4" (3.5"-5") high efficiency wanted please

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I saw some crude FR graphs years ago on these but I can't find them now. The measurements did show a reduced output in the range you mentioned Badman. Actually I seem to recall it extending a bit higher, but this is an old memory. I'd really like to see some good measurements on it.

Dan

100-200Hz is where the 109dB output and 99dB sensitivity requirements fall apart. At 300Hz it's a stretch but possible. Our OP has backed off of 100Hz, and conceded to reviewing lower sensitivity drivers, both of which make his goal more attainable, but doesn't seem to be willing to properly define requirements within the ability of drivers. You have to give up a lot of sensitivity to get sufficient high excursion motors and suspensions, allowing for maximum output in these ranges, but if you move the crossover to 300 or 400 you can minimize the excursion requirements.

If such a driver were to be possible, those trying to squeeze maximum excursion from tiny drivers would have achieved it. But high sensitivity comes at the cost of high Fs and low Xmax. It is this way now and always will be without a major change in motor design or some new, advanced materials for the magnetic circuit.

http://www.creativesound.ca/pdf/CSS-FR125SR-dataS.pdf can just about manage the output requirements from 150Hz and up but is nowhere near high sensitivity.
 
From Post #1
"Am searching for the 4" (3.5"-5") that will produce about 107dB or more on less than 50 watts from somewhere near 100Hz up to 3kHz or so."

This pretty well shows badman cannot read. 50 watts is +17dB over the sensitivity with allows for drivers down to 90dB for 1 watt input. This also shows for drivers up to 5" nominal. All of this was "more or less" so a couple to a few dB in either direction could still fit. Badman is a bad man in that he has taken the most ridiculous extremes and then said I was making those my requirements... something never defined by me and only imagined by him. Again he brings a lot of negative to the thread. A thread bully and not in the spirit of DIYaudio or this otherwise very informational thread concerning high efficiency small drivers.

Everyone else, let us keep this list and posting of efficient and/or loud drivers coming! These type drivers have become unpopular in recent years for many reasons however there are still some out there so keep this fine work going!

Thanks------ :)
 
Last edited:
From Post #1
"Am searching for the 4" (3.5"-5") that will produce about 107dB or more on less than 50 watts from somewhere near 100Hz up to 3kHz or so."

This pretty well shows badman cannot read. 50 watts is +17dB over the sensitivity with allows for drivers down to 90dB for 1 watt input. This also shows for drivers up to 5" nominal. All of this was "more or less" so a couple to a few dB in either direction could still fit. Badman is a bad man in that he has taken the most ridiculous extremes and then said I was making those my requirements... something never defined by me and only imagined by him. Again he brings a lot of negative to the thread. A thread bully and not in the spirit of DIYaudio or this otherwise very informational thread concerning high efficiency small drivers.

Everyone else, let us keep this list and posting of efficient and/or loud drivers coming! These type drivers have become unpopular in recent years for many reasons however there are still some out there so keep this fine work going!

Thanks------ :)

ALSO FROM POST 1

Hello All-

Yes I know there are lots of about 4 inch drivers making about 90dB for 1 watt that will handle 100 watts or more but that is not what is needed. Low power consumption is critically important for this design so high efficiency drivers are needed.

Am trying to build a LOUD speaker. Have a 98dB woofer and a 96dB tweeter but have not found a midrange. Looked at hundreds of possibilities and not much so far. No horns or compression drivers will be used.

From the same post.... You very clearly stated that 90dB was NOT acceptable.
 
