom Andry, who reviewed the SongTowers for EnjoyThe Music.com.  This is from a review of the Klipsch Icon V speakers in Audioholics:

"The Klipsch mains had replaced a set of Salk SongTower QWT's and the center replaced the RBH TK-5C center channel. The first thing I noticed was that I barely had to adjust the calibration on the Denon for the new speakers. The Salks are rated at 88dB efficient while the Klipsch are rated at 97dB. I should have had to make more than the 1-1.5 dB adjustment that I ended up making. This was true for the center and surround channels as well. I was puzzled as the Klipsch speakers didn’t seem to be as efficient as indicated in the specifications."

So get out the salt shaker when you see super high sensitivity ratings. 

[zybar]

Dennis,

No doubt many vendors are "generous" with their sensitivity ratings.  That's why I love the fact that Stereophile performs sensitivity measurements to determine how accurate the vendor's claims are.

That being said, it isn't just the lower sensitivity on the HT3's that make them a little tougher to match up with amplification; it is also that they required (from my standpoint, in my systems) "beefier" amps that could really drive the 10" woofer. 

I didn't find the same issues with the SongTowers. 

Lastly, I have other speakers in house that really do have measured sensitivity in the mid to upper 90's and it does require me to substantially turn down the volume when they are inserted.

George

[BPuckett]

Well, I suspect that Klipsch places the microphone on the horn-loaded tweeter axis when they make their measurements, rather than on the more realistic axis that splits the difference between the tweeter and the other non-horn-loaded drivers.  Doing so would give a higher sensitivity rating than would be experienced during actual listening conditions where the listener's ears are not aligned with the tweeter axes.  Maybe I'm wrong, but I think that's what's happening.

Bob

[funkmonkey]

Very interesting Dennis.  I also found it interesting that when I was auditioning speakers, my personal favorites tended to be the least sensitive designs.  That is something that I noticed only after I had auditioned quite a few designs and was going back for second or third sessions.  Up until that point, I had done my best to ignore all the stats, letting my ears decide what they liked best.  Makes me wonder if there are any inherent advantages to a more demanding load, provided that it is handled properly...

[DMurphy]

Dennis,

No doubt many vendors are "generous" with their sensitivity ratings.  That's why I love the fact that Stereophile performs sensitivity measurements to determine how accurate the vendor's claims are.

That being said, it isn't just the lower sensitivity on the HT3's that make them a little tougher to match up with amplification; it is also that they required (from my standpoint, in my systems) "beefier" amps that could really drive the 10" woofer. 

I didn't find the same issues with the SongTowers. 

Lastly, I have other speakers in house that really do have measured sensitivity in the mid to upper 90's and it does require me to substantially turn down the volume when they are inserted.

George

Hi   There are super sensitive designs out there.  But on average, I think you'll find sensitivity ratings on the high side.   The HT3 is (or at least was) genuinely demanding when it comes to amplifiers.  First, it is not very sensitive.  Second, as is true of most 3-ways with beefy woofers crossed over passively below 500 Hz, the impedance swings pretty low in the midbass. I've addressed the latter issue in the newest HT3's--the minimum impedance is 1.5 ohms higher.  But SET 8 watt amps still need not apply. 

[DMurphy]

Well, I suspect that Klipsch places the microphone on the horn-loaded tweeter axis when they make their measurements, rather than on the more realistic axis that splits the difference between the tweeter and the other non-horn-loaded drivers.  Doing so would give a higher sensitivity rating than would be experienced during actual listening conditions where the listener's ears are not aligned with the tweeter axes.  Maybe I'm wrong, but I think that's what's happening.

Bob

Well, in theory sensitivity ratings are supposed to be made at 1 kHz, which would be below the tweet's operating range.  I really don't know how Klipsch came up with 97 dB. 

