Sorry, I have not modeled it, the filters we make are evolution on evolution , we change listen change and listen, sometimes we even measure if we do something radical. Don't think that there's any simulation way. We may understand the principals of dynamic unlinearities, but to quantify them in models is not really possible. Like how do you quantify the non ideal spring of the spider an surround or the air compression in a flow chamber. Not really possible or desireable. From my previous job I have ofthen seen deviations of more than a 100% in FEM simulation strains compared to actual measured strains. For me FEM is only a tool with pointers, not a tool that holds the truth. Sure no one here really believes the 0,00000 something simulation of LT-spice that flow around here, but to get an idea about the circuit and to see how the currents are, it's really good, in the end of the day you'll have to build an listen
Yeah, my simulations were like that, at which point I handed that (and thermal analysis) over to those who were good at it
jn:
that's pretty good.
with all the remake madness these days, why won't someone do the time tunnel?
mlloyd1
that's pretty good.
with all the remake madness these days, why won't someone do the time tunnel?
mlloyd1
Finite element modeling has come a long way with the use of powerful desktop computers and there are many mechanical modeling programs that can do much with loudspeaker design. With that is the common poor results when we use many lumped sums and overlooking the small details. Take a spider for example and you can easily model or measure the actual force/distance values to determine the spring properties but at the same time if you do not include the air loading on both sides of the spider in actual use those numbers will be inaccurate at best. The same goes for the surround. While many look only at a surround as a centering device and also a secondary restorative device returning the cone to the neutral position little is looked at as regarding how the surround dissipates the energy from the edge of the cone and the inherent acoustical signature of the material that the surround is made of. It is all the small details that are overlooked as to complex that make the models inaccurate. Just as all the little parasitic elements in an electrical circuit can determine the final outcome of that circuit all these little factors in a speaker itself need to be looked at or you are stuck with having to use subjective methods and trial and error to get the speaker right, this does not have to be the way forward. All those lumped sums need to be turned into discrete values that can be used in a better model giving better and more accurate results.
Last edited:
Maybe because it was soooo cheesy and the folks in TV land have scruples and sense to leave a dead horse alone.
Pffffft. I crack myself up sometimes..
To this day, I remain scarred for life. During the initial episode, I watched them driving on the desert, I turned my head for a split second, when I turned back, they were already underground...whattthappened???. I missed it.
sigh...
jn
Kindhornman and Miib, it would appear that you have had some success in loudspeaker design.
Personally, I quit trying about 35 years ago, when I made a pretty good speaker that measured well, made of the best parts that I could find, yet did not sound as good as I hoped. I realized that I just did not know enough 'details' of speaker design in order to do a first class one, on my own.
I find it similar in electronic design, and that is the main reason that I have declined to model much, but instead measure the real thing, and LISTEN to the results. Sometimes, even good measurements are not enough.
Personally, I quit trying about 35 years ago, when I made a pretty good speaker that measured well, made of the best parts that I could find, yet did not sound as good as I hoped. I realized that I just did not know enough 'details' of speaker design in order to do a first class one, on my own.
I find it similar in electronic design, and that is the main reason that I have declined to model much, but instead measure the real thing, and LISTEN to the results. Sometimes, even good measurements are not enough.
Hi,
I put up the relevant parts in another forum here awhile ago. Maybe a search on my name will find it? The ESP site is a copy of that work I published in 1984-5. But more details. The original concept was a subset of an amplifier article in The Audio Amateur 3/1985 ("Low-Feedback Amp design")
It has the actual before and after speaker distortion plots in it. If I can find that article... back in the days when I was prolific in All things audio.... I'll put it up again, here. I'll go look for it.
THx-RNMarsh
Last edited:
The Marsh speaker distortion reducer -
I found it. This is the original concept results. I have never seen this done earlier this way which is why I added it into the amp article:
View attachment Motional FB data reduction.pdf
As you can see, the proof of concept worked very well. Both closed box and ported. It was actually done first using a Dynaco 120 amp. As I said in the article, no attempt was made to get the thd as low as possible..... just that my concept worked well and could be added to the amp design shown. I showed the circuit with the DIY amp design.... a modest nfb vfb amp design.
