[[email protected]]

Hi,

I have this strange project in mind, to write a Bass-Reflex and Closed
Box enclosure simulator for HP 49G+ calculators. These calculators can
run codes which you compile with HPGCC therefore the code will be in
C++.

I think I can manage the circuit analysis part but first I need to
calculate the component values which are present on the equivalent
circuit. I have examined many documents and none of them are clear
enough to give the calculations of all the components, and their names,
like Mac, Mas, Mad differ from one source to another making it
impossible to track the equations.

I have examined the equations from the site below;

http://en.wikibooks.org/wiki/Enginee...closure_Design

but I have realized that they haven't explained how the values of the
circuit elements are calculated.

There are, however, some sources which explain the calculation of he
values of the circuit elements by measurement, but when you use a
software, you only enter T/S values like Qe, Qm, Fs, Re, Vas. You never
enter, Cms, Mac, Rec, Map or alike. Therefore, I need those parameters
in terms of T/S parameters.

Does anybody know a reference where I can find those or can anybody
help me with this?

Thanks in advance,
Regards
Onur

[[email protected]]

wrote:
Hi,

I have this strange project in mind, to write a Bass-Reflex and Closed
Box enclosure simulator for HP 49G+ calculators. These calculators can
run codes which you compile with HPGCC therefore the code will be in
C++.

I think I can manage the circuit analysis part but first I need to
calculate the component values which are present on the equivalent
circuit. I have examined many documents and none of them are clear
enough to give the calculations of all the components, and their names,
like Mac, Mas, Mad differ from one source to another making it
impossible to track the equations.

I have examined the equations from the site below;

http://en.wikibooks.org/wiki/Enginee...closure_Design

but I have realized that they haven't explained how the values of the
circuit elements are calculated.

There are, however, some sources which explain the calculation of he
values of the circuit elements by measurement, but when you use a
software, you only enter T/S values like Qe, Qm, Fs, Re, Vas. You never
enter, Cms, Mac, Rec, Map or alike. Therefore, I need those parameters
in terms of T/S parameters.

Does anybody know a reference where I can find those or can anybody
help me with this?

There are plenty of sources for the conversions. Since they are,
after all, the "Thiele/Small parameters," why not go back to Thiele
and Small and get them from there?

But why bother? The entire point to the Thiele Small model is
that you end up with a single consistent set of units that, in
their entirety, can be used to derive the transfer function of
the system. Again, see the original papers by Thiele in the early
1960's (in IRE Aus, 1961) or their reprints in JAES (1971) and
the original set of the Small papers on direct radiator loudspeaker
analysis in 1971-1972. Look later at some of the papers done
by Bullock, where he describes a single uniform model for closed
and vented systems.

You can also check some of the work done by Leach in the early
1980's on direct synthesis of equivalent circuits of driver transfer
functions if you really want to go down that road Lok also at the
conference paper done describing the origins of LEAP, though I
don't recall the author off the top of my head.

One of the BIG problems in trying to use a pure circuit-analysis
appropach is that there is no direct electrical analog for the
radiation
impedance, nor can one be synthesized out of linear passive
components as even in its simplest model, the characteristics
require describing it with a Bessel function. If you're talking about
putting together a mesh analysis program that includes such an
element, then fine, it will work, but it's a LOT of work.

[[email protected]]

I have the book of W. M. Leach "Introduction to Electroacoustics and
Audio Amplifier Design", in which he solves and simulates every
enclosure using SPICE. He might be a great SPICE fan but what if
someone tries to solve them manually? That book fails to become the
perfect study material as no alternative way of calculation is
supplied.

On the other hand, LEAP can be a good reference. I will look into that!
Thanks for the advice.

I am not planning to code a complex code. My aim is to give a broader
idea of what happens when you tune the enclosure to a lower frequency,
how the response changes when you increase the volume and such. On a
132x70 pixel display, I don't have the luxry to offer much though.

Thank you
/Onur

You can also check some of the work done by Leach in the early
1980's on direct synthesis of equivalent circuits of driver transfer
functions if you really want to go down that road Lok also at the
conference paper done describing the origins of LEAP, though I
don't recall the author off the top of my head.

One of the BIG problems in trying to use a pure circuit-analysis
appropach is that there is no direct electrical analog for the
radiation
impedance, nor can one be synthesized out of linear passive
components as even in its simplest model, the characteristics
require describing it with a Bessel function. If you're talking about
putting together a mesh analysis program that includes such an
element, then fine, it will work, but it's a LOT of work.

