Zaph|Audio

5.5" Driver Comparison

About the Test

All response curves are with the drivers on an infinite baffle. All drivers are countersunk, and all baffle openings are chamfered in back for smooth airflow. Don't forget that in the real world, baffle step and diffraction will affect the curves. These curves are all far field, taken with a 10ms gating window. Below 600 Hz, a near field measurement is merged in for accuracy. Response curves include impedance curves at the bottom.

Close-up impedance curves are provided to show more detail that may not be seen in the response curve. Any variation from complete smoothness is generally a bad thing. Small bumps in the impedance curve, no matter how small, will always mean either linear or non-linear distortion. There may be a peak or dip in the response curve, or an increase in harmonic distortion to coincide with the impedance bump.

Cumulative Spectrum Decay (CSD) charts are 1/12 octave smoothed. Time slices are taken up to 3ms with the start marker right at the beginning of the impulse. All levels are equalized to 30dB to accurately show comparitive energy storage regardless of individual driver efficiency.

Harmonic distortion charts are taken semi-far field with the mic directly on axis. The far field method shows a more realistic top end distortion but does allow some room effect to show up also. The same conditions apply to all drivers. Amp output level was adjusted for each driver so that the fundamental is 96dB at 1/2 meter. Harmonics F2 thru F5 are tracked. F2=Green, F3=Blue, F4=Purple and F5=gray. In general, the higher the order of harmonic distortion, the more offensive the sound. The range is 20Hz to 5kHz and the level is +30 to -110 dB. The only tradeoff with doing HD tests far field are room effects. For normal DIY'ers like me, anything less than a full anechoic chamber is going to introduce some inaccuracies in the results. Important: there is a small room effect showing up at 240 Hz in all measurements.

All T/S parameters are actual measured, not from factory data. All Vas numbers are done delta compliance since that method is more accurate than delta mass. Also note that the SPL number on the T/S parameters is an efficiency number based on one watt. Additionally, this number is calculated rather than actual. The SPL shown on the response graphs is a sensitivity number based on 2.83 volts and is actual rather than calculated. If you want to "believe" one, the SPL shown on the response graphs would be best to use. Also note that the Le number on the T/S parameters is not a good comparison point between software packages because of different calculation methods. The Le is often interpeted by software particularly poorly when the motor is complex with substantial faraday sheilding. For accurate impedance vs frequency, look no further than the driver's impedance curve rather than an arbitrary Le number. Flatter is better.

Occasionally, I run some tests to prove T/S accuracy and I did so this time around. The test starts by obtaining T/S parameters through my normal methods. Then, I take those parameters and insert them in a box modeling program. [image] Next, I take the exact same woofer, build a box of the same size I just modeled, put the woofer in it and do a near field measurement. [image] Finally, I lay the measured on top of the modeled to see how they compare from 500 Hz and below. [image] I do this with a few different woofers to make sure I have a good grasp of reality.

All parts of this test are done with Soundeasy 12.0. This is a modeling package with a lot of extra testing applications included. It is not a full acoustic measurement system such as CLIO or Praxis however, so take these results with a grain of salt. What is important however is that all of these drivers were tested in the same way. Because of this, my test results should not be compared with tests done elsewhere by other people, but comparisons within this group of drivers are perfectly valid. That's why tests like this are best done in groups.

There are a few types of tests that are important, but not shown here in the interests of brevity. Keep this in mind and know that these results need careful interpretation.

