Tribute to Lansing-Heritage-Forum and DIY 2405 & 2446 & E145 build

[berga12]

I want to add that TAD is the Japanese response to JBL....yes of course, they copy JBL and do it better on some parameters several years after.

also the 12" of TAD (the grey one) is a revisited 2202h with some improvements, but basically a copy.


To my mind, building a 4350/55 fully TAD would be one dream, but the cost is 3-4times higher...

I would go with TAD 1603 in your cabinets(with smaller volume, I suppose 150-170Lt) !!!


(and sell to me your E140 for my 4560 horn setup!! ehehe Just kidding).

[Dr.db]

I think the TAD TL-1603 is the way to go.

The 1601b shares the same cone but has a huge Alnico motor... Interesting, but I´m not shure if these are modern designs that would withstand high currents e.g. heat like a JBL 1500AL. Or if they will demagnetize like most Alnico´s from the past...

[Ruediger]

@Ruediger:

I have tried these 3,3ohms resistors just in front of my speaker inputs.

<snip>
The sound was different, but didn´t really cut it

I´m still missing lows so I will be changing the E-145 to something else....

I had no working internet for a couple of days. So this reply is a bit late.


The resistor just changes the driver's Qt (increases it) so that you can find an alignment which matches your boxes. Part of the alignment is a properly tuned port.


You may remember these replies:


<reply #1>
Thiele is good for you. Thiele is very good for you!

Try to get acquainted with the Thiele papers. Start with table i.


The natural alignment for your speaker would be alignment #3. That would give you a horrible f3.


Alignment #5 is tempting. That would give you a pretty f3 and would be compatible with your box size. But your Qt is too small. Look at equation 70 how to make Qt larger. Use reasonable cables and reasonable inductors and Qt will get larger. You need to play around a bit with your pocket calculator.


Your box is still too large? Make the port cross section larger, this is good anyways. It requires you to increase the port length. You can have a port with a small cross section and a small length, or one with a large cross section and a large length (for the same result, box resonant frequency). Eat up the excess volume with a bigger port.


The resulting alignment can be plotted and optimized by some friendly forum members.


First calculate, then build a prototype from chipboard, then modify your beauties.


Ruediger
</end of reply #1>


<reply #2>
A worked example

Below is a worked example. It does not fix the problem completely but it can shift parameters in the right direction. It also helps in making educated guesses.


E145 data:
fs = 35 Hz
Qt = 0.25
Qm = 6
Qe = 0.26
Qa is the same as Qm
Vas = 275 ltr
Re = 5.7 Ohm
Re is the DC resistance of the voice coil


Rg is the "Generator resistance". It consists of the amplifier's output impedance, the cables' resistance and the DC resistance of all bass coils in the xover which are in series with the loudspeaker.


The "damping factor" for a certain load is the ratio between load and the amp's output impedance. A damping factor of 200 at 8 Ohm means that the amp's output impedance is 8 / 200 Ohm = 0.04 Ohm.


The specific resistance of copper is 0.015 Ohm * mm**2 / m. Lets assume a cross section of 2.5 mm**2 and a length of 5 m (times 2, back and forth).


Rcable = 0.015 Ohm * mm**2 / m * 10 m / 2.5 mm**2 = 0.015 * 10 / 2.5 Ohm = 0.06 Ohm.


A typical bass coil (inductor) for an 8 Ohm / 1 kHz crossover 2nd order has an inductance of 1.8 mH and a DC resistance of 0.5 Ohm.


Our "Generator resistance" is 0.04 Ohm (amplifier) + 0.06 Ohm (cables) + 0.5 Ohm (crossover) = 0.6 Ohm. You can try 3rd order crossovers as they need more and larger coils.


We put this into equation 70 of the Thiele paper and get a new Qt of 0.274. With this value we can interpolate between alignments 3 and 4.


For a spot landing at a Qt of 0.383 we would need a generator resistance of 3.27 Ohms. A tube amplifier with output transformers may also help you.


You can experiment with large ceramics resistors at your speakers input terminals to get a clue what the impact would be.


Ruediger
</end of reply #2>


With the 3.3 Ohms resistors you can make a spot landing. f3 == fs, fb == fs, fs == 35 Hz. Vas / Vab = 1.414, thus Vb = 194 liters which is 6,9 cuft.


You don't need to find an exactly matching Thiele alignment. There are approximate methods.


You need to modify your boxes' tuning such that fb == fs == 35 Hz.


What is the net volume of your boxes, and what are the dimensions of your port tubes?


Ruediger

[Dr.db]

The internal net volume is about 8 cubic feet.

The ports are 4" in diameter and 3,5" in length....

[Ruediger]

E145H

Dr. dB, if the data you gave me is correct, and if the E145's parameters are correct, then you should be quite close to a proper alignment with an f3 in the 35-38 Hz region. That would be excellent.

