Acoustic Research ARXP62 Bookshelf Assessment

First, let me note that I consider myself a casual audio listener, not a rarified audiophile. What follows is only my opinion, but it is more than you will find elsewhere concerning this product's performance. It is my sincere hope that this write-up will be helpful to those whom are unable to audition this speaker before deciding on a purchase.

As of December '06, J&R seems to be the sole purchasing option. If you live near a brick and mortar store, I encourage you to go listen for yourself and add your thoughts to this thread. If you own or have otherwise heard these yourself, please do post an opinion, whatever it may be.

-----

Attributes of the XP62:

• 1" extended frequency tweeter - soft dome; unidentified material
• 6.5" aluminum woofer - with true phase plug
• Frequency response: 38Hz-50kHz +/-3dB *
• Nominal impedance: 8 ohms
• Sensitivity: 88dB ^
• Recommended power: 15-125 Watts
• Finish: Black Ash Vinyl with High-Gloss Black Accents
• Dimensions: 13-3/4"h x 7-3/4"w x 12-5/8"d
• Individual speaker weight: ~13.5 lbs
• Video shielded
• Made in China
• Not bi-wireable
• XP series manual: http://www.audiovox.com/manuals/owners/XP_Man_6_29.pdf
• Currently available for $100/pair.

* see post 2; measured bass output does not support this claim at -3dB; I have not run across an official response chart; J&R web spec. doesn't match manual
^ SPL output vs other 88dB speakers suggests actual sensitivity of ~86dB

Speakers used for comparison: B&W DM601 S2

• Are bi-wireable (the DA10N can bi-amp these but this was not done for this comparison)
• Spec Sheet: http://www.bwspeakers.com/downloadFi...fo_DM601S2.pdf
• Dealer price; years ago: $450-$500/pair.

Test system configuration:

• Source: Sony CDP-CX355
• Receiver: Technics SA-DA10N (optical in)
• Wire: four whole strand, braided Cat5 (~12 gauge; same as Audioholics Cat5 V2 cable)

-----

I spent several hours listening and A/B selector switching between the DM601s and XP62s. In that time I compared five selections; specifically, these:

1. Sarah McLachlan / Afterglow - Train Wreck
2. Seal / Seal 94 - Don't Cry
3. Carpenters / The Singles '69-73 - Yesterday Once More
4. Pink Floyd / Dark Side of the Moon - Money
5. Norah Jones / Come Away with Me - Don't Know Why

My listening position was centered about 7.5' away, with the speakers on the same makeshift stands, side-by-side with the channels separated by about 6.5'. Most of the time the seating position level measured 80-90dB, peaking at 95dB. Both sets were auditioned without grilles.

One confounding issue is that the DM601s were louder at the same volume setting. It's odd since the B&W specs state 88dB, matching that of the XP62s. However, they clearly had a 2-3dB greater loudness as measured using a Radio Shack handheld digital meter on program material. Maybe the receiver B amps were hot compared to the As. [subsequent selective measurements did not reveal an amp offset] Whatever the cause, louder tends to sound better and in some of the testing I did try to adjust for this effect but could not be 100% accurate since the Technics does not read out volume digitally. It has only a motorized knob for level indication.

-----

Performance observations:

I'm not comfortable calling something warm, transparent, bright, smooth, 3D, etc. It seems to me that such terms are too easily misconstrued, so I will avoid using them.

Bass: XP62 wins this one, with more of a tactile quality, but not by a lot. Do I need a sub? Not really, but perhaps at lower volumes.

Mids: Surprise; I preferred the XP62's performance here as well! This difference is much of what gives the sets different tonality. I sense that the XP62s have less bass emphasis in the mid which results in less nasal sounding voices. Lyrics are nice and clear. I noted the slightly more nasal sound of the DM601s when comparing Norah Jones' and Sara McLachlan's voices. I'm not saying the DM601s aren't pleasing in this respect; I like them still. Could someone send Sara over so I can compare the genuine article to these?