Commentary and review: 5 Star Rating System, drivers sorted by cost. Cost - Driver cost per one unit in USD. Prices and availability updated 12/14/2008. Usability - This is a function of both linear and non-linear distortion combined. The standard of reference is use in a "normal" stand-alone 2-way speaker with a "normal" tweeter. Some examples: Subtract points for a response curve that requires extra crossover components to fix or peaks in harmonic distortion that push the tweeter crossover point below a normal frequency. Add points for drivers that have particularly wide usable bandwidth or easily controlled response curves. Build Quality - This is a combination of fit and finish, along with well engineered features. Some features may help performance in ways not covered here by the normal set of tests. Some examples: Subtract points for glue gushing out at the edge of the surrounds, an off center dustcap or a rough frame casting. Add points for venting behind the spider, clean tinsel lead connection, or an estimate of low air restriction through a frame design. Consistency - Difference between units. This is primarily a comparison of a response curve and impedance plot, and assumes that I have two units to test. If I have more experience with a driver, this rating will have a greater level of accuracy. If I only had one to test, this category will have a "Not Rated." Value - This is a comparison of cost to everything else combined. Remember that aside from cost, the ratings below my subjective opinion. I am one person, namely John Krutke of Zaph|Audio. The ratings are provided for those who wish to know the opinion of this web site's author. I encourage people to use the objective test data, consider their own needs and form their own opinions. I also reserve the right to change these ratings, as I think about things over time and gain more consistency data. Seas Excel W16NX-001 Cost: $219 Usability: Build Quality: Consistency: Value: Comments: This is a Seas Excel with a unique flange designed to get the most radiating area into a 15cm basket. Typically 25% more Sd than most others in this group. Best harmonic distortion below 200hz in the group, with good extension and good Xmax to back it up. Requires a comparatively small vented enclosure. Typical but well controlled metal cone breakup at 5.2kHz. Minor response hiccup at 500 Hz, but not nearly as bad as the L16 Prestige version. Will sound the best crossed over about 1700 Hz LR4. Tested December 2008. Seas Excel W15CH-001 (Magnesium Hexadym) Cost: $219 Usability: Build Quality: Consistency: Value: Comments: This driver is essentially the same as the W15CY-001, but with a neo hex magnet motor. As far as frequency response and distortion, it's barely different at all. I did notice a slight difference in the breakup node, but I suspect that was unrelated, and I only had one sample to test for consistency. One might say that there is really no benefit to these hex magnet designs, but in reality, I would expect better cooling of the voice coil under heavy usage conditions. Better cooling means a more consistent resistance under load, which means lower compression and less change in the transfer funtion. Is it better driver overall due to the hex magnet? Yes, a wee bit. Better value? No, the cost increase for the hex magnet is a little high. Tested May 2006. Scan Speak Revelator 15W8530-K00 (coated low Qts) Cost: $212 Usability: Build Quality: Consistency: Value: Comments: The build quality and attention to detail in the Scan Speak Revelators is second to none. The suspension consists of variable thickness surrounds and a spider design that has variable rib size and spacing, rather than the plain old spiders seen in all other drivers in this roundup. The motor appears to have the right amount of copper in all the right places, giving this driver the lowest harmonic distortion in the group. The frequency response is not quite as smooth as one could hope for however, but extended and usable on both ends. Qts is still rather high for what Scan speak considers a "low Qts" version of the driver. Tested May 2006. Scan Speak Revelator 15W8531-K01 (non-coated) Cost: $212 Usability: Build Quality: Consistency: Value: Comments: (discontinued) For all practical purposes, this driver is the same as the 15W8530-K00, but without the coating and with a smaller magnet. Unfortunately, without the coating, response appears to be a little more ragged, and the smaller magnet means an even higher Qts, likely limiting this driver to a sealed box. All this combines to move the usability rating down a notch. Honestly though, I still have mixed feelings about the coating, because sticky coatings eventually become furry dust covered coatings. Tested May 2006. Jordan JX92S Cost: $180 Usability: Build Quality: Consistency: Value: Comments: Very wide and smooth frequency response. This is a full range driver that actually sounds good run full range with a simple baffle step compensation circuit. When you get 10 degrees off axis however, the high end disappears from beaming. Decent overall performance, but a poor value. A pair of better sounding 2-way systems could be built for the cost of a single JX92S. If you're convinced there's some sort of magic to a full range driver, this is one