[[email protected]]

wrote:
I have the book of W. M. Leach "Introduction to Electroacoustics and
Audio Amplifier Design", in which he solves and simulates every
enclosure using SPICE. He might be a great SPICE fan but what if
someone tries to solve them manually? That book fails to become the
perfect study material as no alternative way of calculation is
supplied.

Well, he has an AES article which does get you there.

On the other hand, LEAP can be a good reference. I will look into that!
Thanks for the advice.

While the LEAP concept is VERY comprehensive, in thatg
it includes mutual coupling of multiple drivers and ports, as
well as a semi-realistic radiation impedance model, it is,
in essence, essentially the same approach that Leach took,
except the LEP article has next to no implementation
information, as it was describing a commercial product.

I am not planning to code a complex code. My aim is to give a broader
idea of what happens when you tune the enclosure to a lower frequency,
how the response changes when you increase the volume and such.

Then you absolutely DO NOT have to go through the gyrations
of attempting to model it with an equivalent electrical circuit.
Doing it that way is a complete waste of time. The standard
Thiele/Small model equations will give you everything you
need.

On a
132x70 pixel display, I don't have the luxry to offer much though.

Then all the advantages of an equivalent circuit based model
are lost, so why bother.

For example, t model the behavior of a selaed box system \
in the piston range requires a single 2nd order equation,
that's it. The only relevant parameters are the system
resonance and the system Qtc. Plug these in to the
equation and you're done. For vented systems, it's
a fourth order equation, but it's still FAR easier to do and
will give results as accurate as an equivalent circuit
approach, if you ignore higher order effects like mutal
driver/vent coupling mass and such. But that can be
modelled in either case as well.

Take a look at http://www.rdrop.com/~billmc/, and
follow the link "DIck Pierce's Spreadsheets". This is a set
of spreadsheets I did 20 years ago under the old Unix
SC spreadsheet calculator that does essentially what
you seem to want to do. The SC format is human-
readable text, so you should be able to follow it through.
It;s a beginning.

Goodness, and I see I even have real documentation
there as well.

[[email protected]]

wrote:
Right below your link, there is a code for HP 48 series (in RPL) ! It
is interesting that, your code and the web page on which your links
are, do not show up when you perform a search using a search engine.

Well, I found 'em in about 5 seconds with Google. I had long
since moved on to other methods and, frankly, didn't keep
a copy around. Bill McFadden was good enough to have made
them available on his web page.

Well, as for your code. The only human readeable file I could find was
the file with the ".ps" extension.

I would suggest you look at all the files that end with the
extension.sc. They are all quite readable and, if you alledge
to be sufficiently proficient at C/C++ to have decided to launch
out on your own, then reading these files could be easy.

I have started going through the
figures you have given and tried to form the relations between the
electro-mechanical parameters and Thiele/Small parameters. In figure
7.1 those calculations vere given. Yet, there are some points which are
confusing. Like;

If you look at Cms, you can see that it uses cells C21 and C7. You jump
to C21 and it is Qms allright. You branch from C21 and you see that it
uses cells C22 and C23, and you see that they are efficiency and SPL
values? Lets' scratch that, there are some cells which are being used
like D35 and D34 which are out of the boundaries of the table, and
there are some cells which contain a formulation which is refering
itself like the case in C12. On the next page there is a finished
example, therefore I have thought that these are not errors but
possible typing mistakes?

Look, as I said, the last time I worked on this stuff was 20 years
ago. All the stuff was working fine when I released it into the
public domain.

If you say that you have the complete
documents, would it be too much for me to ask you to kindly forward
them to my email address which is :

Everything is one that one page.

Just another question. In many text (Res + Re) is used to show the
maximum impedance value at drivers resonance frequency, however, it is
used in calculating Rms, which makes it mandatory to be measured? Does
it cancell at some stage in the calculations, therefore no one ever
need to measure it?

Huh?

[[email protected]]

For those who have followed this discussion, please visit Bodzio
Software, the coders of SoundEasy, and find the documentation on
BoxCAD, the link is given below ...

http://www.interdomain.net.au/~bodzio/

The first 3 chapters are all about this topic and the equations are
given suitable for coding.

Regards
Onur