[jsalk]

FYI -

When we originally worked on the Sonicap upgrade package for the HT3's, we also offered a bypass cap upgrade for an additional $200.

We recently re-worked the basic upgrade and added bypass caps to critical components only.  Since this added only $50 to the Sonicap upgrade price, we decided to offer it as a standard part of that upgrade and eliminate the bypass cap upgrade option.

- Jim

[Papajin]

Hi Jim,

Do existing orders with the original upgrade path get the new version of it, or will they still get the old version?

[Wally King]

If one gets no other upgrades besides the cap upgrade, exactly what audible improvements can be expected vs. stock?

[jsalk]

Papain -

Hi Jim,

Do existing orders with the original upgrade path get the new version of it, or will they still get the old version?

Yes.  The bypass caps are now standard with the upgrade.

- Jim

[jsalk]

Wally -

If one gets no other upgrades besides the cap upgrade, exactly what audible improvements can be expected vs. stock?

The stock crossover is extremely good.  In fact, much that has been written about the HT3's to date is based on the stock crossover.

But if you have good source equipment, you will notice a very slight increase in transparency.  It is like it removes the last little hint of "graininess" (for lack of a better term).  In short, it takes a great sounding speaker over the top.

Is the difference dramatic?  No.  It is quite subtle.  In fact, if you A/B both versions, it will take several iterations before you can discern a difference.  But once you know what to listen for, it is audible.

Is it worth the upgrade price?  That is something I cannot decide for someone else.  My personal pair has the standard crossovers, primarily because when people come for auditions I want them to hear the standard version.

If I were to build a pair strictly for personal use, I would probably put the Sonicaps in, just to know I maxed things out.  But the HT3's do a great job without them.  I know of no one who is in the least bit dissatisfied with the sound of the standard crossover.

I hope that gives you a flavor for the difference.  It is quite subtle.

- Jim

[rblnr]

What type of crossover is used: first order, second order, etc.?

[jsalk]

rblnr -

What type of crossover is used: first order, second order, etc.?

The passive crossovers are 4th order with an additional trap circuit on the midrange.  This is critical with the W18 in order to avoid cone resonances.  The W18 is an incredible driver when cone resonances are taken into consideration.  


In the active version, 96db slopes are used.  In this case, the additional trap circuit is not needed.

In both cases, the driver integration is seamless.

- Jim

[zybar]

In both cases, the driver integration is seamless.

That statement is no BS - it is truly seemless!!

George

[skrivis]

rblnr -

What type of crossover is used: first order, second order, etc.?

The passive crossovers are 4th order with an additional trap circuit on the midrange.  This is critical with the W18 in order to avoid cone resonances.  The W18 is an incredible driver when cone resonances are taken into consideration.  

You mean a notch filter? So you need a 4th-order crossover and a notch filter to tame the W18? Ouch.

In the active version, 96db slopes are used.  In this case, the additional trap circuit is not needed.

In both cases, the driver integration is seamless.

- Jim

This was the first time I had seen an active version mentioned, so I just checked it out on the web site. It definitely sounds like the way to go. Nice work!

Have you tried a pair of Frank Van Alstine's OmegaStar 240/EX amps with the HT3a? The pair would be less expensive than an ATI amp, and I would bet you'd get better performance.

I'm hoping that lots of people buy DEQX units so they'll go into volume production and drop the price.

[DMurphy]

"You mean a notch filter? So you need a 4th-order crossover and a notch filter to tame the W18? Ouch"

Hi  whether you call it a trap or a notch, it's just an inductor, cap, and resistor in line.  The crossover is 4th order acoustic, not 4th order electrical, so you don't end up with as many components as you might think.  A trap plus 4th order electrical would give you at least 7 components just for the low pass.  The 4th order acoustic with the lcr filter only uses 4 components.

[skrivis]

"You mean a notch filter? So you need a 4th-order crossover and a notch filter to tame the W18? Ouch"

Hi  whether you call it a trap or a notch, it's just an inductor, cap, and resistor in line.  The crossover is 4th order acoustic, not 4th order electrical, so you don't end up with as many components as you might think.  A trap plus 4th order electrical would give you at least 7 components just for the low pass.  The 4th order acoustic with the lcr filter only uses 4 components.

I just hadn't heard it called a trap before.

[rblnr]

The crossover is 4th order acoustic, not 4th order electrical

I think I understand the difference between acoustic and electrical in this case, but could you elaborate?

Thanks.

[DMurphy]

The crossover is 4th order acoustic, not 4th order electrical

"I think I understand the difference between acoustic and electrical in this case, but could you elaborate?"

Hi.  I can try.  Let's just worry about the woofer circuit in a 2-way.  And let's suppose the woofer's response is perfectly flat from 40 Hz to 20,0000 Hz.  And it's impedance curve is a perfectly flat 8 ohms above 100 Hz.  Of course, nothing like that exists on this planet, but just assume it for the moment.  Then, in order to roll THAT woofer off at the top at a 4th order slope--24 dB per octave--you would need to use four components.  A series inductor, a capacitor to ground, another series inductor, and then another cap to ground.  That's a 4th order "electrical" circuit.  It assumes a woofer that essentially behaves like a simple and prefect resistor. Real-world woofers aren't flat.  Not only do they have peaks and dips at the top end (say between 3 kHz and 5 khz). but they have a natural roll-off at the top end.  And their impedance increases rapidly as you go above 1 or 2 kHz.  If your goal is to roll THAT woofer off at 24 dB per octave--so that its "acoustic" output falls by 24 dB per octave--all bets are off concerning what crossover components will be needed to accomplish that.  With the W18, all that's needed is a series indcutor followed by an "lcr" circuit to ground.  That's an inductor, capacitor, and resistor in series with each other, but connected to ground after the series inductor.  The "lcr" circuit is a tuned "trap" that suppresses the W18's big breakup peak just below 5 kHz.  That suppression, combined with the natural roll-off of the W18, is enough to produce a 24 dB "acoustic" slope when it's preceded by a series inductor.

[Christof]

DMurphy

Out of curiosity, are additional components needed to knock down the upper Hz resonance in the Mag driver after you have hit your target 24db/oct with 4th order XO?

Thanks for your informative posts
c.[/b]

[JoshK]

DMurphy

Out of curiosity, are additional components needed to knock down the upper Hz resonance in the Mag driver after you have hit your target 24db/oct with 4th order XO?

Thanks for your informative posts
c.[/b]

I am not Dennis and can't speak for this design, but in general that is a yes.  That is the notch or trap spoken about earlier.  I imagine Dennis was keeping the explanation simple to make it clear between electrical and acoustic target slopes.

[DMurphy]

DMurphy

Out of curiosity, are additional components needed to knock down the upper Hz resonance in the Mag driver after you have hit your target 24db/oct with 4th order XO?


Hi   I'm not quite sure I understand your question.  The "upper Hz resonance" is The what the lcr circuit is designed to eliminate.  But in the process that circuit, in combination with the first series inductor, is enough to give you the desired 4th order acoustic roll off.  No other components are needed.  That was a major breakthrough in the design of the W18.  The previous W17, and the aluminum L17 series both had a much more complex breakup mode that required more components.

[Christof]

Dennis

You answered my question.  I was basically wondering if the w18 has the breakup that the L17 has and how it is delt with.  If it ain't broke I guess there is no point in fixing it