Houston, one of my Revel Salon's may have a problem

[Irvrobinson]

I agree a re engineering is a major task, and not undertaken lightly.

I stand by the fact that the crossover points are almost certainly a major factor in these failures.

Here is the power response versus frequency.

So look at the added power demands on the mid range driver with a crossover at 150 Hz versus 350 or 400 Hz.

In fact this is worse than it looks, as in the design in question, the mid range driver is going to be taking the lions share of BSC so that the power to that mid will be going up first order from around 550 HZ. So the power demands on that driver are huge.

In essence if you think about it, this speaker is really a bookshelf speaker with small drivers, on top of a sub with 8" drivers.

Actually, I think it is a tribute to the build of that driver that it sustains that punishment as long as it does.

You are entitled to your opinion as I am mine. I maintain and believe strongly that the design considerations for this speaker were not prudent with regards to long term reliability. They are choices I would never have made.

I agree that these speakers, and the Salon2s (and several other speakers I'm aware of) are really more like bookshelf speakers with an integrated sub. But, like them or not, sub-200Hz crossover frequencies are common. As an avid solo piano listener I suspect I may prefer these ill-conceived ))designs, because sticking the knees of crossovers well inside an octave either side of middle C sounds like a bad idea for piano reproduction. My previous speakers, the Legacy Focus, had a crossover point of 180Hz. Coincidence or a sign of a preference? Previous to the Legacys there were a couple pairs of ADS speakers, like the L1530, which did have a 450Hz crossover frequency. And one of my complaints with the L1530 was that it didn't sound as natural as I thought it should on my own solo piano recordings. I have to admit, though, I didn't even realize the Salon2 had a 150Hz crossover until after I bought it. It just never occurred to me to care.

 

[AcuDefTechGuy]

Yeah, I prefer to have FOUR 6.5" midrange drivers per speaker myself.

 

[RichB]

Yeah, I prefer to have FOUR 6.5" midrange drivers per speaker myself.

You're at 4 times the risk of glue failure

- Rich

 

[GranteedEV]

This does make me wonder about what'll happen someday to my Salon2s when they go out of production. Every driver is in-house, full-custom. In ten years, if I keep them, I may have a pair of 175lb paper weights on my hands.

Those low diffraction cabinets alone would be tempting to toss some high-end off-the-shelf drivers in, if they fit the mounting holes, and make all-new crossovers. :drool:

 

[AcuDefTechGuy]

You're at 4 times the risk of glue failure

- Rich

No pain, no gain.

Imagine the risk on those line array speakers with a dozen or more drivers. But I'm sure the reward is worth the risk.

 

[Irvrobinson]

Those low diffraction cabinets alone would be tempting to toss some high-end off-the-shelf drivers in, if they fit the mounting holes, and make all-new crossovers. :drool:

I'll send you a PM when they become paperweights. At least I know what to do with them now.

 

[TLS Guy]

Those low diffraction cabinets alone would be tempting to toss some high-end off-the-shelf drivers in, if they fit the mounting holes, and make all-new crossovers. :drool:

That's exactly what I thought!

 

[TLS Guy]

You're at 4 times the risk of glue failure

- Rich

I don't think so. There is nothing like having multiple voice coils for reducing the risk of failure.

 

[MidnightSensi2]

I agree a re engineering is a major task, and not undertaken lightly.

I stand by the fact that the crossover points are almost certainly a major factor in these failures.

Here is the power response versus frequency.

So look at the added power demands on the mid range driver with a crossover at 150 Hz versus 350 or 400 Hz.

...*snip*...

I maintain and believe strongly that the design considerations for this speaker were not prudent with regards to long term reliability. They are choices I would never have made.

As an avid solo piano listener I suspect I may prefer these ill-conceived ))designs, because sticking the knees of crossovers well inside an octave either side of middle C sounds like a bad idea for piano reproduction.

Is that the trade off? Reliability versus sound quality versus power handling?


..or are we just totally over thinking this?