Xmax is not hard clipping as in a solid state amp

Xmax is pretty much and simply the difference between the voice coil length and the gap length (height on some specs) divided by 2. The Xmax number is a guide to the motor and is not a guide to the suspension characteristics or to the linearity of the system. It is safe to say when Xmax is exceeded then distortion will increase. It is equally safe to say when the signal size is increased above 1 microwatt the distortion will increase. Almost every driver ever tested here increases distortion with increasing drive signal. Sometimes there is lots of distortion long before Xmax is reached and sometime not, depending on a lot of different things about a particular driver. Occasionally, some drivers have even decrease distortion above Xmax. Xmax does not represent hard clipping like in a typical solid state power amplifier where the amplifier has no more ability to output greater voltage because of the "clipping" effect. Drivers do clip which can easily be heard as a snap or clack sound when some moving part of the driver hits some part of the frame or magnet which is not moving. That is driver clipping and is signal size (voltage) and frequency related and completely unrelated to Xmax. Many believe Xmax is some hard boundary like clipping. It is not. Xmax is not even a very good indicator for distortion onset in many drivers. This is true for numerous reasons. Because in almost every driver increasing signal means increasing distortion the hobbyist or designer has the chance to make the decision of how much distortion is okay. Based on a lot of sub woofers tested here 30% at 100dB and 40Hz is not at all uncommon. Not saying this is good or bad, just saying it is not uncommon. Exceeding Xmax does not necessarily lead to rapidly increasing distortion at a particular frequency because there are so many factors involved in the motion of the cone and suspension the single factor of Xmax plays just a part of the entire event. Those who use Xmax as an upper limit know little of the reality of driver design and loudspeaker results in the real world. These are the people who never built anything and use (hopelessly inadequate) computer models to do their thinking for them. In the computer models there is no place to put in the increasing resistance from air motion around the windows, the very non-linear effects of the spider, the effect that the voice coil does not produce linear force with position along the pole piece or any of numerous effects known by the skilled and experienced designer. Only those who assume some model is actually the real world make these very common mistakes and DIYaudio is (sadly) full of people who believe the computer answer is very accurate.

As a pioneer of computer modeling of speakers back in the middle 70's I learned how handy models could be as a general guide to driver understand and behavior. I also learned how worthless trying to perfect an exact model was. Never could the model even come close to what a real world speaker did except at milli-watt levels. At the upper limits to cone motion (above Xmax) and upper limits of power these models fall apart like a house of cards in the wind. The models fall apart because the complexities of the event are not at all modeled well by the simple math used in the computer models.

In conclusion, those who use and believe in models for the maximum motion available and for maximum power are simply fooling themselves believing those results have anything but a very casual relationship to what happens in the real world. Further, it is up to the speaker system designer to determine how much distortion at what SPL and frequency is acceptable. This is again not the choice of Xmax or a computer but a designer and real world measurements with real drivers as these things are impossible to model at this time. With these things in mind the truth about Xmax and computer calculated driver distortion are suddenly much less important than a mostly worthless computer model might suggest. This is why drivers are chosen by designers and listeners and never by computers. Spewing forth and extolling the benefits of the computer model and Xmax at the upper limits of power and driver motion is only done by the fool. And there always seems to be a fool who has little to no real world experience shouting about the computer modeled results on DIYaudio claiming expertise in mouse clicking. Impressed? I am not.:D
 
Let's see a driver that works the way you claim then. How about a response and distortion graph (at 109dB) for that panasonic? No designer worth their salt would use an evenhung driver or near to it (as is the case for all high sensitivity 4-5" drivers) with half its coil out of its gap, and that's exactly the case with real world drivers of this type. You have not shown any evidence for your position. Do so. Linked is a Klippel Analysis of a quality Seas 4" driver. This is a MEASUREMENT not a model. Not surprisingly, things go to heck above Xmax, with suspension, BL, and inductance all working together to clip the waveform.

HTML Report - Klippel GmbH

Spewing forth and extolling the benefits of the computer model and Xmax at the upper limits of power and driver motion is only done by the fool. And there always seems to be a fool who has little to no real world experience shouting about the computer modeled results on DIYaudio claiming expertise in mouse clicking.

Enough with the bragging and attempts at insults. You're not impressing anyone. I have enough real world experience to not need models for anything but confirmation of my intuition. You are completely off base in your arguments. Stick to the discussion, and give a counterexample if you have so much experience in operating drivers HUGELY outside of Xmax. I wouldn't argue if you were talking about 5% or even 10%, you're talking about exceeding it by as much as 1000%, but more importantly than that ratio, exceeding it by 50% of gap height for the type of motors we're discussing
 
more of the same nananana

Skip the next paragraph for those interested in audio as badman insists on this stupid engagement.