Highs: There is another small difference here. The XP62s are only slightly less intense. Maybe it is due to the greater degree of mid/bass emphasis. I should note that I'm 37, so according to sources, I no longer hear above ~16kHz, which begs the question of what the rated 50kHz response really could contribute. Micro-dynamic harmony? I suppose. If there is a winner here, it's not clear enough to me. [see also post 28]

If I were not privy to which set were selected but had heard the selection on both recently, these speakers are different enough that I'm reasonably sure I could identify which is in use. If an unfamiliar artist/selection were played on only one randomized set, I'm not so confident. I would try to pin it down using the difference in bass response in that case.

In conclusion, my B&W dealer is fortunate I didn't have the XP62s in an A/B setup at the showroom when I bought the DM601s. I would have confidently walked out that day with the XP62s! Factor in the price difference and it is no longer a serious contest.

-----

Additional details of the XP62 per direct inspection:

• Has 5-way binding posts with plastic nuts; even the DM601s use plastic nuts.
• There are odd plastic top/bottom front retainers for the wire mesh grill. They remove easily, but do need to stay since the speaker directly sets on the bottom one and the top looks funny without it. The top cap is loose enough without the grill to make me wonder if it will eventually rattle, causing its removal.
• Behind the grill is a surface covering shaped foam piece that is not intended for removal, but may have acoustic benefits and looks nice anyway.
• Utilizes a single rear port, slightly above center, attached to a 1.5" diameter tube of about 7" length and flared on both ends.
• Cabinet and finish seem solid with maybe a top/bottom that should be thicker. I passed that judgment after simply rapping a knuckle on each surface.
• The rear wall is not a relatively solid plank like most speakers. Instead there is at least some side-to-side cross bracing and a large plastic housing. This housing is a 1.5" deep (at center) convex, semi circular, protrusion attached to the entire rear of the cabinet (axis running top to bottom). The shape echoes that of the cabinet's front grill.
• Using a telescoping mirror to look around in the cabinet, I spotted yellow acoustic batting on one side and bottom with the top and opposite side being bare. The same batting is visible at the rear of the cabinet behind what appears to be the main rear brace/wall placing it in the aforementioned plastic housing.
• Mid/bass surround material appears to be primarly composed of foam but may be a rubberized hybrid.

-----

XP62 Photos:



Thanks to 'Groch' for the following photos from posts 21 & 22:

Given that the manual specs for the ARXP62 bass extension are rather difficult to believe, I broke down and conducted a crude set of sine wave based frequency measurements.

I used the Radio Shack SPL meter to record the C weighted, slow response level of bass frequencies ranging from 31.5-104Hz. This was done with one speaker of each set at a distance of 2 feet from the test speaker.

While I was at it, I checked for a difference between the A and B amp set of the receiver. I found no difference when switching between them using the same speaker so the previously reported loudness difference of these two speaker sets does not seem to be due to a constant attenuation offset within the receiver's amps.

Measurements have not been corrected for the relatively well known Radio Shack meter offsets. They are also subject to unknown room effects so I don't expect others to be able to faithfully reproduce these specific results. I wanted only to get a less subjective feel for the lower response limits and bass signature deltas of these speakers. Measurements were spot checked by way of sub-sampled repetition after the full sweep on each speaker. They did prove to be repeatable.

I am surprised to find that the two speakers measured so similarly. Absolute standard deviation between readings is .95dB with only one instance of a >3dB delta at 54Hz among 74 samples. Despite my own expectations, the DM601 did not roll off any higher than the XP62. I did verify that the receiver's sub-woofer cut-off filter was disabled.




1/1/07 Update - Happy New Year!:

I spent a bunch of time today figuring out how to properly use HomeTheaterShack's free Room EQ Wizard. From that I obtained complete sweep graphs and posted them below. In the process I discovered that my PC's onboard C-Media sound device response drops precipitously above 7kHz, but it was measurable and the 'Wizard' applied correction factors for the sound card as well as the Radio Shack SPL meter.