 

[GranteedEV]

Is that the trade off? Reliability versus sound quality versus power handling?


..or are we just totally over thinking this?

It's a combination of a lot of things.

I think the uppermost 8" woofer on the Salon 2s wants to be crossed over higher, as pethaps does the 6" midwoofer. The problem is that it's very complicated to still integrate the other two 8" woofers. A hybrid-active solution with a separate amp for the other two bringing them in gradually as frequency drops would allow it, and you could have your cake and eat it too... It would also allow for higher midbass efficiency. And it lets you avoid any passive attenuation on, say, the Be tweeter circuit. Do a passive crossover and get the inter-driver phase matching right, but use tailored active EQ to drop the HF level (and the subsequent trims).

Market dictates though.

 

[Irvrobinson]

Is that the trade off? Reliability versus sound quality versus power handling?


..or are we just totally over thinking this?

I think TLS Guy is either overthinking it or underthinking it, depending on the point he's making. Other times he's spot on.

 

[TLS Guy]

Is that the trade off? Reliability versus sound quality versus power handling?


..or are we just totally over thinking this?

Not really. You can't take a small mid range driver and cross it over in the lower power band, and give it BSC duty and not expect failure, unless you are content with a speaker of limited power with power limiters installed. You can not make a high powered monitor that way.

That is why high powered studio monitors are not configured that way.

I think the limiting factor of designs like the ATC monitor, is the large woofer and therefore large front baffle. Baffle reflections are the limiting issue I suspect. Even so they are very fine speakers and used world wide in fine studios.

In that market you can not be known for unreliability. If you get that reputation you are soon out of business. Those speakers must perform day after day, month after month, year after year for many years without a failure, belting it out in mixing and production rooms.

And quite honestly, I can't abide unreliability in equipment of mine or others. A vital piece of equipment down is a bad day. My be its just me, but I like those days very far and few between. In speakers it should pretty much never happen, if designed properly.

 

[GranteedEV]

And quite honestly, I can't abide unreliability in equipment of mine or others. A vital piece of equipment down is a bad day. My be its just me, but I like those days very far and few between. In speakers it should pretty much never happen, if designed properly.

I know what you mean.

Then again, as far as power handling design goes, even with the low crossover point and BSC, the Revel speakers use true 4th order slopes (which have better power handling than anything lower or higher) an octave higher than most people are crossing their subwoofers to midwoofers the same size.

If the issue is glue, then it's a manufacturing issue.

 

[Irvrobinson]

I think the limiting factor of designs like the ATC monitor, is the large woofer and therefore large front baffle. Baffle reflections are the limiting issue I suspect. Even so they are very fine speakers and used world wide in fine studios.

I'm not a speaker designer like you, but I think this is a good guess, since my major gripe with the 150s was imaging. It absolutely flunked my jet flying over your head test. Monkish knows the track I'm referring to. I also wonder if that 15" woofer's dispersion doesn't match too well with the dome midrange.

 

[TLS Guy]

It's a combination of a lot of things.

I think the uppermost 8" woofer on the Salon 2s wants to be crossed over higher, as pethaps does the 6" midwoofer. The problem is that it's very complicated to still integrate the other two 8" woofers. A hybrid-active solution with a separate amp for the other two bringing them in gradually as frequency drops would allow it, and you could have your cake and eat it too... It would also allow for higher midbass efficiency. And it lets you avoid any passive attenuation on, say, the Be tweeter circuit. Do a passive crossover and get the inter-driver phase matching right, but use tailored active EQ to drop the HF level (and the subsequent trims).

Market dictates though.

You are spot on!

It really does take at least a hybrid solution.

This is how my speakers are configured essentially. Both 10" drivers share duty below 60 Hz, for cone excursion reasons and sensitivity. Both have separate amps. Below 60 Hz the signal is mono. So the below 60 Hz is the same to all four 10" drivers. This reduces excitement of unwanted room modes.

