In designing enclosures, most people don't compensate for volume taken up by the driver, bracing and padding - or anything else for that matter.
If you don't use ideal volumes, then everything becomes more critical. I even looked up average air density and pressure for the area I sold in. As the box is reduced from optimal your tuning peaks and is far more critical on driver parameters as well. Thankfully if you design to the proper volume, things are pretty forgiving. Bigger boxes though.
However you learn these things through frustration when you're starting out and your designs don't work out. These days we have these factors figured out, and we have the internet. Not so when I began.
The passive crossover adds impedance to everything and messes with your design as well. All stuff you have to learn, there isn't really a list of steps you can follow.
If you don't use ideal volumes, then everything becomes more critical. I even looked up average air density and pressure for the area I sold in. As the box is reduced from optimal your tuning peaks and is far more critical on driver parameters as well. Thankfully if you design to the proper volume, things are pretty forgiving. Bigger boxes though.
However you learn these things through frustration when you're starting out and your designs don't work out. These days we have these factors figured out, and we have the internet. Not so when I began.
The passive crossover adds impedance to everything and messes with your design as well. All stuff you have to learn, there isn't really a list of steps you can follow.
I think the last 30 or so posts have been very much on topic. I have learned some things, so I assume many others have learned something too.
My belief is that making measurements which are good enough for speaker design is hard, but not because the process is complicated. It is hard because there are so many pitfalls, so many opportunities for error. If someone believes that making measurements is a trivial process of just connecting up some equipment and recording a value, they are very likely to make an error. This discussion helps bring awareness to this.
It is helpful to have reasonable expectations about accuracy and repeatability, and this discussion also makes us aware of that.
Actually the fameus quote "Measuring is knowing" is wrong and contrdictional, it is more like "to knowing through measurements ".That is exactly what I was trying to say!
Also 100% agree with the last part.
Some people use the expression that "measuring is knowing".
I always like to add that it's only true when "you know what you're measuring".
Meaning that results can still be totally useless if you don't understand what they mean.
Which can lead to potentially serious problems that someone is totally unaware of.
All that being said, let it absolutely not withhold people from starting doing measurements themselves!!!👍
Key here is just staying REALLY consistent and knowing the limitations of the measuring setup.
Get a couple of basics right and you can basically measure the vast majority of things almost like a pro.
The rest will come with knowledge and mostly experience.
So I really encourage anyone to start with it!
In other words: an observation can only lead to more understanding when it can be measured somehow.
The difficulty obviously lies in "can be measured" ;-)
My goodness. C'mon fellas.
When you make measurements you know more than if you didn't make measurements. (All other things being equal.) It's as simple as that.
You can get all cerebral about this and that and obsess about the pitfalls of various measurement techniques, but all-in-all you still have increased knowledge of your system/project by measuring it.
Dave.
C'mon fellas.When you make measurements you know more than if you didn't make measurements. (All other things being equal.) It's as simple as that.
You can get all cerebral about this and that and obsess about the pitfalls of various measurement techniques, but all-in-all you still have increased knowledge of your system/project by measuring it.
Dave.
B_force I would like to suggest you to slowly start a guide for measuring loudspeaker drivers. You seem very knowledgeable in the field, so it could be a great contribution.let it absolutely not withhold people from starting doing measurements themselves!!!
@anatech is it possible to fast make 10 comments to ones own thread and ask admins to unlock them for edit indefinitely?
-I dont know how many relevant measurements there are, but it would be beneficial that each has its own post.
@b_force Dont put too much into the following. They are only my notions and not based on thorough examinations😄:
My reservations are that you write very long. Could you maybe with the help of others input make its short and consice?
I rarely see you admit to being wrong or bend your initial opinion. Could you work together with those you've had harsh discussions with before, to improve and develop such a guide?
I would love to see such a sticky.
Cheers!
You worded it different, but mean the same. ;-)My goodness.
When you make measurements you know more than if you didn't make measurements. (All other things being equal.) It's as simple as that.
You can get all cerebral about this and that and obsess about the pitfalls of various measurement techniques, but all-in-all you still have increased knowledge of your system/project by measuring it.
Dave.
Except the measurement needs to be relevant, which may sound like kicking in an open door, but just the Rsense differences or voltage differences in otherwise identical measurement setup, shows it is not that obvious or straight forward at all.
My choice for Rsense of about 0.135 ohm (and 1, 2.8 and 4 volts rms) icm with Arta and an amplifier is based on:
- Is in line with how Purifi measures.
- Is "in line" with real world use of the driver.
It is my choice and gives me an understanding of the driver behaviour in its intended use. And if so needed a clear communication with the people of Purifi.
I personally give very little for standards if they do not match the real world use case.
Which f.i. in my view is the case with the TS parameter standard(s).
@JanRSmit
At the risk of being labeled circular......
Measurements don't have to be conducted in real-world conditions to yield valid data.
Nor does testing in non-real-world conditions necessarily make the results less relevant. (In fact, stress testing at beyond real-world conditions is valuable.)
Anyways, the topic of this thread is "driver measurements which are needed for speaker design."
Driver impedance measurements are indeed needed for speaker design. (In most cases.) How you make those measurements should be understood in the context. This is one of the possible "pitfalls" I was talking about.
Cheers,
Dave.
At the risk of being labeled circular......
Measurements don't have to be conducted in real-world conditions to yield valid data.
Nor does testing in non-real-world conditions necessarily make the results less relevant. (In fact, stress testing at beyond real-world conditions is valuable.)
Anyways, the topic of this thread is "driver measurements which are needed for speaker design."
