Here is the problem. "Paper" is a composite. Glass fibers, many times carbon fibres. Different lengths on types of wood fibers and then binders. The binders can be thermoset or air curing depending on the cone makeup. The cone itself can be pressed, one sided pressed. two sided pressed. Find me a program that can develop a randomized fiber matrix and then calculate out all the potential differences in the modulus of mystery particles. COne manufacturers will not tell you their secret recipes.
So an engineer will tell you take a sample and perform a deflection test on it and generate a materials property and then simulate this. For that cone run it is a valid assumption. If the cone is aluminium it is a very valid assumption. If it is a composite, not so valid an assumption. The next time a composite cone of Paper pulp is made will it have exactly the same composition? That depends on time of the year for temperature and humidity. It depends on if the same batch of wood fibers was used, these are variable. It even can depend on pressing time. If the heating elements were at the same temperature or did someone increase the temperature and shorten the pressing time.
If you use the tools of simulation for what they can tell you they are useful. They can help you to discover reflections between a cone profile and the surround. Or if you modify a surround profile what it may allow in the area between 1.2 and 1.5 kilohertz on a 6.5 inch driver for example. What has been mentioned also is the empirical method of simply looking at existing examples will tell you as much or even more than any simulation.
Nothing in audio reproduction is simple. There is never a engineering choice that gives with both hands. You get something and you generally lose something. Some times you get what you want and also lose what you want. I read many times art of loudspeaker design. That is kind of true. It is basically a trade. I am a trained furniture maker, and have been trained in a few other trades as well. You are taught by someone with enough experience to get things made well. Some trades people continue learning and applying new and augmented methods. Some do everything exactly the way it was always done. The fact is we take building blocks of magnet, low carbon steel, cone composites and synthetic rubbers and fabrics for dampers, metal for frames and we make something. 99.9% of the time it's pretty much like everyone else's parts. We just changed the recipe a little. And we arrived at something either nearly the same or a little different. It still a woofer, or a tweeter or a midrange. There are thousands of them. The toughest ones to make are for car audio. Tiny budgets. Tight margins and they can sound pretty damn good. They impress me every time I listen to a good inexpensive loudspeaker. That is true talent. Making the most, with the least.
I think I have already seen very old Philips drivers with structures in cones.
https://www.textreme.com/product/0-90/090-biobased-flax
I've been doing work with Composite sound since last August. Never really looked around at the other stuff Oxeon/Textreme the parent company offers. Might this be interesting to you gentlemen?
I've been doing work with Composite sound since last August. Never really looked around at the other stuff Oxeon/Textreme the parent company offers. Might this be interesting to you gentlemen?
It sure is: I have a keen interest in all sorts of materials, even more so when those might be used for cones or domes. That being said, I haven not yet been able to detect the claimed superiority of many novelties, including Textreme. The persistent use of classic off the shelf surrounds, be it DKM or Chinese, does not help either.
But now that we're on it: here is a question for Mark and Lars.
Suppose we have an old school AR 3A with a relatively soft paper (almost felt like) coned 11" woofer in a closed box.
Does in your learned view that cone deflect and (slightly) deform in the 40- 80 Hz range if fed by a handful of Watts (so the cone has to move!) or does it stay purely pistonic, notwithstanding it relative softness?
Suppose we have an old school AR 3A with a relatively soft paper (almost felt like) coned 11" woofer in a closed box.
Does in your learned view that cone deflect and (slightly) deform in the 40- 80 Hz range if fed by a handful of Watts (so the cone has to move!) or does it stay purely pistonic, notwithstanding it relative softness?
40-80Hz is extremely low.
So my guess would be that if we assume that the cone isn't really rigid, either distortion goes up as well as Rms probably.
We will be in piston region nevertheless.
Just with losses. But that doesn't change the region we are in. Only performance.
Pistonic region ends roughly at minimum impedance or about 2-3 times Fs.
Although there is an argument to be made it ends just before the first cone resonance.