of the best, far surpassing most others I've come across. Tested March 2008. Seas Excel W15CY-001 (Magnesium) Cost: $162 Usability: Build Quality: Consistency: Value: Comments: This is Seas' definitive high end driver. Response within the usable band is flat line smooth, and harmonic distortion is excellent, though a notch down from the Scan Speaks and the Peerless Exclusive. The breakup node is easily controllable with a simple RLC notch, though I consider the optimum crossover point of this driver to be around 2kHz, maybe higher depending on how much that 2.3kHz 3rd order harmonic distortion bugs you. The 5th order harmonic at 1.5Khz is low enough to be off the audible radar. It's not the cheapest 5.5" around, but it's probably worth the extra cost. Note: a couple stars off consistency for a wandering breakup node over the years. Old/Other tests available. Tested December 2007. Seas Excel W15LY-001 (Nextel) Cost: $151 Usability: Build Quality: Consistency: Value: Comments: This is the new Nextel Excel. It's basically a gray painted paper cone. Response is smooth and extended except for a wideband dip centered on 1kHz. This is not as bad as it seems, because in a normal width box, there is usually a diffraction peak there. on top of the 6dB 2pi to 4pi space conversion, there is often another couple dB of diffraction ripple starting with a peak at 1kHz and a dip at 2.3kHz or so. Bottom line, under normal usage this driver is very smooth, and usable with a very high or shallow slope crossover point. Harmonic distortion is generally pretty good with a mild increase in 2nd order below 1kHz, which may be audible but not too annoying. Tested May 2006. Morel HU531 Cost: $127 Usability: Build Quality: Consistency: Value: Comments: This Morel driver almost looks alien from the back with a very unique frame and magnet mounting system. It has a huge 3" voice coil. The response is relatively smooth and extended, and the voice coil should remain cool under heavy usage, but that's pretty much it for the items on the good list. On the bad side is the harmonic distortion. Basically, it's terrible. Notice that the F5 is nearly as high as the F3... not good. While I was running the sweep, I noticed very high tall order spectrum noise. I have to wonder if we will ever see 3" and 4" inset magnet speakers with good motor designs. Currenly, not from Dynaudio, Hi-Vi or Morel. Tested May 2006. Eton 5-880 Cost: $120 Usability: Build Quality: Consistency: Value: Comments: The Eton 5-880 has a multi-layer cone with a hollow hex core. The response curve is relatively smooth but will require a RLC notch to tame a breakup node. Harmonic distortion is average, and due to the 3rd order harmonics induced by the breakup, I recommend a 2kHz crossover or lower. In the build quality dept, the frame is nothing special. I would expect venting behind the spider at this price point, but it's solid with no airflow. The surround rubber is inset from the outer frame metal and looks like it doesn't fit quite right, but that really only affects the visuals. Overall, there's much better drivers out there for this kind of money. Tested May 2006. Davis 13KLV5MA Cost: $118 Usability: Build Quality: Consistency: Value: Comments: (discontinued) The Davis drivers I tested had some issues. The first issue was a bad case of cone sag, with the effective Xmax being about -1/-7mm. The 2nd issue was that one of my samples had a problem with a rubbing voice coil at some frequencies. The cone sag was caused by bad packaging from the manufacturer. Drivers with surrounds that protrude past the frame need some sort of spacing to keep the box from pressing the cone in. This had none, and the cone sag was permanent, not springing back after any kind of breakin period. 3rd order HD was kinda high between 1 and 2kHz, and I suspect a very high 2nd order HD below 300 Hz was related to the cone sag. Tested May 2006. Seas L16RN-SL Cost: $86 Usability: Build Quality: Consistency: Value: This is a driver from Seas with a unique flange designed to get the most radiating area into a 15cm basket. Typically 25% more Sd than most others in this group. Great harmonic distortion below 200hz, with good extension and good Xmax to back it up. Harsh harmonic distortion on the high end requires a very low crossover when used in a normal 2-way, at least 1400 Hz LR4. Minor issue in the response curve at 500 Hz. As used in a normal 2-way speaker, a few points are lost in usability and value. This driver is best used in special bass/midbass applications where it's small size and great bass performance can be put use. Tested October 2007. Seas U16RCYP Cost: $78 Usability: Build Quality: Consistency: Value: Comments: Like other new 16cm Seas woofers, this is designed to have a lot of radiating area that fits in a normal 15cm cutout. Works well and goes deep in a smallish vented enclosure of .25 to .38 cu ft and even smaller sealed. Very smooth and extended response, thanks to the woven poly cone. Merely average harmonic distortion through the midrange but a very clean low end. Smooth response except for a minor ripple at 1200Hz. Tested June 2009. Seas ER15RLY Cost: $69 Usability: Build Quality: Consistency: Value: Comments: This woofer is very close in performance to the older CA15 listed below. In