Driver impedance measurements are indeed needed for speaker design. (In most cases.) How you make those measurements should be understood in the context. This is one of the possible "pitfalls" I was talking about.
Cheers,
Dave.
That's good to hear!
You were clear in the beginning not to derail to much, so I try to be very aware of that and respect that.
If we all together stay a bit aware of that, we don't have to go on to many side tracks.
Although the lines can be very blurry, so you decide
No, the point is if someone doesn't have a freaking clue what they are doing, results also don't mean anything at all.My goodness.
When you make measurements you know more than if you didn't make measurements. (All other things being equal.) It's as simple as that.
You can get all cerebral about this and that and obsess about the pitfalls of various measurement techniques, but all-in-all you still have increased knowledge of your system/project by measuring it.
Dave.
Since the person in question kind see the difference between actual useful data or noise/errors.
Worst cast is when measurements look promising for the untrained eye but in fact contain false data.
With the speaker unplugged, the result should not depend on the test signal, because there is only the noise and no test signal ;-)
The slightly more important issue might be that without any signal from the speaker/DUT, the software may have trouble to identify the time range where the impulse response would be, so it might bark at you. Of course that depends on the software and the configuration for automatic detection of the DUT response.
I have given workshops in the past, but they require a lot of time.
Even more so when the group of people all have different backgrounds and different levels of knowledge and experience.
Which makes it hard to find a balance between no going to fast, or skipping steps that seem obvious for others.
For example, I have also given practical classes at a physics school.
In that case the focus is much more clear.
Which brings me automatically to this point.
What I see people doing very often, is just dumb statements without explanation.
This is where most of the miscommunication start.
Nuances are extremely important to understand the entire picture.
In my opinion even forum posts are already very difficult to explain all the ins- and outs.
I am no fan of quick statement dumping with zero explanation and almost 100% assumption.
Learning requires reading and understanding.
That's very interesting, because I am an extremely open minded person actually.
If I made a mistake or similar, I always make sure to rectify this (and I very often do so actually).
I also have no problem changing my opinion at all, as long as someone comes with a proper argument or evidence.
But in my opinion there a couple of rules to have a good, respectful and mature conversation;
- Be pro-active in a discussion, which means also show initiative, come with explanations, evidence, proof and/or experience
- Followed up by this, don't automatically assume and judge things, but ask further
- There is a big difference between subjective opinion and just objective facts of physics.
Neither of the two is wrong, but they are very different animals and don't mix them up
- We all make mistakes or misread stuff. Keep in mind that there are many different personalities, cultures etc etc here.
The problem here, and this is why discussions sometimes drag on forever, is that "relevant" is undefined.
Without talking about context, and people just dumping quick statements, that's a useless discussion.
Some find it overdone because in their mind they only care about a simple impedance measurements for a crossover.
To some extend I find certain opinions kinda ironic, because some people seem to suffer from distortion anxiety.
Yet at the same time brush of core elements as if it's totally unimportant.
If we talk about so many significant digits of distortion, you simply can't deny BL(x), Kms(x) and Le(x).
As well as what these parameters do with even the system response itself.
And for those who have trouble understanding, that is not a subjective opinion.
That is just how the physics work behind all of this.
Take a deep breath man.
These are all adults out here and likely aware of the possibility of irrelevant measurements.
You're point is taken. No need to beat the drum over and over.
Dave.
A good example of miscommunication or misreading again
In my experience I am not beating any drums atm, I am having a happy conversation and it just wasn't clear to me if the point was understood well.
Also, it doesn't matter for me if somebody things differently about that point either.
You still have the ANALYSIS of the measured signal! And these react different to noise, some more sensitive as others.
As I'm working with an AP system the time reference is no problem. For programs like REW you should first do a proper measurement, do your windowing and stick with that window then everything is fine.
Never do full blind automatic measurements in acoustics - that's not an oscilloscope
We have time, and you already spend a lot of it here😊. You don't have to finish in a week month or a year. A common work with you updating OPs when you see fit. I suggest you start tomorrow 😄. No. just a wishfull idea😊
As for my reservations remember i said without examination. If i haven't seen you take something back doesn't mean at all that you haven't. Could also be, that you are rarely wrong and that's not bad.
Cheers!
I like multitasking a lot, meanwhile I am working at the same time.
Don't know why, but that's how my brain works.
Thanks, I actually have started multiple ideas.
Including actually writing a book, mostly because the current literature is very hard to digest for most people and most of it is very outdated.
If somebody feels seriously helping (and I mean on a SERIOUS level), anyone is free to contact me.
I think we should define which measurents (or parameters) are needed
1) for evaluation of drivers and choosing the most appropriate units for a project. Woofer/mid /tweeter duties ask for different properties...
2) for xo design/simulation with chosen drivers in a project, in the baffle/box being used. Both electrical and acoustic properties
1) for evaluation of drivers and choosing the most appropriate units for a project. Woofer/mid /tweeter duties ask for different properties...
2) for xo design/simulation with chosen drivers in a project, in the baffle/box being used. Both electrical and acoustic properties
Yeah, ... it is interesting that we are up to 180 posts talking about impedance measurements. I offered that topic up originally because I thought it would be very non-controversial. It turns out that there are some nuances that deserve to be discussed.
One thing I would like to get some opinions on is using an impedance curve to detect and assess resonances. For example
This is a small woofer in a sealed box. Notice at 640 Hz, there is a small burble in the curve. Zooming in a bit:
Here we see that the burble extends from 500 Hz up to around 650 Hz. After some investigation, I concluded that this was a small, insignificant cabinet resonance. My question is this: Is there a mathematic technique to assess this resonance? To assign a numeric value to it, based on the impedance curve?