Altec horn driver resonance question
geocla
This weekend a friend brought over two Altec horns to measure - an 811B
with 806 driver and a Mantaray with its associated driver. Measuring
the frequency responce using an IMP system showed a response usable
between approx. 1kHz and 6 Khz for both horns. The problem is that
looking a the impedance curves shows a resonance at 1.5 to 2.5 kHZ on
the 811 and two resonce peaks one at approx 700 Hz and one at 2.5kHz on
the Mantaray. As both these horns are used in systems that crossover
below these resonance points the question is how are these resonaces
handled in the commercial systems (if they are) and are these horns
usable for faithful reproduction at a 1 kHz crossover which is beow
their resonance peaks. How will the resonance effect the sound? Thanks
for any help.
George
Greg J Szekeres
I can't picture the models, but some horns with to narrow a mouth
can produce these problems.
greg
Randall Bradley
The 811 is a very different horn than a Mantaray.
You didn't say what model or size (or vintage) Mantaray
you have there. Presumably a small one, since it appears
that the small format driver fit onto it.
The Mantaray is designed to provide the "elimination"
of so-called "waistbanding" which is a pinching of the
polar response pattern at certain freqs. To do this they
designed a horn that consists of different flares and
therefore frequency "breakover" points in the *vertical*
and *horizontal* axis. Plus the horn is made of sequential
conical sections rather than a continuous curve like
an exponential or hyperbolic/exponential or tractrix.
What this appears to do is to load the driver somewhat
differently than does a "standard" horn of good design.
Thus, I suspect, the cause of the two impedance peaks
at different freqs.
Also, although they don't say so explicitly, it appears
to me that the horn is actually "fed" via a slot radiator
formed by the "throat" section. Thus the "slot" provides
very uniform radiation in one plane. (let's say, left to
right) The JBL "Bi-Radial" drivers use a somewhat similar
trick, although a slightly different throat geometry -
still ending in a slot.
Hope that sheds a bit of light...
_-_-randy
BEAR Labs
Mark Pindler
gfre...@northnet.org wrote:
> > This weekend a friend brought over two Altec horns to measure - an 811B
> > with 806 driver and a Mantaray with its associated driver. Measuring
> > the frequency responce using an IMP system showed a response usable
> > between approx. 1kHz and 6 Khz for both horns.
>
> Actually the response is usable to 20 kHz if you are willing to use
> equalization. The rolloff on a 511B horn with a 808-8A shows a 6 dB/ octave
> slope starting at 5 kHz and increases to 8 dB/ octave from 10 k to 20 k (
> Altec specified these drivers as 500 - 20 kHz ). The 806 was just a lower
> power version of the 808 with similar frequency response curves. The 811B
> has a low frequency limit of 800 Hz so your going to see a rolloff on the low
> end due to the decreasing Q of the horn ( the vertical radiation angle
> increases ). The only Mantary horn made for these drivers ( 7/8" throat ), as
> far as I know, was asymmetrical. The recommended crossover for that horn is
> 1500 Hz and the actual low frequency limit is 1200 Hz.
>
> > The problem is that
> > looking a the impedance curves shows a resonance at 1.5 to 2.5 kHZ on
> > the 811 and two resonce peaks one at approx 700 Hz and one at 2.5kHz on
> > the Mantaray.
>
> What is the magnitude of the impedance peaks? I just quick checked a
> 808 driver and found nothing of significance, I know it is not the same
> driver but the 806 was very similar. Why would you think this is a problem?
>
> As both these horns are used in systems that crossover
> > below these resonance points the question is how are these resonaces
> > handled in the commercial systems (if they are) and are these horns
> > usable for faithful reproduction at a 1 kHz crossover which is beow
> > their resonance peaks.
>
> If the Mantaray horn is the one I suspect then the crossover frequency
> is too low. The 2.5 kHz might be a driver resonance, try removing a driver
> from the horn and measuring the impedance again. The drivers are usable to
> 500 Hz with the cutoff frequency of the horns being the limiting factor in
> your case. The impedance curves really don't tell you anything about the
> acoustic output, I assume that you removed the crossovers before doing your
> measurements.
>
> > How will the resonance effect the sound?
>
> Got Me.