Some data is approximate (8 cuft) or not equal to the required value (6.9 cuft), the port tubes leave room for discussion, the E145 Vas parameter could be incorrect, and finally:

You might be expecting too much bass

Below I show the calculations so that you can make them yourself, and apply them to other projects.

Box resonant frequency:

fb = c / (2 * pi) * sqrt(A / [Vb * L])

c = speed of sound, ca. 344 m / sec @ room temperature
A = vent area
Vb = box volume
L = effective vent length (Leff below)

The vent can consist of one or more tubes of equal length.

If the vent shape is not circular an "effective radius" Reff is calculated by considering a circular vent with equal area:

A = r**2 * pi -> r = sqrt(A / pi) (Reff below)

The "effective length" Leff of a vent tube is its actual length L plus inner end correction Li plus outer end correction Lo:

Leff = L + Li + Lo

A free standing port has an end correction of 0.85 * Reff,
a flanged port has an end correction of 0.61 * Reff, so for a vent which is free standing on the inside and flanged on the outside the calculation is:

Leff = L + 0.85 * Reff + 0.61 * Reff = L + 1.46 * Reff.

These values are approximate.

If the vent consists of several tubes (2 in your case) then there are two cases:

When the ports are close to each other the calculation is that for one vent with twice the vent area. Replace "twice" with "number of ports".

When the ports are not close to each other then the end correction must be considered separately for each port.

Your partially rounded port is not easy to calculate, and it is neither really flanged nor really free standing at the inside. The openings are neither really close to each other, nor really apart from each other.

My guesstimate is: two ports with end correction = 1.5. Simple mixed-case formula from above: end correction = 1.46. Not a big difference.

Now the calculations.

Vb = 8 cuft = 8 * 28.3 liter = 226 liter

Two ports with
diameter = 4" = 10 cm
length = 3.5 = 8.9 cm

Total port area A = d**2 * pi / 4 * 2 = 78.5 cm**2 * 2 = 157 cm**2

Effective radius (ports combined)
Reff = sqrt(157 cm**2 / pi) = 7.1 cm.

Combined ports: effective Radius = 7.1 cm,
7.1 cm * 1.5 + 8.9 cm = 19.55 cm

Separate ports: Radius = 5 cm,
5 cm * 1.5 + 8.9 cm = 16.4 cm

The difference is 3.15 cm. Build with temporary ports, measure and adjust. Better: avoid such uncertainties in the 1st place.

Box resonant frequency:

fb = c / (2*pi) * sqrt(A / [Vb * L]) =

(344*100cm/sec) / (2*pi) * sqrt(157cm**2/[226*1000 cm**3 * 16.4cm])

"cm" can be cancelled completely, the remaining unit is "1/sec" ->

1/sec * (344*100) / (2*pi) * sqrt(157 / [226*1000 * 16.4]) =
35.6/sec = 35.6Hz

This would be a perfect match for Thiele alignment #5 if Vb was right: Vas / Vb = 1.414 -> Vb = Vas / 1.414 = 194 liter = 6.9 cuft. Let's calculate with proper Vb:

1/sec * (344*100) / (2*pi) * sqrt(157 / [194*1000 * 16.4]) =
38.4/sec = 38.4Hz

This would be a perfect match for Thiele alignment #5 if fb was right.

Summary: If the published E145 data Vas is correct the calculations look promising. I attach a flow chart about finding an approximate solution.

Ruediger

[AudioFan]

Ruediger,

Excellent work from you as usual. Dr. db was also looking at the JBL 2235H as a possible substitute in this enclosure. With the same enclosure volume, what would the 2235H do in this box with optimized ports for same? I have built several 2235H based systems that worked out very well in the past as dedicated subs or as the LF component of a multi-way speaker in a similar sized encloure.

Thanks for your hard work.

AudioFan

[Ruediger]

Ruediger,

Excellent work from you as usual. Dr. db was also looking at the JBL 2235H as a possible substitute in this enclosure. With the same enclosure volume, what would the 2235H do in this box with optimized ports for same? I have built several 2235H based systems that worked out very well in the past as dedicated subs or as the LF component of a multi-way speaker in a similar sized encloure.

Thanks for your hard work.

AudioFan

Hi AudioFan,

the box is much too large for the 2235H. Below I attach a few notes about what can be done with the 2235H.

But why settle with just one bass speaker per box? Why just shake the house when you can rattle the foundation?

With Thiele alignment #3 the E145H will have an f3 of 62 Hz in a box of 62 liters without any Qt-changing resistors. Qt will remain 0.25.

Dr. db's box has a volume of ca. 226 liters. Minus 62 liters and some more for the woodwork to accomodate the E145H.

That should leave ca. 130 liters for a 2235. See the attached notes for an approximate design (36 Hz, 129 liters).

