Slot Loaded Open Back / Slot Loaded Push Pull


This project was stimulated by a friend, prior to which I had but a vague awareness of its existence.
Several years ago, a friend gave me a book titled, Question Everything which was true of my nature which has its root when I was a kid. My dad taught me how to think, not what to think.

I take little as so-called gospel unless, of course, the source is known to be credible. While i wouldn't place that friend as credible in such a technical matter, I did accept the credibility of the author but my curiosity got the better part of me, beside which, I needed another project to occupy my time.

I did find out that the author was right, however slightly exaggerated that claim may appear. The only way to find out is an A-B blind test but in all my years in audio, I found only one who would consent to such a test. We did agree that he may not like the result and I was right, he didn't. That which he thought would be preferred was opposite to that which he thought.
The test involved claims that a kevlar cone was better than a paper cone. I will state now that that statement is true from a purely technical point of view. However, technical analyses have little to noting to do with personal preferences.

The test was done in the mid eighties and involved a pair of 10 inch woofers, one a $20 paper cone and the other a $110 kevlar cone. Two visually identical enclosures were designed and built. Each had slightly different internal volumes for optimum performance in the bass according to BassBoxPro, v6.Once done, the friend came over for the audition. the speakers were placed against the wall opposite the listening position, some 16 feet distant. The room measures 20ft by 17.5ft. The speakers were placed behind a sheer black sheet and the room very dimly lit. this disallowed seeing the yellow kevlar cone. The speakers were placed side by side to eliminate differences in room reflections. A silent switch was used to switch between the speakers. In was unknown which was played first and whether or not a change was made. All I did was ask him to say if he preferred the sound that was playing and I made note thereof.

After about half an hour and several different types of music, the result was tabulated and much to his surprise, he preferred the sound of the $20 speaker's bass more than 90% of the time. See, we get used to things and a change is usually rejected because we don't like to think we were wrong. In this case, he was influenced by what is called bias expectation. He had never heard the kevlar speaker but was told that it was better and believed that it would sound better. Again, depending on one's preference, that may be true but for the most of us, it isn't. A person familiar with how a live performance sounds,

Regarding the SLOB/SLPP design, no audition was performed as my only desire was to determine how much the second harmonic distortion was reduced. I don't doubt that some can hear it but I doubt that I can at my age as the reduction is subtle, percentage wise.

I remember Paul Klipsch stating in one of his papers that an attendant at one of his demonstrations repeatedly demonstrated his ability to differentiate the sound between a Marantz Model 9 and a McIntosh Mc275.

Anyway, my apologies for the diatribe.




The chamber.  It measures 11"x9.625"x1.5" for a volume of 189 cubic inches.  This includes the added volume (12 in^3) of the slot in the yet to be attached front baffle and the combined volumes of the frusta of the two speaker cones. (18 in^3)

The slot area is 16.5 in^2

The larger chamber including the frusta and the slot volume is 274 in^3



The front baffle attached.  It serves little enhancement of the low frequencies as it's longer dimension, 25" will support frequencies above about 500hz.  This size was chosen because it fits the 3 ft^3 enclosure upon which it sits and can be used for future tests.



The assembly mounted on top of another enclosure to get the center of the slot midway between the floor and the ceiling, 48 inches.  the CLIO is internally calibrated to cancel first reflection from the floor and ceiling with the mic at a distance of 1m from the source.

The LMS allows adjustment of these distances and once entered, it will turn off the signal before the first reflection gets to the mic, thus simulating an anechoic response.  Both units will fail below about 400hz as the wavelength is too long to get a good sample before the reflection arrives.



This was a later experiment to see what happens if the chamber volume were increased by 80in^2 by increasing the chamber width by 0.75 inch.  There were two tests done here.  One with the wider slot and another, shown here, with the slot reduced to the original size and both with the larger chamber.  The results were unworthy of consideration and therefore, not reported here. (see fig 2)

It seems the chosen volume may be close to optimum.  Consideration was given to making the chamber smaller but that has yet to be studied.  The main purpose of this experiment was to determine if second harmonic distortion is reduced, as claimed.

It wasn't.  In fact, it got worse. (see figs 5 and 7)



The speakers in a sealed box of 3 ft^3.  Second harmonic distortion was lower than that of the slot system and, obviously, the lower bass was enhanced due to the blocking of the rear wave. 



The two eights on an open back baffle,




Z   BLK=2 in parallel in SLOT Chamber; RED=2 in parallel in 3 cu ft sealed

At first glance, it may seem counterintuitive that fs would be lower with the pair facing each other in a small chamber. The rear of the speakers is radiating into open space, unlike the pair in the closed box.  Despite the fronts of the speakers radiating into a small space, that space is open at the slot and while it may impose a heavier load on the speakers, the speakers may see it as additional mass, hence the lower resonant point.  The open back imposes a far less load than the 3 ft^3 sealed box thus resembling an open air condition.