>
> Greg
>
> -----== Posted via Deja News, The Leader in Internet Discussion ==-----
> http://www.dejanews.com/ Now offering spam-free web-based newsreading
I found that most of the diaphram failures in these horns were fractures due to
low frequency info getting to them at high output levels. The poor things just
rip or shatter while the voice coil is just fine as the systems had enough clean
power to drive them into oblivion. Remember the excursion limits of the
diaphrams. There is alomost none! To move air at that frequency and at that spl
level creates a tremendous pressure wave for the diaphram to deal with, and, it
can't do it for long! It is safe to use a driver at 2 times the resonance
frequency and up to avoid destructive modes of excurtion. Also, only measure the
driver mounted on the horn to be used, they are a symbiotic unit. Tests apart are
useless for real world apps.
Mark Pindler
gfre...@northnet.org
geocla
Mark Pindler wrote:
> low frequency info getting to them at high output levels. The poor things just
> rip or shatter while the voice coil is just fine as the systems had enough clean
> power to drive them into oblivion. Remember the excursion limits of the
> diaphrams. There is alomost none! To move air at that frequency and at that spl
> level creates a tremendous pressure wave for the diaphram to deal with, and, it
> can't do it for long! It is safe to use a driver at 2 times the resonance
> frequency and up to avoid destructive modes of excurtion. Also, only measure the
> driver mounted on the horn to be used, they are a symbiotic unit. Tests apart are
> useless for real world apps.
> Mark Pindler
mounted. The magnitude of the impedance was about 40 ohms vs 8 ohms
nominal. The horns were used in a two way system crossed over at
12db/octive at 800hz for the 811 and 1200Hz for the Mantaray. Could
this have damaged them? My friend would like to use one of these horns
in a new two way. the Mantaray freq response was very flat (within 2db)
from about 1Khz to 6Khz were the roll off begins. I guess we'll build
one up and see how it performs. Thanks again
George
gfre...@northnet.org
> > I found that most of the diaphram failures in these horns were
> fractures due to low frequency info getting to them at high output levels.
>>The poor things just rip or shatter while the voice coil is just fine as
>>the systems had enough clean
The only ones I ever saw that shattered or damaged diaphragms were in bi-amp
systems that didn't have a protection cap in series with the horn. The
protection caps are to protect against low frequency damage from amplifier
turn on or power line transients. Altec warranted all their drivers for life,
the A-7 voice of the theater and its many derivatives, with the radial horn,
all crossed over at 500 or 800 Hz. Altecs power test was band limited pink
noise 500 - 5000 Hz at full power for 24 hours, this is far more demanding
than the typical sine wave power ratings.
>>It is safe to use a driver at 2 times the resonance frequency and up to
>>avoid destructive modes of excurtion.
This rule of thumb doesn't apply to these drivers.
>>Also, only measure the driver mounted on the horn to be used,
It is quite safe to do impedance measurements without the horn, just
don't do it at full power.
> Thanks all for the info. I measured the drivers with the horns
> mounted. The magnitude of the impedance was about 40 ohms vs 8 ohms
> nominal.
Wow! I measured a 808 driver, 500 - 1200 Hz was between 7.8 to 10 ohms,
rises to 14 ohms at 1400Hz and 20 ohms at 2 KHz. Rolled of to 8 ohms at 3
kHz and stayed between 8 - 10 ohms up to 10kHz ( that's where I stopped ).
> The horns were used in a two way system crossed over at
> 12db/octive at 800hz for the 811 and 1200Hz for the Mantaray. Could
> this have damaged them?
Not a chance! I used the radial horn ( crossed over at 800 hz ) and the
Mantaray ( 1500 hz ) for live sound for years and at far higher power levels
then you will ever encounter in a domestic environment. The pressure
sensitivity is about 108 dB @ 1 watt - 1 meter, you will suffer serious
hearing damage before you even come close to driving them to full power.
André Huisman
Mark Pindler heeft geschreven in bericht <3574DD84...@earthlink.net>...
> I found that most of the diaphram failures in these horns were fractures
due to
>low frequency info getting to them at high output levels. The poor things
just
>rip or shatter while the voice coil is just fine as the systems had enough
clean
>power to drive them into oblivion. Remember the excursion limits of the
>diaphrams. There is alomost none! To move air at that frequency and at that
spl
>level creates a tremendous pressure wave for the diaphram to deal with, and,
it
>can't do it for long! It is safe to use a driver at 2 times the resonance
>frequency and up to avoid destructive modes of excurtion. Also, only measure
the
>driver mounted on the horn to be used, they are a symbiotic unit. Tests apart
are
>useless for real world apps.
I don't see the pressure wave story. I always suspected the diaphragm to
shatter because it banged itself to the phase plug. This is always due to low
X-over freq., low order of X-over, in any case because there is just too much
LF content in the signal. BTW, you would have to be deaf not to hear the unit
shred itself to pieces. The sound coming from that isn't quite pleasant to the
ears.
