OK look to get everyone on the same page: a primer (2 hours) from Klippel on the NFS:
A shorter one (minutes)
Ultra-short time-lapse (seconds)
A shorter one (minutes)
Ultra-short time-lapse (seconds)
The benefit of the Klippel NFS is the processing that can remove the room from the measurements. Therefore allowing high resolution, quasi-anechoic, even at low frequency.
Boden posted a DIY attempt at doing this same processing, but I haven't read the whole thread so I don't know if it was successful.
Boden posted a DIY attempt at doing this same processing, but I haven't read the whole thread so I don't know if it was successful.
I'd try contacting Amir at Audio Science Review if I wanted my speakers Klippel tested. He's already tested several speakers sent in by members so he might agree to perform private testing for a fee. Personally I'd be worried about the cost, effort and safety of shipping my "babies" for testing.
If Erin is on the bench indefinitely would he be willing to rent his Klippel to you? He'd get some income from a Klippel collecting dust and you escape the very high purchase price. Just a thought.
If Erin is on the bench indefinitely would he be willing to rent his Klippel to you? He'd get some income from a Klippel collecting dust and you escape the very high purchase price. Just a thought.
Brandon's, and my upcoming speakers. 😎
For sure a great spinorama doesn't guarantee a great speaker. But it does give more insight into the speaker's directional behaviour than the single on-axis measurement that is published by some manufacturers.
I'd say it takes out 30% of the guesswork of saying a speaker is fatally flawed or well designed.
In fact, based on Erin's 100+ measurements of speakers on his website, there's probably only a handful I'd even audition, at home, before I'd buy.
We don't yet have a speaker that does DC to 20Khz from a single source, plays 0-120dB with zero amplitude or phase error. Equally, not everything that can be measured matters, and not everything that matters can be measured.
But that doesn't mean we aren't pushing ahead...
Last edited:
Well Brandon, I still like would like to know what is wrong with ARTA or REW, provided these systems are properly set up, a turntable is used for off axis measurements, and measurements are done in large and high ceiiing rooms, such as school auditoria or sports halls as per Zvu.
@Boden, one small thing is the absolute repeatability of the Klippel measurements, you move the mike a couple of mm:s and your curves are going to look different, if you don't spin the speaker around it's axis perfectly for every exact 10 degrees your data is going to be different, any measurements taken with a USB mike such as the Umik1-2 are practically useless for spin data and Vad work etc, etc.. But I find that the speed along with the accuracy that the Klippel measurements produce is a game changer. I can sit and change the DSP run it again for a reliable result at 0-90 degrees in less than a minute. I totally understand if you feel that you can do without that speed and accuracy though.
The German government pumped a lot of money into the University of Dresden and Prof. Wolfgang Klippel has a CV that speaks for itself, the Klippel NFS is an outcome of this collaboration. I am aware that the DIY community is full of positive thinking individuals that feel they can equal that and that's OK. Really depends on your level of commitment and ambition as a speaker designer.
If you really want to understand what they offer I can recommend attending the three day seminar that they give:
https://www.klippel.de/know-how/education/3-day-lecture-sound-quality-of-audio-systems.html
The German government pumped a lot of money into the University of Dresden and Prof. Wolfgang Klippel has a CV that speaks for itself, the Klippel NFS is an outcome of this collaboration. I am aware that the DIY community is full of positive thinking individuals that feel they can equal that and that's OK. Really depends on your level of commitment and ambition as a speaker designer.
If you really want to understand what they offer I can recommend attending the three day seminar that they give:
https://www.klippel.de/know-how/education/3-day-lecture-sound-quality-of-audio-systems.html
Last edited:
There are countless examples of speakers designed using Klippel, from every single Seas driver to Majico speakers. Ask Lars Risbo of Purifi what he considers to be the SOTA in speaker measurement. Your own technique of arguing could really do with some polishing though..
Made a video about Klippel vs DIY. I forgot to mention, even though you might think the <1khz resolution of the Klippel is cool but not a big deal, a lot of diffraction and resonance issues occur here and our smoothed responses just don't see them.