badman- Have no idea why you dwell on this Panasonic. There is no modern driver available so far which satisfies my specific needs. The point of this thread is to identify "possibilities" and several are somewhat close. Have not laid down any hard and fast rules for the driver either other than "let us all check what we see and post on this thread." You have feebly attempted to put me in a corner about a driver I like but to what end? That driver is no longer available for a very long time, like 30 years, so what IS your point? If nothing else comes of this effort by so many good people other than a guide to some efficient smaller drivers that will be enough to assist the DIYaudio community as a handy reference source. Possibly there is a driver which suits my needs also. There used to be drivers that did but those are long gone. Have no idea why you insist on this inane challenging as this thread is just getting going and there has been little time to even receive the first of the existing drivers let alone any prototypes. Of course reality says a new driver or appropriate driver will take longer to find or develop than an empty headed refrain about "prove it now" or "you are wrong." badman- you are still a bully and still write like the jerk you must be in real life. You made you point so move on and allow this thread to grow. You keep repeating the same thing over and over as though that will change something making you more knowledgeable or intelligent. It will not. Do you have any idea how annoying you are? Of course, you, as several others seem to enjoy thread wrecking. I am hoping very much as so many sites on the web are now doing... kicking bullies out of the fun and games and soon my wish is this will expand to the forums.

For the good people of DIYaudio- keep those driver finds coming. Wonderful parts are found in the most unlikely places and your help is more than appreciated. Several great off topic finds have resulted from threads reading and your kind help!!:):) Thank you. =SUM
 
Don't mean to pile on, but I'm siding with Badman on this one.

I would not take Panasonic's claims of efficiency too seriously. Why? Mass-market drivers aimed at non-hifi markets have specs that are largely fictitious. Last time I was in India, I saw boomboxes in the stores that claimed a ludicrous 480 watts of "PMPO" power - and it was powered by six D cells! Maybe it would put out 480 watts if you dropped it across a 3-phase power line - for about a millisecond - but not any other way.

Speaker specs for the non-hifi crowd (autosound, low-end PA, DIY disco) are similar. If you make a really cheap, really peaky speaker, yes, the peaks will be really up there - 10 to 15 dB above the flat region. If the speaker is peaky enough, there will be no flat region at all, just a series of hills and dales, and naturally the claimed efficiency is right at the top of the peaks.

I also don't understand the motivation for seeking a small, high-efficiency driver. Why? Small means low efficiency, and more importantly, low power-handling. A high BL product (high magnetic force in the gap) and a large, lightweight cone are the time-accepted ways to get high efficiency and high power handling. This has been true since the Rice & Kellog driver was invented in the late Twenties. That's the reason that high-end console radios in the Thirties used 12" and 15" cones; amps weren't that powerful (10 watts was a monster amplifier), but the speakers were quite efficient.

Or, use horn loading, which multiplies efficiency (in the horn passband) by 10 dB or more. But the size of the mouth controls the LF limit, and there's no way around that. A horn that goes down to 100 Hz is the size of a refrigerator, while a horn with a 4" mouth is basically a supertweeter.

A 99 dB/meter/watt speaker is a high efficiency speaker by any standard, midway between a Klipsch LaScala and Cornwall in efficiency, and neither is all that small. If you want something the size of a BBC LS3/5a with its 5" midbass driver, you're going to have to accept 85 dB/metre/watt. The nearfield Yamaha NS10's seen in studios all over the world have similar efficiencies. Both run out of headroom around 95~98 dB SPL, but work just fine at a 1 meter listening distance.

Take your pick: efficiency or small size. You don't get both, unless you compromise in both directions. That gives a you a typical bookshelf speaker with efficiencies in the 87~90 dB range, and midbass drivers that are 6.5" to 8" in size. The prototypes for all of these were the AR-2 and the KLH Model Six, introduced in the late Fifties. The speakers made a half-century later have slightly better cone materials and much better tweeters and crossovers, but are basically the same. There have been no breakthroughs in efficiency, sorry.
 
Last edited:
I agree. The BBC/Spendor LS3/5a is probably closer to 82 dB/meter/watt, in the range of classic electrostats like the Quad ESL57. Things really start to come apart at 95 dB SPL, and more amplifier power just destroys the speaker all the faster. These speakers are about the subtler things in life.

You want loud, go big. That's all there is to it. Think about the size of the speakers behind the movie screen. Do you think cheapskate theater owners would spend all that money on not one, but three huge loudspeakers if there was any other way to fill the theater with sound?
 

Attachments

  • Theater_Speakers.jpg
    Theater_Speakers.jpg
    80.9 KB · Views: 239
Last edited:
I also don't understand the motivation for seeking a small, high-efficiency driver.
Different strokes for different folks? You say tomato and I say potato? Variety is the spice of life?
If Sumaudio wants to pursue a design that utilizes small highish efficiency drivers, then who is he hurting? Just because you and some others in this thread don't agree with Sum's design choices, that doesn't make you right and Sum wrong, and it's a shame that some in this thread have nothing better to do with their lives than to direct ill-will and rancor towards Sum because they disagree with what he has to say. Good grief, live and let live and let the guy have his fun - if anyone thinks that what Sum has to say is **, then show some maturity and don't click on this particular thread and save yourselves some aggravation.
 