[DMurphy]

Very interesting Dennis.  I also found it interesting that when I was auditioning speakers, my personal favorites tended to be the least sensitive designs.  That is something that I noticed only after I had auditioned quite a few designs and was going back for second or third sessions.  Up until that point, I had done my best to ignore all the stats, letting my ears decide what they liked best.  Makes me wonder if there are any inherent advantages to a more demanding load, provided that it is handled properly...

Sensitivity is usually dictated by the woofer, and the higher the sensitivity, the less deep it will reach, all else equal.  So in theory, the only systematic difference you should have heard would have been in bass extension.  A more demanding load, again all else equal, is a bad, not a good. 

[zybar]

Hi   There are super sensitive designs out there.  But on average, I think you'll find sensitivity ratings on the high side.   The HT3 is (or at least was) genuinely demanding when it comes to amplifiers.  First, it is not very sensitive.  Second, as is true of most 3-ways with beefy woofers crossed over passively below 500 Hz, the impedance swings pretty low in the midbass. I've addressed the latter issue in the newest HT3's--the minimum impedance is 1.5 ohms higher.  But SET 8 watt amps still need not apply. 

Dennis,

Glad to hear that the minimum impedance is a little higher on the newest HT3's compared to the ones I had.   

That improvement might allow a few more tube amps to mate with the HT3's.

George

[BikeWNC]

Dennis, can you say when the impedance change was made?  Was it when the new woofers were added to the HT3 or will my speakers with the TC Sounds woofer which I received in May be part of this change?

[DMurphy]

It was just made last month as part of the redesign for the new woofer.  But if you're not having any amp difficulties with your HT3's, then the change is pretty much academic. 

[BikeWNC]

It was just made last month as part of the redesign for the new woofer.  But if you're not having any amp difficulties with your HT3's, then the change is pretty much academic. 

No, not having problems, but I have a tube amp I wanted to try with the HT3s and was interested.  Can someone describe what these problems might sound like if an amp is not able to handle the load of the speakers?  I have a 50w tube amp, which I know is on the small side for the HT3s but if I plan to keep to moderate volume levels what issues should I look for?  Thanks.

[jsalk]

I also found it interesting that when I was auditioning speakers, my personal favorites tended to be the least sensitive designs.

This is not too surprising.  Most speaker manufacturers today look for small drivers that play deep in a small cabinet. 

As Dennis pointed out, all things being equal, higher sensitivity drivers do not play as deep.  They also tend to require larger cabinet volumes.  So, if you were a driver manufacturer, would you put all your R&D into higher sensitivity drivers that don't play as deep and require larger cabinets?  Not if you want to sell very many. So it is not surprising that the latest state-of-the-art drivers tend to be lower sensitivity drivers.

There was a time when amps were not that powerful and there really wasn't a choice.  But today, watts are cheap.  So if you want a speaker the the best-of-breed drivers, you will probably end up with a lower sensitivity speaker requiring more power. 

Of course, another alternative is to use multiple drivers, but you end up introducing all kinds of comb filtering issues in the process. As I've said many times in the past, speaker design is all about balancing trade-offs.  There is no free lunch.

- Jim

[Nuance]

Thanks for the info, Dennis.  How was your vacation, by the way?

I never believed Klipsch's sensitivity ratings, as I never believe any manufacturer's listed specs.  I do, however, trust you and Jim because you have no reason to lie and actually have the tools to measure them.  You guys rock!    

My ST's are rated at 88db, right (did the ribbon change that)?  Well, they rock loud, and I can't imagine needing them to be a higher sensitivity.  I had them at 100dB on the Radioshack meter and they stayed clean without distortion, compression or the woofers bottoming out.  They had more room to play, too.  Nice!

[oneinthepipe]

deleted.  Jim answered my question in a prior post that I didn't see.

[funkmonkey]

Thanks Dennis & Jim, you guys are spearheading my education. 

[zybar]

I also found it interesting that when I was auditioning speakers, my personal favorites tended to be the least sensitive designs.