Still, these metrics are really only useful for relative comparison of the two speakers in my write-up. Room effects remain undefined, as before. It's nice to see my original bass results are confirmed by these charts.

Apparently, the 4kHz DM601 crossover point is reflected in the graph. Judging from that, it seems the XP62 is using a similar frequency. In both cases, I would have hoped for a smoother transition.

20Hz-24kHz sweeps with 1/3 octave smoothing:



20-250Hz, 0-250ms waterfalls:

Photos have now been posted.

openwheelracing:
I wonder if the sides of the VP26 and those of the XP62 are the same thickness. The XP62 sounds thinner on the top/bottom, but it could be due to cross bracing for the sides. Did the VP26 also have this characteristic? Knuckle rapping on the DM601s sounds remarkably similar to that of the XP62s.

dknightd:
I'm an audio/electronics pack rat. I would sell a lot of my other junk long before punting one of these sets. I'll keep both sets in service and may demote a much older set of ARs in my bedroom. At least for the short term, I will be listening more to the XP62s. Who knows?; I might find material they hate...then revise my assessment, though I hope not.

Despite having owned the DM601s for years, I had to educate myself a bit about phase plugs.

The XP62 mid/bass driver also seems to be different than that of the VP26 (and presumably the entire VP series). The XP62s are definitely using a true phase plug. The close up mid/bass photo shows the gap between the cone and plug.

In fact even the DM601s are using a quasi plug attached to the Kevlar cone. I was expecting to find that the XP62 design was the lesser of the two, but that is not necessarily so per this quote from Audioholics (in the speaker gimmicks article):

"Be cautious with some loudspeaker manufacturers that use a quasi phase plug which looks like a phase plug for marketing purposes, but does not physically separate from the cone of the driver. These 'so called' phase plugs can sometimes do more harm than good as they can provide an inconsistent surface area and can also increase mass of the cone yielding poorer frequency response and linearity.

A real phase plug essentially eliminates the need for a dust cap, which in turn, reduces cone mass, permitting extended frequency response. It allows more airflow through the motor structure, which improves cooling and thus increases power handling. A Phase plug also helps to reduce on-axis beaming by dispersing higher frequencies that the cone is producing."

I have updated the second post in this thread to include some full range sweep comparison measurements.

Enjoy!

I ordered a second pair of these now; they are that good for the price. I'll need to figure out how to run them together...maybe A+B with one set flipped on top and foam between...hmm.

You win the award for first independent opinion to post!

Of course you'll probably know that lots of people also describe bright as detailed, even crisp/clear. As a reference, I own Axiom M80s and the audio community albatross of that brand is that it is 'bright', an opinion too which I never subscribed.

Anyway, I count detail as a positive when it is not overweighted in comparison to the remaining frequencies. Do you consider the bass or mids to be recessed at the same time? If not, I would be reluctant to tag these with the 'bright' term...just my two cents.

I hope you do get them adjusted to your liking, but I would advocate listening to them without treble adjustment for a while to see if they break you in (which is actually what some say 99% of break-in really is...haha).


Nope, I'm using them in a stereo only setup. The driver difference in the center would also concern me a bit. Then again that size isn't uncommon in centers. It would be nice to know if the mid/bass drivers use true phase plugs though. I doubt they do but would you mind checking for that when you go to listen?

-----

I received the second pair today. Running them A+B provides a 3dB gain at the same volume setting...a nice little boost plus the flipped stacked orientation creates a makeshift D'Appolito driver configuration which may provide associated advantages.

Thanks, and yes, the plug should not be attached at all.

Edit - Looking at the J&R image, the center seems to have the same golden outline which matches the XP62's cone, making me think it is also a true plug design. Surprising if so. How many budget speakers offer this feature nowadays?

