Drivers List › Dayton Audio Drivers › UM18-22

Dayton Audio kindly sent in a review sample of their new big boy subwoofer the UM18-22. The other long throw 18” driver in the Dayton lineup, the RS18-HO,  was reviewed about 2 years ago and found to offer a lot of performance for a very reasonable cost. The largest Ultimax driver proves quite capable as well and offers a larger amount of excursion for deep bass headroom, and a bit more thermal power handling, at only a modest cost increase over the RS18-HO.

The Dayton Audio UM18-22 is a nominally 18” driver built on the familiar 12 spoke cast aluminum frame. The driver features a triple stacked set of ceramic magnets, about 7.5” in diameter and roughly 25mm thick. The magnets are covered with a rubber, protective magnet boot and the top and back-plates of the motor appear to be around 10mm thick or so.  The back-plate and T-pole features a large center pole vent of about 1.5” in diameter for enhanced cooling and the frame also has large vents underneath of the spiders to help cool the former and coil. Additionally there are also a pair of aluminum shorting rings in the motor in order to reduce distortion due to eddy currents and help with heat sinking. The voice coil is a 75mm or 3” diameter unit which is 4 layers in total and is wound with copper wire as a pair of 2ohm coils. That being the case the driver can be wired as either a 1ohm load or 4ohm mono load. A dual 2ohm load is also a possibility of course. The former that the coil is wound on is a black anodized aluminum piece. The cone for this driver is listed as a single piece of Nomex honeycomb which is covered by a concave piece of woven glass fiber. This dish shaped cone in turn appears to also be connected to the former and reinforced at the spiders with a very steeply raked cone of pressed paper. The result is a very nice looking cone cosmetically and one which is also quite strong and resistant to damage or marking. In fact there is a video of this driver being stabbed with various implements and otherwise tortured in a manner that would destroy typical cones and emerging relatively unscathed. The suspension system supporting all of this is composed of a pair of linear spiders that are around 7.5 to 8” in diameter and mounted in a mirrored configuration to further reduce suspension distortion effects and a large high roll profile rubber surround. The terminals are two sets of spring loaded units that accept very large gauge wire. The leads are sewn into the top spider and appear to offer plenty of slack to allow large cone displacements. Considering the street price of $280 shipped, this is a nicely appointed and well put together driver.

The driver function and mechanical noise was checked with a 20Hz sine wave in free air while breaking the suspension in a bit. The UM18 is rated at an xmax of 22mm one way which I believe is a coil/gap geometry mathematical figure with no correction applied. The driver operated smoothly with very little air or suspension noise up to around 1.5 to 2” peak to peak which is consistent with the xmax spec. In fact the driver was impressively quiet when driven to this amount of excursion with only a bit of air noise from the gap and pole vent. Pushing the driver a bit further to what was roughly 2.5” peak to peak caused a noticeable increase in distortion and noise. This seemed to coincide with the driver being pushed past its linear range of movement. I did not push the driver too much farther than this but it seems that the mechanical limitations on the incursion side of the stroke are at least 35mm or greater.

After this break in and function check, the driver was mounted to the test stand and a series of impedance measurements were taken in order to calculate the small signal parameters for the driver. We use the measured physical cone area plus 1/3rd of the surround for the SD calculation here at Data-Bass so as usual the SD came in somewhat lower than the manufacturer spec. The Dayton spec for SD is 1219cm, which is no doubt including half of the surround area as most manufacturers now do. The measured SD using 1/3rd of the surround area came in at about 1115cm. The measured Re of the driver with the coils wired in series was also a bit lower than the factory spec at 4.1ohms versus the factory 4.36ohm rating. The measured Fs of this particular unit was 21.2Hz which is very close to the factory spec of 19.5Hz and being a fresh unused driver it will likely drop down a bit after it gets a number of hours of play time. The BL measured right at 22Tm which is just a bit lower than the factory rating of 23.76, put pretty close after considering the slightly lower Re of the driver as well. The factory specs give a BL^2/Re of 129.5 versus about 120 for this particular driver. The impedance curves also aligned quite nicely with both showing a similar shape and Fs and both showing the same resonance issue right about 500Hz. The suspension compliance measured about 0.128mm/N which is a little softer than the factory spec but not by too much. The suspension compliance is a source of significant variation from one driver to the next. The Vas was quite close at 223L measured versus 212L factory and the inductance came in a bit lower than the factory rating which is always nice on a driver with this much stroke. The factory 1Khz rating is 2.22mH while the measured 1kHz rating is only 1.48mH. Judging by this and the shape of the impedance curve above the driver resonance, the dual shorting rings in the motor are doing their job admirably. One significant difference in the measured versus factory specs is the mass of the moving assembly. The factory spec is a very heavy 655g while the unit measured here calculated out to just 440g which would be pretty average for a driver of this size and build. I do not know whether the Dayton spec is an actual weight measured from a cut up assembly of the components or not. If it were it would obviously be the more accurate spec. However 655g seems very heavy for this driver. There are drivers with much larger heavier voice coils, longer formers, larger heavier spiders and surrounds, that do not weigh near 655g so if that is the true weight of the moving assembly the majority of the extra weight must be in the cone assembly. The coils are copper wire which is much heavier than aluminum though so that could account for some of the extra weight. Either way the difference in Mms is not that big of an issue it mainly affects the suspension compliance calculation, the Vas and the overall sensitivity. Since the Mms calculates to only 440g the result is a somewhat lower Qes and Qts than factory and a bit higher voltage sensitivity. The overall picture is of a high value driver which should have a well behaved response shape, low noise, predictable behavior and low distortion when kept within its linear range of operation.