This is not too surprising.  Most speaker manufacturers today look for small drivers that play deep in a small cabinet. 

As Dennis pointed out, all things being equal, higher sensitivity drivers do not play as deep.  They also tend to require larger cabinet volumes.  So, if you were a driver manufacturer, would you put all your R&D into higher sensitivity drivers that don't play as deep and require larger cabinets?  Not if you want to sell very many. So it is not surprising that the latest state-of-the-art drivers tend to be lower sensitivity drivers.

There was a time when amps were not that powerful and there really wasn't a choice.  But today, watts are cheap.  So if you want a speaker the the best-of-breed drivers, you will probably end up with a lower sensitivity speaker requiring more power. 

Of course, another alternative is to use multiple drivers, but you end up introducing all kinds of comb filtering issues in the process. As I've said many times in the past, speaker design is all about balancing trade-offs.  There is no free lunch.

- Jim

Jim,

Watts are cheap depending on what type of amp you want to use...big tube watts are still very expensive. 

I tried multiple tube amps with the HT3's and while I preferred their performance over all the ss amps in the mid-range and highs, I never could get the woofer control and bass I wanted with tubes.

BikeWNC, if you listen at low spl's (70's to low 80's) and don't play music with lots of bass (vocals, folk, small jazz bands, etc...) you shouldn't have any problems with running tubes.  In my experience, it wasn't that the tubes amps I tried didn't "work" or couldn't handle the load, it was simply that the bass wasn't as good sounding as I know it could be.  When you hear a woofer that is underdamped or not controlled properly by the amp, the bass can become slow, bloated, lack detail, can be wooly sounding, and just generally uninteresting.  If you crank up the volume with your 50 watt amp, you also might clip the amp.  With a tube amp, this will sound like distortion and should be fairly easy to recognize.

George 

[JoshK]

I read a trick about an easy way to estimate whether you would like a tube amp with a particular pair of speakers when you usually use SS amps.   Take a 1-2ohm power resistor and put it in series with the speakers and then use your SS amps.  The effect is similar to that of a tube amp.

Many PP tube amps have a source impedance in the 1.5 ohm range.  Speaker designers typically assume zero source impedance, which is typical practically speaking for SS amps with ungodly amounts of feedback.  The impact of a non-zero source impedance is that it changes the tuning of the box, thus impacting the Q.  This can cause there to be a peak in the bass/midbass area and high Q which is underdampened bass.  One can also design a box for a non-zero source impedance so that the Q is optimal for the tube amp (or whatever else) but speaker designers almost never do that, since it would require knowing what the end user was going to use. 

Taking a typical SS amp and placing a 1-2 ohm power resistor in series with the speaker simulates the tube amp's impact on the bass tuning.  Its a cheaper experiment then selling your SS amp for a big expensive tube amp and finding out you don't like the result.  Sorry it doesn't exactly help your situation if you have a tube amp already and are considering the HT3s, but maybe you can visit an HT3 owner and bring some power resistors with you.

[carusoracer]

Great thread!  It is nice to hear from the Men behind the Curtains

I tried the 50 Watts/Ch on my HT3's. As prefaced, if you are not playing at loud volumes and heavy techno or bass heavy tracks you are fine. It took me a few minutes to readjust to the sound, but Oh the glorious sound of the EL34 midrange is quite intoxicating, especially on a cold winter night basking in the warm glow of tubes.

Nice trick JoshK with the resistors.

Any pictures of the new HT3 woofer? I plan attending BRM's GTG, weather permitting, will a prototype be there?

[BikeWNC]

The above info is good to know.  I just didn't want my amp to meltdown trying to push the HT3s.        I figured it would be a listen and see experiment and having an alternative/different sounding amp can be a nice change with the right music.  Thanks.

[TomW16]

I tried high sensitivity single speaker drivers and I didn't get the full spectrum of sound that I wanted.   I, therefore, traded sensitivity for watts and now really like the sound.

Cheers,
Tom