EDIT: just noticed I kept saying "mic stand" when I was describing the speaker stand
EDIT: just noticed I kept saying "mic stand" when I was describing the speaker stand
Last edited:
I feel like this is a silly question. What we gain from designing a speaker using a Klippel NFS is measurement accuracy the likes of which we probably cannot replicate elsewhere. The badge of "designed with klippel" I feel should be less interesting to us DIYers than the prospect of having access to that quality of data. At the end of the day, it comes down to the designer. You could place me in a room with the greatest workshop in the world and I would struggle to make you a simple wooden box. If you are skeptical about the helpfulness of using a Klippel NFS, I think the question would be better posed as "Does accuracy really matter?"
No offense meant 🙂 hope everyone is having a good day
Almost, the question is more ‘does this level of accuracy really matter OR increase the quality of your diy effort?’and then ‘what is the % increase over a low cost or free product?’ And finally “how do you measure that increase’…. Sorry for the unpolished arguement but expensive analyzers lead to (unfounded, subjective) arrogance.
It's not a silly question at all.
It's a bit extreme I agree with you, but there is most definitely something to say for the fact that there are multiple approaches.
People can get about 90-95% of the data of a Klippel system with just being smart about it.
The downside is that it takes more work and effort, but you can do quite the same.
There are some other measurement techniques recently described on this forum recently for example, that are most definitely doable for any DIY'er OR professional.
To justify $100k in spending for most companies, is rather difficult.
The majority of companies I know only do a full redesign every 3-5 years or so.
And most of those designs are only just a different flavor of what they did before.
That is a lot of money to spend and invest for something that is very doable with other techniques as well.
So in the end it's mostly a luxury product.
In the end of the day it still doesn't make anyway design (and sometimes understand) better speakers if you still don't know what to look for, or understand the underlying physics. That is 100% true.
Still the proof is in eating the pudding.
Anyway, going back to the original question @augerpro
For me personally, the only thing that is interesting, are the raw driver measurements.
Erin has a couple on his website, VoiceCoil magazine do have a couple.
But unfortunately it seems that they are only interested in speakers in the higher price range.
Which to me are always extremely boring reviews to being with, since we already except those drivers will perform very very well. So what's left is the check in consistency and if manufacturers are actually giving us the right information.
There are many, many other speakers out there that perform remarkable well.
That being said, I would think about the initial investment (apparently people have just 100k lying around, but that's a different story), and how to either profit from it, or make something work to let itself pay back over time.
I think working something like a patreon system would work well here.
Most hobbyists are not willing to spend more than 10-30 bucks, at least on this side of the pond.
It's a bit different for the more exclusive projects, but they seem to be scares these days.
So you have to get it from bulk, I think there are more people willing to spend like 5 or 10 bucks once, than more money per every time again. Hack for 10 bucks on average, you only need 1000 people.
Some are probably willing to tip in a little more even.
You can ask the same question for commercial products as well?
It’s totally up to you, as I keep saying. It’s kind of though to form an opinion though before you tried it, don’t you think?
The world is ripe with folks who are absolutely convinced that what they make (in all kinds of fields) themselves is superior to commercial efforts, or the work done in academia. This is a forum for and that’s mainly populated by these folks and I applaud this spirit.
@Indigent Audio
btw, it's not only about raw accuracy, but also just convenience and consistency.
A lot of people don't even have to space to measure loudspeakers for example, they find the whole measuring thing very complicated and confusing, or don't want to spend around 150-200 bucks for the equipment.
So to some extend, yes it will make them better speakers, since they only have to focus in making them, instead of understanding all the measurement techniques behind them.
The strength here is the practicality of it.
btw, it's not only about raw accuracy, but also just convenience and consistency.
A lot of people don't even have to space to measure loudspeakers for example, they find the whole measuring thing very complicated and confusing, or don't want to spend around 150-200 bucks for the equipment.
So to some extend, yes it will make them better speakers, since they only have to focus in making them, instead of understanding all the measurement techniques behind them.
The strength here is the practicality of it.
I would think that any practicality here would be lost in the idea augerpro posed though, except for those that live near
That is definitely a thing in the US, yes. Since shipping is quite high over there.
But in the end, that's the whole answer to the main question to being with.
No matter what direction you're going with it.