Ascend Acoustics new Klippel optimized speaker

[TurtlePaul]

It looks like the score is higher than you would “expect” because:
- The wobbles in the on-axis response correspond to areas where they improve the PIR. Often these wobbles make both the on-axis and PIR worse bot they may be intentional here.
- I think the bass hump makes the PIR flat below the omnidirectional frequency, but also masks that the -6 dB on this is in the low 40 hz range. That gives a good bass score.
- The directivity index isnt smooth, but DI isnt in the score, PIR is.

I would take the score with a grain of salt. We often see +/- 0.5 with different reviews numbers. For example the KNS setting for number of sample points clearly impacts score. Also, the DI turns distinctly negative in lower frequencies which frankly isnt likely true, making me this some of the settings arent perfectly dialed in.

I wish it was crossed lower or less steep on the tweeter/steeper on the woofer to make the speaker wider in the 2khz range. Then again, the DI data looks a little off. Still looks like a very interesting speaker.

 

[Sancus]

There's really no point to the score for speakers at or around 7 points. These are already firmly in the highest scoring +/- 1pt bucket. Once you've reached this point, you can be assured that tonality and basic off-axis performance are good and so other factors become more relevant.

That said, the measurements do indicate this is probably about as good as it gets for 2-way wide-dispersion, standard design, basic bookshelf speakers. The directivity isn't as even as the best waveguided monitors, but those have narrower directivity so are less suitable for those that prefer the wide special effects. The distortion isn't available but given the drivers involved I'm sure it's fine.

There are still some things you just can't get from standard design bookshelves. Good vertical directivity without lobing is one, this is ignored by the spinorama but it's certainly not inaudible in some rooms and applications. Also, low frequency directivity control. But that last one only comes in at much higher price points for now at least.

 

[BrokenEnglishGuy]

Weird.
Waiting for their distortion measurements.
That being said, hope their ribbon tweeter also measures well

 

[617]

I'd expect distortion to be extremely good with the exception of around 1200hz or so. Tweeter will be operating very low.

This is a very appealing alternative to revel and kefs Chinese made offerings.

 

[RMW_NJ]

Dave posted a price of $1398 plus $50 shipping for the initial run. Likely $50-100 more after the pre-order. I placed an order.

 

[Sancus]

It's very cool that he's offering to include NFS measurements at an additional charge. For those who want to EQ for the last bit of perfection, that's actually useful.

 

[pierre]

I sent Dave F. An email to see if he would agree sharing the data (and publishing). Depending of what he tells, I will publish data/score/EQ …

 

[mmi]

I sent Dave F. An email to see if he would agree sharing the data (and publishing). Depending of what he tells, I will publish data/score/EQ …

It might also be worth asking if you can publish the 2EX data posted in his forums and here previously.

 

[RMW_NJ]

Distortion.

 

[ribosradagast]

Distortion.

is that, errm, good or bad?

 

[TurtlePaul]

is that, errm, good or bad?

That is good. I would say that the bass distortion for a speaker this size is very impressive - especially at 86 dB, which would correlate with moderately loud levels in most medium sized rooms (75-80 dB at listening position).

Distortion is less than 1% if there is a 40 dB gap between the signal and the distortion. This speaker can achieve this well below 100 hz, which isn't common on bookshelf format speakers. At 96 dB, very loud levels, you see the bass distortion rising but it is still reasonable which you don't often see in a speaker this size. If the distortion rises to 30 dB below the fundamental, then that is 3% ditortion. Typically, below 100 hz, distortion isn't an issue until it rises to above 3-5%.

Outside of the bass, it has very good distortion in the midrange between 500 hz and 2 khz. The main acoustic issue to note is at 2.8 khz. This may be the woofer break-up frequency. It is somewhat well controlled at < 1% THD even at 96 dB, but you can see it on the graph. A trap could be added to the crossover to reduce this (or maybe it is already reduced) but it would add another 2-3 components worth of expense to control what is already only 0.5% distortion at 86 dB.

Of course, it would be ideal to have all of this data validated by a third party such as Amir, which would make sure it is consistent with the methodology used for all the other reviews here.

 

[617]

is that, errm, good or bad?

Given the bandwidth of the tweeter required for the DI shown in the first graphs, it is quite good.

This is an excellent performing speaker, close to state of the art for a direct radiating speaker with no waveguides or anything.

The NFS validates this design, but the basic idea for this speaker is - use really high end drivers, cross over low and gradually. Basically the same idea as 80% of other small audiophile designs, but executed really well.