Hello All-

Am searching for the 4" (3.5"-5") that will produce about 107dB or more on less than 50 watts from somewhere near 100Hz up to 3kHz or so. I have this old Panasonic 4" nominal that is 97dB for 1 watt with Fs of 80Hz and it will handle 10 watts RMS which is 107dB at 10 watts.

Does anyone know of anything like this or better out there? Two of us have been looking for days and best thing so far is this:

Atlas Sound - FC104

This Atlas will do 104dB on 10 watts which is the power limit for this driver. Would really like 110dB on 50 watts as a goal. 50 Watts is +17dB over the 1 watt rating so a 93dB sensitivity speaker handling 50 watts would work. Really want something more like 96dB for 1 watt and handling 50 watts. That is 113dB and really plenty loud.

Yes I know there are lots of about 4 inch drivers making about 90dB for 1 watt that will handle 100 watts or more but that is not what is needed. Low power consumption is critically important for this design so high efficiency drivers are needed.

Am trying to build a LOUD speaker. Have a 98dB woofer and a 96dB tweeter but have not found a midrange. Looked at hundreds of possibilities and not much so far. No horns or compression drivers will be used.

Here is the tweet and woof for those interested:
Vifa DX25TG09-04 1" Fabric Dome Tweeter | Parts-Express.com
Eminence Sigma Pro-18A-2 18" Cast Frame Driver | Parts-Express.com


There are so many here on DIY with so much experience maybe someone will be able help.:)

Thanks in advance for suggestion!

=SUM

I am too lazy to read this entire thread, so some of these may have come up already.

First off, Faital Pro sells a 3" woofer that's designed for line arrays which has the highest efficiency that I could find in that size. I purchased a box for a project that never got off the ground. The build quality is very nice.

Second, JBL sells a number of "so-called" 5" and 4" drivers that have a cone that's about 3" in diameter. The frame is very very beefy, and this explains why they're called 5" woofers. The JBL 400GTI and LE120 come to mind.

The thing that all of these drivers have in common is that they use unconventional motors. Either neodymium or alnico. There is no free lunch, and the only way you're going to increase the efficiency to astronomical heights is by concentrating a lot of energy into the gap, and using the lightest cone and surround humanly possible.

As for your complaint that no one makes drivers like this anymore -

Of course they don't. Woofers aren't very well behaved when the cone weighs a few grams. Most engineers would trade sensitivity for bandwidth and improved distortion numbers.

The main reason that I've used drivers like this is that I do a lot of car audio work, and in the realm of car audio, enclosures are very very very very small. So the very high efficiency of these small drivers is a way to get the output of a 6" woofer in a box that's much much smaller.

But there is a price to pay. If size was not a factor, I agree with Badman(?), and would use a larger and better behaved woofer.
 
>>> If you make a really cheap, really peaky speaker, yes, the peaks will be really up there...

Yup, and choosing the highest of the high peaks and spec'ing your driver there is really not being honest.

From the original poster:
“Efficient drivers are a forgotten technology and manufacturing practice.”

I looked at the FR for the Panasonic and it does tickle 100db. Regardless, its response drops after 300hz so it too does not make the grade. I was and am still hoping someone will find a hidden gem for us to build systems around. I’m tired of sub 90db four inch drivers! 99db sounds FUN!

Right there on page one of this thread the best advice is probably given… from badman.

“Change your requirement to 300Hz- 3k and your goal is much more likely...”

>>> JBL sells a number of "so-called" 5" and 4" drivers that have a cone that's about 3" in diameter. The frame is very very beefy...

But they may not be as good as those old school, inexpensive Panasonic.

This is one of my favorite threads! It's helping us understand current limitations of drivers this size. And the back and forth debate makes it entertaining to read.
 
Here is another interesting one. Easy 92dB with power rating to +17dB or 109dB output, yummy.
4 1/2" 4 OHM SAMMI SR100A25 MIDRANGE

At any single frequency below about 200, with sufficient power to achieve 109dB in the passband, your coil is ENTIRELY out of the gap. Even with a 200Hz highpass, this happens around 200.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.