Depending on definition.
Edit:
This is just following lumped model of a loudspeaker. Aka mass-damper-spring system.
A less rigid cone would just translate in losses at those lower frequencies.
By the law of conservation of energy.
(Well lack off in this example)
So my guess would be that if we assume that the cone isn't really rigid, either distortion goes up as well as Rms probably.
We will be in piston region nevertheless.
Just with losses. But that doesn't change the region we are in. Only performance.
Pistonic region ends roughly at minimum impedance or about 2-3 times Fs.
Although there is an argument to be made it ends just before the first cone resonance.
Depending on definition.
Edit:
This is just following lumped model of a loudspeaker. Aka mass-damper-spring system.
A less rigid cone would just translate in losses at those lower frequencies.
By the law of conservation of energy.
(Well lack off in this example)
Last edited:
it’s quite difficult to measure Sd(x) directly. I did a setup with a driver in a ported box and a microphone inside the box. By adding DC the position of the cone could be stepped whilst measuring the AC responses. This confirmed the simulation results for half rolls. Switching from a had roll to the neural surround for the same cone and super linear motor showed a 20dB drop of the second harmonic at 150Hz where models showed that Sd modulation would be dominant.
I first did aluminium cones with custom profile and custom surround in 2014. I can tell you that they behave very differently to most other cone materials.
Here are the differences. If you know how to specify the cone to begin with. You can design a cone that is rather neutral. Then we come to the surround. Which I can tell you effects a lot more of the sound characteristic of a loudspeaker than you may believe.
Lars may have a different point of view considering how much work they put into their surround design.
I have been able to take open tooling aluminium cone and surround and gotten very low distortion in an underhung design. Posted below.
So this is 90db 2 meters ground plane measurement. So one meter equivalent for a 2 Pi measurement. It is not a close mic cheat where you drop down the SPL and pretend that is the distortion products at one meter.
That's a cross section of the driver. No sane person will be making this one. It was a lot of magnet. Unfortunately I had a small run to make for a client, and this was the only magnet in stock and available for something like this.
This driver has been redesigned and uses a much more modest magnet and Has just arrived here and will get some exhaustive testing.
So why post this? Well it's an example of what you can do if you think a little. I don't chase the lightest aluminium cone, I chase the best compromises. Chasing mass ends up with a cone that is a screamer. Engineering a cone that is not gives you about a 0.6db efficiency penalty. I can live with that.
Oh one last thing.
Ground plane, you can use these measurements up to the diameter of your mic capsule. So for most of them that is well beyond 18 kilohertz.
A vacuum chamber with just a standard current measurement and laser would show this immediately probably.
In fact you can probably get this already from just the current measurements.
We just have to keep track of the rate of change of Fs.
Besides the compliance, Mms would be the other variable here.
Mmd wouldn't change here, since that's a inherent part of the system.
But a non-linear Sd would change the Mms.
The difference between the rate of change between vacuum vs non-vacuum will give you a very decent indication of how significant a non-linear Sd will be.
Btw. Copyright and trademarked 😄😄😄😄
Last edited:
So still about half roll?
This has always been rather close for us. I actually use a plastic digital caliper and measure inside roll close to cone to outside roll away from cone. Within 10% changes very little overall when you use the numbers to simulate an enclosure.
So all this complicated insanity and still, you will end up with 50% of the surround if it is a half roll surround.
The fun and games begin when you do not have a half roll surround. Then you need a stroboscope and some quality time in the dark
Well actually, since Mms as well as Cms (Kms) are equally proportional to Fs, we need about drop 25% of Sd to be equal to 10% distortion.
If we assume a linear relationship, this would mean that a drop of 3.125% in Sd will result in 1.25% distortion.
I haven't done the math, but what's the worst case change in SD at full xmax?
If we assume a linear relationship, this would mean that a drop of 3.125% in Sd will result in 1.25% distortion.
I haven't done the math, but what's the worst case change in SD at full xmax?