With some fiddling it should be possible to get the two drivers into the boxes. One could gain a few liters by mounting the 2235H reversed, with the magnet at the outside. That would improve the cooling too.

Should somebody go this way he would need to design the crossover 1st, in this scenario the E145H behaves like a 62 Hz highpass of fourth order and should be complemented by a properly designed lowpass for the 2235H.

Alignments for the JBL 2235H 15" Bass Loudspeaker

JBL 2235H

fs = 20 Hz
Re = 6.0 Ohm
Qt = 0.25
Qm = 2.5
Qe = 0.28
Vas = 460 Liter

Thiele Alignment #3, QB3
Qt = 0.259
f3 = fs * 1.77 = 20 Hz * 1.77 = 35.4 Hz
fb = f3 / 1.25 = 35.4 Hz / 1.25 = 28.3 Hz
Vb = Vas / 4.46 = 460 L / 4.46 = 103 L

Thiele Alignment #15, B6 (requires an active filter)
Qt = 0.299 must be made fit (see below)
f3 = fs = 20 Hz
fb = f3 = 20 Hz
Vb = Vas / 2.73 = 460 L / 2.73 = 168 L

Keele's Method (see flow chart, f3 and Vb can be adjusted)
Qt = 0.25 must be effective value
f3 = 36 Hz
fb = 29 Hz
Vb = 129 L

Notes
Qm and Qa are the same
Vab and Vb are the same

Qt (total Q) can be made larger if required
Qt can be calculated from:
Qa acoustical Q (also mechanical Q)
Qe electrical Q
Re DC resistance of loudspeaker voice coil
Rg Generator resistance = Output impedance of amplifier plus
loudspeaker cable resistance plus DC resistance of
inductors which are in series with the loudspeaker

1/Qt = 1/Qa + (1/Qe) * [ Re/(Rg + Re) ]

[AudioFan]

I read this article many years ago and built several subs based upon the information it included, (http://www.lansingheritage.org/image...subs/page7.jpg). I built 12cf 2245H, 8cf 2245H, 5cf, 6cf 2235H. I also read about this sub: Certified Bass for the Certifiable by James S, Sherwin in Audio Magazine around 1990. Link on site: (http://www.audioheritage.org/vbullet...oning-for-subs). I built that dual 2235H sub unit as well. I also built one with 2-2235H drivers as pictures in one of my earlier posts to this thread. It can be seen below a center channel system composed of all the drivers from a pair of JBL L7 systems. They all seemed to work fairly well with the JBL B63A crossover inserted in front of the sub amp. I tried everything from Phase Linear 400 and 700B to UREI 6290 and Ashly MFA6000 and MFA8000. However, they are so efficient that you really don't need a ton of power to reproduce very high levels of LF output. I think that a bridged UREI 6260 is adequate to power either one of these subs.

AudioFan

[Dr.db]

Thanks a lot for your extensive explanations!

I have to admit, these calculations are a little confusing to say the least... But I believe they are absolutly right.
What I have actually done was copying the JBL Everest DD55000 enclosure.
That box is a proven design so I thought I´d just stick with that... If you have a look at the puplished freq.resp. it indicates a weak bass. A 4344 is much more linear in the bass than the DD55000...

At this point I have to say I´m absolutly not a bass-addict that aims a boomy response. This is what I usually dislike with most commercial speakers... But with the E-145 alone the bass is too thin and lacking body in the music.

Using the E-145 in a small subchamber and mate it with a 2235 was an idea I was allready thinking about. But at the end I would say they are still too close and not different enough to run in the same enclosure.
This is why I thought about using a 2235 with a 10" midrange instead...

[Dr.db]

I was very lucky to find a pair of barely used TAD woofers locally!

In around a week I while have time to swap them in and compare them to the E-145. I will reduce the internal volume to about 6 cubic feet e.g. 170liters. This is the enclosure size I could read about in other threads so far. I guess the two ports should be about 8" long to obtain a 35hz tuning....
What do you guys think about enclosure-size and tuning?

[JeffW]

A tribute to Lansing speakers using TAD drivers....

[Dr.db]

A tribute to Lansing speakers using TAD drivers....

A tribute to Lansing-Heritage-Forum, not speakers....

Please read the title and my first post accurately as this thread is a tribute to all forum-members that have helped me over the years on this board.
JBL choosed to stop selling parts to the DIY-community and if you want to purchase a current model you have to go with other brands. Otherwise I would have bought the 1500AL, but no way to buy them anywere...

[grumpy]

Even so, there is a Lansing heritage element in many TAD products 😎

[JeffW]

I was mainly joking, shoulda put a smilie in there.

I don't really care what drivers folks use, and I have a few TAD myself.

[Dr.db]

I was mainly joking, shoulda put a smilie in there.

I don't really care what drivers folks use, and I have a few TAD myself.

No harm meant