Groch:

Thanks for the interior pictures! They are very nice/informative. Interesting that the basket is plastic. At least it appears to be a rigid design.

I had measured the mid/bass driver as you did but when I then measured the DM601, M80ti, and older SVS 12" drivers, finding them to consistently leave off 1" from the spec, I figured it was an industry standard. If so, driver measurements often include the basket mounting flange, accounting for the observed deviation.

I'm also wondering too what degree a break in period alters frequency response. My second pair arrived four days ago and was immediately put in service while flipped and set atop the first pair. Soon after, I heard excessive peaks when listening to Keri Noble's 'Fearless' CD. Switching to just the first pair, things improved. Moving the new pair back to a normal orientation and playing them solo, the peaks again seemed to reduce somewhat, presumably due to external acoustics likely coupled with an axis response variation. It's not surprising that positioning would be influential.

I'm now running the second set in to see what happens as time passes. I wrote the OP assessment based on a pair that was run in for about thirty hours so after about a week I'll know more and plan to post an update.

-----

masteraleph:

This comfirms that all the XP mid/bass drivers share a true phase plug design. Did you listen to them and/or buy anything for an in home demo? Adding comments of any type regarding the XP62 herein is appreciated!

Power handling is good by my estimate. Using highly compressed popular music, I've turned them up a bit past 100dB at my listing position for short periods. I don't want to push them too much more than that. Once they reach loud levels, I start looking for the invisible sub. Based only on the specs, it should be possible to reach peaks of 109dB at 1 meter anechoic. However, I do believe the sensitivity spec is higher than actual, as has been noted twice in this thread. I suspect it is closer to 86dB vs the claimed 88dB. That said, they should get plenty loud in all but large rooms which challenge all similarly sized speakers. Since I have not driven them to their physical limits, I have not noted any point at which there is audible distortion, much less a stressed tweeter. I'm not exactly sure what a stressed tweeter would sound like, short of a strong but short spike followed by the complete loss of high frequency reproduction...


The vinyl finish is consistent and of good quality. Under scrutiny, seams meet up with just good to average alignment. They look like most faux wood black boxes I've seen before without appearing to be cheap.

Ok, I've run in the second set long enough to conclude that the initial brightness does reduce somewhat. What could be ear flinching peaks, lost the flinch response. Both of my XP62 pairs now have the same characteristics given the same setup. However, I have not measured the second set with my meter and have no plans to do so at this time. In the process I ran the new pair on a second amplifier (Sony TA-F707ES) to see if this had an obvious impact compared to the Technics. It did not seem to change things in any way I could readily identify.

Selections tested in the initial post do not seem to push the highs as significantly as some recordings. A good example of more peaky material is the previously mentioned Keri Nobel 'Fearless' CD. Playing that CD on the B&W's also creates a heightened sense of high frequency response. So my conclusion is again that compared to the DM601 S2s, the XP62s are remarkably similar..yes, even bright, in the way the DM601s are bright. Call it detail if you don't mind the DM601 sound. I would say that compared to a set of Axiom M80s, the DM601s and XP62s are brighter. Of course my M80s are in another, larger room with some acoustic damping, so take that with a grain of salt.

The major amendment to my initial assessment is that this perception of brightness is the primary weakness of the XP62 so far as mass appeal is concerned. Even I will be tempted to use the treble adjustment on select material. For the time being, I choose to view this as the speaker's unforgiving nature versus a truly unbalanced tweeter response. I expect only a professional response analysis could effectively settle the issue.

These would need a creative solution for any kind of wall mounting. The rear port would cause some extra stand-off needs. Only the top, bottom and sides seem like viable attachment/support points. It should be possible to drill into those, if need be. I wouldn't trust the plastic housing on the back to provide adequate support on its own. These really are most at home on stands. It's a shame AR didn't keep the nice VP bookshelf mounting features...that was a definite step backward.