Both upper 10" drivers are fed the BSC signal for the 7" mid range drivers in MTM. The crossover also smooths the response of the 10" drivers, and cuts them off before break up. This really unloads the 7" drivers.

A big advantage of active BSC is that it can be set for room and speaker position. One of the big limitations of any speaker is that BSC will only be correct by luck. The BSC will be different for any given speaker in different rooms, and even more to the point in the same room in different positions. This problem can only be overcome with an active solution. This really is the key element in how I voice a speaker to the room.

The bass is not cut off from the 7" drivers. They are robust enough to handle it, as they are tough mid/woofers with a superior motor structure. Since they are in their own TL, they do not off load from the enclosure, like a reflex ported enclosure would. The high pressure in the line really limits cone excursion. The increased VC excursion that a crossover would introduce increases air movement in the gap and aids VC cooling. The smaller line is tuned close to half an octave above the long line. This provides a nice smooth assist to the drivers.

The odd shape of the enclosure is to avoid the 7" drivers and tweeter being on a large baffle.

The downside is that you do get a crossover in the speech discrimination band. In this case it is 2.8 KHz. This is 700 Hz lower than I consider optimal.

Although these 7" drivers are bass/mids, they do make good mid range drivers and are used solely as mid range drivers in a number of designers including by Salk.

One advantage of this design is excellent phase response and a spectacularly good impulse response.

It should be no surprise that two and 2.5 way speakers are so popular. Three ways and higher just present a whole host of additional challenges. Once you get to three way designs and above the case for at least a hybrid design, if not fully active. becomes close to overwhelming, especially if you are looking for really spectacular performance.

 

[TLS Guy]

I'm not a speaker designer like you, but I think this is a good guess, since my major gripe with the 150s was imaging. It absolutely flunked my jet flying over your head test. Monkish knows the track I'm referring to. I also wonder if that 15" woofer's dispersion doesn't match too well with the dome midrange.

I think you are correct. That is why I favor two 10" drivers over a 15" The cone area is the same. As long as the 10" has a low Fs, in my case 20 Hz then all will be well.

 

[TLS Guy]

I know what you mean.

Then again, as far as power handling design goes, even with the low crossover point and BSC, the Revel speakers use true 4th order slopes (which have better power handling than anything lower or higher) an octave higher than most people are crossing their subwoofers to midwoofers the same size.

If the issue is glue, then it's a manufacturing issue.

I can tell you from my driver designing days, that nothing causes glue to fail like heat! Even with the fourth order crossover, I think you are failing to appreciate the massive power burden of BSC with a crossover that low.

 

[GranteedEV]

I can tell you from my driver designing days, that nothing causes glue to fail like heat!

I don't think we disagree. It's just that especially with chinese manufacturing, the glue itself might not be up to spec. Kevin Haskins, who designed my subwoofer driver, went out of business because of one bad asian batch. My sub is still working great with with thousands of watts (not really ever) going into it, but the drivers from that batch all had the wrong glue on the shorting rings, causing them to become dislodged. It's not a cheap driver... I think I paid about $450 for it and it weighs around 60lb.

This guy had eight of the same driver, except from the bad batch with the "chinese swapped-out glue", and all of them failed almost instantly when power was applied:

Granted, good batches with good QC also exist, and I'm positive my driver is from a good batch because it's still pulling along. But it lends credence to the idea of only buying North American manufactured stuff and more importantly, I hope the driver companies move manufacturing away from China where ethics and communication are at a premium. They're playing with fire, and even if the customer doesn't, they WILL get burned.

My next sub driver is going to be the Mach V Phallus, which is made right here in Canada. The build linked in my sig uses a high quality American custom manufactured Acoustic Elegance woofer too.

Even as far as tweeters, I feel more safe with the RAAL and Transducer labs stuff, and even though Jeff Bagby mentioned how great it sounded, CSS found they had to discontinue their LD25X tweeter because of chinese manufacturing being unreliable WRT QC.