fact, the differences are hardly even notable - a bit lower Fs, slightly higher sensitivity, and mild response differences. It's all too close to call. All of the Seas build quality and consistency is there, and the response curve is flat and extended. This is a woofer that's easy to implement and easy to enjoy. Tested June 2007. Peerless HDS Exclusive 830882 Cost: $66 Usability: Build Quality: Consistency: Value: Comments: (discontinued) This is the 5.5" flagship of the Peerless line. It's touted as having low distortion, and it lives up to that. Harmonic distortion was near Scan Speak levels with the exception of some issues between 400 and 500 Hz. The 450 Hz peak in HD was exactly the same in both samples, leading to an extra star for consistency but holding back one star for usability. If I had to speculate on the 450hz HD peak, it's probably related to the dustcap-cone configuration. The frame casting was a little on the rough side, and the phase plug was not centered as well as I have often seen in Seas speakers. Overall, it's still a great driver and a good value but honestly I would have hoped for a little more attention to details. Tested May 2006. Seas CA15RLY Cost: $66 Usability: Build Quality: Consistency: Value: Comments: Coated paper driver from the Seas standard line. It has an extended response curve and is generally well-behaved in linear and non-linear distortion. It's not the lowest distortion driver around, but well worth the money. It has the typically excellent Seas build quality. Tested May 2006. Aurum Cantus AC-130F1 Cost: $60 Usability: Build Quality: Consistency: Value: Comments: This mineral filled poly cone driver from Aurum Cantus has an extremely smooth response curve. While there is a breakup around 5kHz, it is smooth on both sides and easily controlled with a notch filter. Build quality is excellent. The voice coil has venting under the spider, something not quite common at this price point. There is a 3rd order peak in HD at 1.8kHz which will be audible on some content but probably not noticable most of the time. One negative for many DIY'ers is the squared off frame, near impossible to countersink with normal tools. Tested May 2006. SBacoustics SB15NRXC30-8 Cost: $53 Usability: Build Quality: Consistency: Value: Comments: This is a paper cone driver with excellent build quality, excellent motor design and very low harmonic distortion in the midrange, near Scan-Speak levels. Merely average HD in the bass and midbass. A mild dip in the response at 1700 Hz won't be too noticable, but the breakup starting at 5kHz will be a challenge to filter out. Aside frome the minor response curve issues, this is a great driver and one of the value leaders in this price range. Tested December 2008. Peerless HDS Nomex 830873 Cost: $47 Usability: Build Quality: Consistency: Value: Comments: This driver is one of the cheaper models in the Peerless HDS line. It is very similar to the Exclusive, but without the phase plug, without as much faraday sheilding in the motor and it has a trunctated frame. It does not have a distortion issue at 450 Hz like the exlcusive, but distortion is somewhat higher throughout the spectrum. The response curve is very smooth, and the top end is a little easier to control than the Exclusive. Overall, it's a decent performer and value. Tested May 2006. Tangband W5-1685 Cost: $46 Usability: Build Quality: Consistency: Value: Comments: This is a paper cone driver with a monsterous underhung motor design and very low harmonic distortion in the midrange and low treble. Merely average HD in the bass and midbass. The response curve is smooth and controllable with a mild breakup at 8kHz that may need to be addressed depending on your crossover point. The driver can generally cross over as high you you need it to. This is a value and performance leader in this price range. Tested December 2008. Dayton RS150S-8 Cost: $37 Usability: Build Quality: Consistency: Value: Comments: The Dayton RS line has been a long time price/performance leader from Parts Express. The harmonic distortion level is not seen in any other driver at this price point. The response curve is a little bit ragged, and the breakup node has 2 peaks, but for the most part it is controllable with a few extra crossover components. The isn't a Scan Speak or Seas Excel level performance, but at this price anyone can be happy with this choice of driver. Bonus points for looking like a stealth fighter. Admit it, looks are important. Tested May 2006. Dayton RS150-4 Cost: $33 Usability: Build Quality: Consistency: Value: Comments: This is the 4 ohm unshielded version of the Dayton RS150 above. Functionally, it is exactly the same, with the difference in sensitivity expected from the lower impedance voice coil. The rubber boot on the magnet is a nice touch. Tested May 2006. Hivi M5a Cost: $33 Usability: Build Quality: Consistency: Value: Comments: A well made cast frame driver with good to average performance. On the positive side, decent sensitivity and smooth midrange response. The down side is harmonic distortion performance that is consistent with a very simple motor design. Overall a reasonable performer, but there are a couple other woofers in the under $30 class that perform better in various areas for less