What makes this speaker somewhat unique is the excursion of the woofer, which at 22mm is very, very high. I think the SB acoustics woofers at this size are something like half that - this is closer to what Purifi offers.

Ascend is a small shop, so they can't do all the engineering. This is kind of cool, because it means that they buy really high end drivers for their speakers and offer them at very good prices. The SEAS titan woofer retails for USD $170, which is quite a bit of woofer for a speaker in this price bracket.

 

[ob1]

I do not want to start a riot, but looking at Pierre Aubert's take on the spin (MANY THANKS FOR YOUR WORK, btw @pierre), I can't help but scratch my head : when looking at the data tab, the sensitivity is quoted as an incredibly low 75.4dB (300Hz-3kHz)

https://pierreaubert.github.io/spinorama/speakers/Ascend%20Acoustics%20Sierra-LX/Ascend%20Acoustics/index_vendor.htm

Since the data has just been posted, chances are this could be a typo or else, it is not standardized and one should not make much of it?

Yes, watts are cheap and the distortion figures are awesome @86dB and still very good @96dB, but I can't help thinking something is off.

If one clever ASR member could enlighten me, please?
Thanks in advance!

 

[tuga]

Dave Fabrkiant bought a Klippel a few months back and he’s using it to design a new speaker based on the SEAS Titan tweeter/woofer. It appears to be a replacement for the Sierra 1.

A Klippel to design a speaker?

 

[Beave]

I do not want to start a riot, but looking at Pierre Aubert's take on the spin (MANY THANKS FOR YOUR WORK, btw @pierre), I can't help but scratch my head : when looking at the data tab, the sensitivity is quoted as an incredibly low 75.4dB (300Hz-3kHz)

https://pierreaubert.github.io/spinorama/speakers/Ascend%20Acoustics%20Sierra-LX/Ascend%20Acoustics/index_vendor.htm

Since the data has just been posted, chances are this could be a typo or else, it is not standardized and one should not make much of it?

Yes, watts are cheap and the distortion figures are awesome @86dB and still very good @96dB, but I can't help thinking something is off.

If one clever ASR member could enlighten me, please?
Thanks in advance!

Must be a typo or other kind of mistake. From what I see in Ascend's plots, sensitivity is around 81-82dB.

 

[Shazb0t]

A Klippel to design a speaker?

Is it crazy to think having the ability to easily generate accurate anechoic 360° sound field measurements of your prototypes would help you to optimize things and create a better design?

 

[brandonhall]

Here's what jumped out at me from all of his posts:

Even as I type this, a new truly 3-dimensional standard is being developed, of which I have been asked if I would like to participate in.

Also, it must be said that his focus on engineering and breakdown of the design process is excellent content. Even if there are some calculation mismatches along the way, it's clear this is an objectively excellent speaker.

 

[617]

Must be a typo or other kind of mistake. From what I see in Ascend's plots, sensitivity is around 81-82dB.

Calibrating a frequency response chart to a standardized output level is not a simple task. I'm not sure how the Klippel NFS does it.

Quantifying speaker output in an echoic condition is sort of tricky for us DIYers since the room will contribute to output. Do you include that in your sensitivity specification? I'd say ideally not.

To do accurate sensitivity measurements you need to start with a known pressure source, then measure the microphone voltage produced by that pressure source, then the gain from the mic preamp, then figure out what that corresponds to after ADC.

Which is a roundabout way of saying, I'm sure the Klippel can product accurate sensitivity figures, but it's not trivial, and it's not that important for design purposes.

Generally sensitivity will be around 6db less than the quoted woofer sensitivity due to baffle step loss in the bass. Woofer is in the 86-84 db range, so you can expect system sensitivity to be around 78-80, which is to say, very low.

 

[hardisj]

Calibrating a frequency response chart to a standardized output level is not a simple task. I'm not sure how the Klippel NFS does it.

It regulates the voltage output so if you tell it to spit out 2.83v, it does that.

Microphone sensitivity can also be entered.

As long as Dave got that right (which, I’m sure he did), it’s pretty much foolproof on the Klippel side of things. Even if he tested with a different voltage, the software will adjust based on what you tell it to. For example, 2.83v. Or if someone wanted to make up their own standard they just type the desired voltage in. I do that for my powered speaker tests, where I use 0.30v.

 

[hardisj]

For what it’s worth, I believe I saw where Dave made the statement that he is the first internet direct manufacturer to design a speaker using the near field scanner. That is not accurate. Buchardt - an internet direct manufacturer - used the NFS to design their S400 well over two years ago. Which means Buchardt is likely the first to fit that description.