I'm not against all Asian drivers of course. It seems SB Acoustics, which is located in Indonesia, has very high quality assurance. Their drivers are basically Scanspeak-quality, and indeed their high end Satori stuff is Scanspeak price too. Dennis uses the SB Acoustics 8" in my Philharmonic 2s, and it's fantastic.

Even with the fourth order crossover, I think you are failing to appreciate the massive power burden of BSC with a crossover that low.

Yeah, but would that cause more failures than bookshelves crossed over to subwoofers at 80hz? I'm not denying that the Salon mids are crossed over lower than I like. I'm just saying they're also crossed over higher than the norm. Even with more power going into the high-end speaker, how much of it actually reaches the 6" mid itself? My gut tells me the least efficient drivers in that system are the 4" mid and the 8" midwoofers, and the most efficient drivers (and hence padded down) drivers are the 6" mid and the Be Tweeter.

I'm not arguing for or against, just thinking out loud. The case just seems too exceptional.

 

[TLS Guy]

I don't think we disagree. It's just that especially with chinese manufacturing, the glue itself might not be up to spec. Kevin Haskins, who designed my subwoofer driver, went out of business because of one bad asian batch. My sub is still working great with with thousands of watts (not really ever) going into it, but the drivers from that batch all had the wrong glue on the shorting rings, causing them to become dislodged. It's not a cheap driver... I think I paid about $450 for it and it weighs around 60lb.

This guy had eight of the same driver, except from the bad batch with the "chinese swapped-out glue", and all of them failed almost instantly when power was applied:

Granted, good batches with good QC also exist, and I'm positive my driver is from a good batch because it's still pulling along. But it lends credence to the idea of only buying North American manufactured stuff and more importantly, I hope the driver companies move manufacturing away from China. My next sub driver is going to be the Mach V Phallus, which is made right here in Canada. The build linked in my sig uses a high quality American custom manufactured Acoustic Elegance woofer too.

Even as far as tweeters, I feel safe with the RAAL and Transducer labs stuff, and even though Jeff Bagby mentioned how great it sounded, CSS found they had to discontinue their LD25X tweeter because of chinese manufacturing being unreliable WRT QC.

I'm not against all Asian drivers of course. It seems SB Acoustics, which is located in Indonesia, has very high quality assurance. Their drivers are basically Scanspeak-quality, and indeed their high end Satori stuff is Scanspeak price too. Dennis uses the SB Acoustics 8" in my Philharmonic 2s, and it's fantastic.



Yeah, but would that cause more failures than bookshelves crossed over to subwoofers at 80hz? I'm not denying that the Salon mids are crossed over lower than I like. I'm just saying they're also crossed over higher than the norm.

Actually the larger cone movement of the bookshelf will cool the voice coil more than the equivalent mid range. To my ears most bookshelves have only partial or no BSC.

If you make a big speaker like the Salons, a reduced tenor range will not go down well. I think people do not expect really high spl from a bookshelf and they don't generally provide it. A less than robust sound is more easily forgiven and tolerated in a bookshelf than a large expensive speaker I think.

 

[GranteedEV]

To my ears most bookshelves have only partial or no BSC.

Fair point.

Actually the larger cone movement of the bookshelf will cool the voice coil more than the equivalent mid range.

That's a good point. Let's examine it, since I brought it up and others might be interested, and since we can.

Same driver, the Accuton C173 with 150w input into both. X-axis is frequency and Y-axis is excursion.:

Exhibit A (white) is a ported bookshelf (tuned to 45hz) with 2nd order electrical BW @ 80hz as THX-recommended receiver crossover
Exhibit B (yellow) is a sealed midrange crossed at 150hz, 4th order acoustical to approximate something like an LR4 (a true LR4 has two Q = 0.707 2nd order high passes, this one's a bit more makeshift for brevity's sake)

So yes, I suppose the bookshelf would probably provide better voice coil cooling from the added driver motion, although due to the difference in electrical filters, it's debatable which loudspeaker has more current running through it and would depend on the source material.