The open air resonant point of these speakers is 32hz and in the SLOB, that resonant point has lowered to about 29hz, very likely due to the effect of the air load in the slotted chamber being pushed by one speaker against the other

The "blips" at about 170hz and 340hz are nor floor bounces as both speakers were measured in the same location and under the same conditions. They may be due to reflections from the back wall as they only show during the open back sweep.

See the note at the bottom of this page.




big box is 79 in^3 larger and its slot is area is 24.75 in^2  Its slot volume adds 18 in^3 to the chamber.

RED=small box small slot  ORN=2nd harmonic

GRN=big box big slot  BLU=2nd harmonic

BLK=big box small slot  GREY=2nd harmonic

It seems that there is little effect with the above combinations although the small box with the small slot has less 2nd harmonic distortion below 230hz



1w1m small slot and chamber RED=THD

 GREEN=2nd harmonic; ORANGE=3rd harmonic

This shows that the total harmonic distortion is mostly 2nd harmonic

Fig 4 shows the same comparison at 10 watts





GREY=small slot and chamber RED=THD

 GREEN=2nd harmonic; ORANGE=3rd harmonic

The major result of the 10 watts is an increase in response level by 10dB, which was expected..  What wasn't expected was the distortion components not increasing more than 10dB.




GREY=3 cu ft sealed YEL=2nd harmonic

RED small slot and chamber ORN=2nd harmonic

Comparison between the two eights in a closed box(GREY) with its 2nd harmonic(YELLOW) and the SLOB(RED) with its 2nd harmonic(ORANGE)

In their passbands below 500hz, the closed box has much less 2nd harmonic distortion than the SLOB, contrary to claims of the opposite.

Keep in mind that the response below 150hz or so is affected by the room hence the higher bass output of the closed box. 






OPEN BACK GRN=1w1m Blue=2nd harmonic BLK=10w1m GREY=2nd harmonic

Usually, the 2nd harmonic curves will be very similar but spaced 10dB apart.  Here, the two curves differ in shape, between about 70hz and 200hz.  Also, the difference in SPL far exceeds the expected 10dB; it's more like 20dB. This is attributed to the speakers' cones exceeding Xmax.  This was not only observed but also heard.

For obvious reasons, the sweep was not repeated a second time.




1w1m DARKBLUE=OB LTBLUE=2nd harmonic BLK=SMLCHAMBERslot GREY=2nd harmonic ORN=SEALEDBOX YEL=2nd harmonic

The dark blue(OB) and orange(SB) upper traces show the increased bass with a sealed box.  The lower traces show that the OB has more 2nd harmonic distortion.

The upper black trace is that of the SLOB.  It shows less distortion below about 150hz and 3 peaks at about 170hz, 340hz and another just past 500hz.

In short, 2nd harmonic distortion is reduced with the SLOB/SLPP configuration but this comes at the cost of reduced bass output, typical of an open back system.  As can be seen by the dark blue and black traces, the SLOP/SLPP does have an improved bass output over the traditional open back.

It can be humorously said that to further reduce the distortion, reduce the bass.  Taken to an extreme, low frequency distortion can be brought to zero by eliminating the bass altogether.




The note at the bottom of this page (FIGURE 8A & 8B)

Air has a mass of 1.12kg/m^3 or 0.0184g/in^3.  The 189 in^3 chamber contains 3.469g of air, which is 1.734g/speaker.  This increase in mass attached to the cone will show a lower resonant point in an impedance sweep. 

LMS was used here because it has certain features that the Pocket CLIO doesn't have.  The more sophisticated CLIO probably does but it costs about $3000.

The LMS will allow a start and stop point to be selected with as much as 800 data points.  This sample was swept between 20hz and 70hz with 300 data samples.  This allows a much more accurate point of resonance to be located using the cursor (not shown), which can move from one data point to the next using the left and right arrow keys.  Here, we have a resonant point at 33.345hz with an impedance peak of 48.206W (BLACK) with 1.7g added to the cone.  The PINK trace is the impedance without the added mass and peaks at 33.736hz with an impedance peak of 49.169W  The GREEN trace is the impedance with 3.4g added and has a resonance at 32.245hz with an impedance of 47.875W

What has all this to do with second harmonic distortion?  Probably little to nothing but my curiosity had to be satisfied, despite the veracity of the above.  If nothing else, it was better than nothing, a shot in the dark.

The LMS doesn't perform distortion and spectrum analyses; CLIO does.


An enlargement of the above.


Back to the loudspeaker main page

Table of Contents