money. Tested June 2009. Tangband W5-704D Cost: $25 Usability: Build Quality: Consistency: Value: Comments: This is one of the value leaders in this group, and an excellent driver for the price. It has a cast frame, poly cone, rear vented pole piece, vented spider, double roll rubber surround and a catenary shaped dustcap. The response curve is respectably smooth and extended. One of my 3 samples had a voice coil lead that disconnected at full Xmax, thus a strike against consistency. The harmonic distortion throughout the spectrum is average for this price level, but it does not have any breakup related peaking like some stiffer coned woofers do. As a result of that and the smooth response, you can cross over the woofer just about as high as you want to, eliminating the need for an expensive tweeter. Old/Other tests available. Tested January 2008. Peerless 830656 Cost: $21 Usability: Build Quality: Consistency: Value: Comments: Budget level woofer with stamped frame. The frame is designed for surface mounting, and is well ventilated behind the spider. The pole piece is not vented, but the cone is perforated near the voice coil former for very turbulence free operation. It reaches it's full Xmax without air noise. Very smooth response curve, and also a smooth harmonic distortion profile, though not particularly low in the midrange. Cross it over as high as you want to a cheap tweeter. Tested November 2008. Silver Flute W14RC25-8 Cost: $21 Usability: Build Quality: Consistency: Value: Comments: A long time value leader, we don't even expect to see cast frames at this price point. The cone is made of pressed wool. The motor is vented through the pole piece. There is a little wiggle in the response curve at 1700 hz, but the more pressing issue is the breakup induced 3rd order harmonic centered at 2.3kHz. it's a wide bandwidth peak due to the shape of the breakup. The best crossover point for this driver is probably around 1800Hz. I've used it higher than that, but hindsight is always 20/20. Originally $14, the price has went up and value went down. Tested May 2006. Dayton DA135 Cost: $19 Usability: Build Quality: Consistency: Value: Comments: At this price, the DA135 is well worth the money. It has a stamped steel frame, aluminum cone, rubber surround, and no voice coil venting at all. It might be a little overly critical to point out flaws in a woofer at this price, but the top end of it's response is a little ragged and the bottom end is a little muddy in the harmonic distortion. Not quite up to the value and performance of it's larger 7" brother. Tested January 2007. MCM 55-3870 Cost: $18 Usability: Build Quality: Consistency: Value: Comments: This is also one of the value leaders in this group, and an excellent driver for the price. The cast frame is one of the best in the group, almost as strong and well ventilated as the Seas. Very smooth and clean through the midrange, but not quite the same bass clarity as a couple of the other drivers in this price range. There is a breakup node, but it's very high and easy to deal with. Will work well up to 2.5 kHz with an LR4 filter. Tested June 2008. Ology SL1364 Cost: $15 Usability: Build Quality: Consistency: Value: Comments: A low cost driver with a stamped non-round frame. It was tested surface mounted, likely the most common usage for a flange of this difficult shape. Highish Qts will likely mean a sealed box is optimal. An LR4 crossover at 1800 Hz seems natural given the driver's response shape and non-linear distortion. This would tame some of the driver's 3rd HD peaking, and the dip in response around 2kHz with the breakup node above will work with a metal cone style filter - single inductor with an LCR series notch across the terminals. Tested September 2007. North Creek 13W06S Cost: $14 Usability: Build Quality: Consistency: Value: Comments: (discontinued) This is a cast frame, shielded driver with a coated paper cone. The voice coil is vented through holes behind the dustcap. At this price point, you'd probably be happy if the driver worked at all. Good news, it does work well, albeit with a few limitations. Qts is on the high side, limiting usage to a larger sealed box, or possibly aperiodic damping. A mild cliff in the response curve at 2kHz means a lower crossover point is a good idea. This driver is part of the North Creek Echo kit, which pairs it with another good value driver, the North D25 tweeter. Tested May 2006. MCM 55-1290 Cost: $9 Usability: Build Quality: Consistency: Value: At $9 each, you'd probably be happy if this driver worked at all. The surprise is that it does work well, although with quite a few limitations. The limitations are as follows: very low Xmax - if run full range, this will begin distorting quickly. High Qts - also if run full range, there will some midbass boominess. The response curve has a dip at 4kHz, but with some smart filter work it should be acceptable. A crossover point of 2.5kHz with a LR4 slope is possible. Looking at these limitations, it's obvious - this driver would make a decent high value midrange driver. The response curve and harmonic distortion is excellent if used in it's optimum range. At the price, it scores high in value. Tested June 2007.

Page done by John "Zaph" Krutke © 2007
Also visit -Zaph|Audio-