Comments invited.... would like to see it expanded further for lowest distortion reduction. With the naturally occurring high rising distortion at low freqs, a reduction of 50% or more is quite audible improvement with no additional cost for more expensive speaker driver.
THx-RNMarsh
I found it. This is the original concept results. I have never seen this done earlier this way which is why I added it into the amp article:
View attachment Motional FB data reduction.pdf
As you can see, the proof of concept worked very well. Both closed box and ported. It was actually done first using a Dynaco 120 amp. As I said in the article, no attempt was made to get the thd as low as possible..... just that my concept worked well and could be added to the amp design shown. I showed the circuit with the DIY amp design.... a modest nfb vfb amp design.
Comments invited.... would like to see it expanded further for lowest distortion reduction. With the naturally occurring high rising distortion at low freqs, a reduction of 50% or more is quite audible improvement with no additional cost for more expensive speaker driver.
THx-RNMarsh
Last edited:
Thanks for that, Richard. It looks to be consistent with Pavel's data also.
Still, it looks to me like the motional feedback is positive.
😎
Mr curl, you're 100% right, simulation is for sections, where you can clearly define realistic border conditions and simplfy what you analyze and why, anything else is too complex and is at best a waste of time, in my past I did FEM/FEA analysis work for a Danish military sub contractor, IRL tests for strain and fatigue was always a part of the performance/test phase. Just like listening to speakers or amplifiers is what it, in the end, is all about, measurement is still a part of the testing, but not ever never as a design goal. For speakers the way we listen and understand music is too complex to be put in numbers.
For me I used FEM/FEA (still have some goodwill and a few MSC tokens) quite extensively in order to create the shape of the cone I use for the dynamic drivers, as we solely work with stiff cones, the mechanical problem was down to three variables mass, stiffens in shape and amplitude in fundamentals, with that in mind the job can be isolated to few realistic borderline conditions and linear elastic domains, then simulation will work with credible results.
For me I used FEM/FEA (still have some goodwill and a few MSC tokens) quite extensively in order to create the shape of the cone I use for the dynamic drivers, as we solely work with stiff cones, the mechanical problem was down to three variables mass, stiffens in shape and amplitude in fundamentals, with that in mind the job can be isolated to few realistic borderline conditions and linear elastic domains, then simulation will work with credible results.
Pass marsh negative output impedance...? Like you push, I push harder. Or is it in fact or you push I'll ease the pressure a bit, I think It can be handled by resistor load network where you take the feedback,
Also those base resonance impedance bumps is an electrical manifestation of mechanical resonances, it's the missing damper in the system, they can be handled or greatly reduced with good internal flow damping of the cabinet air system. This is by far more important than whether the cabinet material is wood, aluminum or like Wilson Bakelite.
Also those base resonance impedance bumps is an electrical manifestation of mechanical resonances, it's the missing damper in the system, they can be handled or greatly reduced with good internal flow damping of the cabinet air system. This is by far more important than whether the cabinet material is wood, aluminum or like Wilson Bakelite.
Last edited:
The point I tried to make was that there is no such thing as a Möbius coil.
The followers of ‘alternative’ theories are attracted by the unexplainable. They use terms, bits and pieces from very specialized disciplines of science and they stick them to their hazy mental constructs for to give them some sort of credibility.
https://www.youtube.com/watch?v=xLxxre80oQE
Mathematical topology is a branch of science for the brave hearted.
Any connection btn these 'coils' and Möbius band can be only visual and intuitive (at best).
There are aspects of the mathematic work of August Ferdinand Möbius that are visually beautiful as beautiful are mathematics in general (but don’t try too hard to make your kids agree with you on this)
https://www.youtube.com/watch?v=JX3VmDgiFnY
We see the name of A. F. Möbius (*) in Visual Mathematical Art works, mostly thanks to M. C. Escher work.
M.C. Escher – Gallery
George
(*) not to be confused with the work of the French comics graphic artist Moebius (Jean Giraud)
Moebius on his art, fading eyesight and legend: ‘I am like a unicorn’ | Hero Complex – movies, comics, pop culture – Los Angeles Times
Last edited:
Then what will the correct term for this induction free (freeish) winding method be..?? To me it looks like the same kind of winding
- Status
- Not open for further replies.