If you use a Compression driver in an active setup, don't forget to use an
extra capacitor to block on/off transients from the amplifier. (Many people do
forget). Just make sure the capacitor does not interfere with the X-over
slopes too much.
André Huisman
New Line licht & geluid
XXXne...@usa.netXXX
http://home.worldonline.nl/~newline/
remove X-es to reply
--- pardon my French, I'm Dutch ---
SGuttag
>The 806 was just a lower
>power version of the 808 with similar frequency response curves.
You got that backwards. The 808 is a high-power version of the 806 (802).
>The rolloff on a 511B horn with a 808-8A shows a 6 dB/ octave
>slope starting at 5 kHz and increases to 8 dB/ octave from 10 k to 20 k (
>Altec specified these drivers as 500 - 20 kHz
Altec never specified the 808 with a 20KHz top end. The 808 (a Symbiotik) was
a trade off of top end for power handling...I believe that 15KHz was its top
rating. The 808 evolved into the 908 and has subsequently been replaced by the
909 (Pascalite) which has the high-end response of the 902 (802, 806) with the
power handling of the 908 (808).
>The only Mantary horn made for these drivers ( 7/8" throat ), as
>far as I know, was asymmetrical. The recommended crossover for that horn is
>1500 Hz and the actual low frequency limit is 1200 Hz.
Not true at all. There are several Mantaray horns made for the 800/900 series
of drivers. Some as raw parts and others as part of a speaker system. The
most popular Mantaray horn for the 800/900 series driver is the MR994. It has
a 500Hz lower limit (like the 511). The MR994 is a 1" version of the MRII594A
which is used with the 288/299 drivers (1.4"). These horns are wonderful
sounding, by the way.
SG
gfre...@northnet.org
> >The 806 was just a lower
> >power version of the 808 with similar frequency response curves.
>
> You got that backwards. The 808 is a high-power version of the 806 (802).
Um.............isn't that what I said, glass is half empty or glass is half
full.
> >The rolloff on a 511B horn with a 808-8A shows a 6 dB/ octave
> >slope starting at 5 kHz and increases to 8 dB/ octave from 10 k to 20 k (
> >Altec specified these drivers as 500 - 20 kHz
>
> Altec never specified the 808 with a 20KHz top end.
They most certainly did and I still have several pieces of Altec literature
that confirm that. Technical letter # 230 " Application of Component
Loudspeaker Power Capacity Ratings To The Design Of Two Way Systems " by Mark
Engebretson. It lists all their hi frequency drivers including the 806, 807,
and 808. All three are listed 500 - 20,000 Hz. Also from the " General
Services Administration Federal Supply Service " catalog, contract:
GS-00S-07116, which is the catalog the Federal Government used. The 802 and
808 drivers are listed and specified as 500 - 20,000 Hz as rated frequency
response. Under the specifications for the " Altec 1204B " the frequency
response is 35 - 20 kHz and the high frequency driver is specified as the
808, this data sheet includes a graph of the frequency response. I also have
the spec sheet for the 808 driver and is specified for 500 - 20 kHz mounted
on the 511B and 800 - 20 kHz
> The 808 (a Symbiotik) was a trade off of top end for power handling...
The 808 and 807 were more efficient by 3 dB compared to the 802, power
handling was 3 times that of the 802 and 806 . I apologize for the
discrepancies as the specifications for the various drivers is scattered
through the literature, I do not have a " one sheet " spec for the 802, 806 or
807 drivers and have had to piece them together.
>The 808 evolv>
<input type=
SGuttag
>> Altec never specified the 808 with a 20KHz top end.
>
> They most certainly did and I still have several pieces of Altec literature
>that confirm that. T
I have indeed found Altec literture that does have the 808 speced out to
20KHz...I apologize for my incorrect statement...
The 808 did have a much more substantial high-end rolloff compared to the other
800 series drivers. Its main purpose was higher power at the expense of
high-frequency. This carried over to the 900 series (902 vs 908). It wasn't
until the 909 came out that one could have their cake and eat it too
SG.
gfre...@northnet.org
> I have indeed found Altec literture that does have the 808 speced out to
> 20KHz...I apologize for my incorrect statement...
Thank you, this is a rare phenomena and should be commended.
>
> The 808 did have a much more substantial high-end rolloff compared to the
> other 800 series drivers.
If I have any frequency response plots of the 802/806 drivers I haven't
been able to locate them.
> Its main purpose was higher power at the expense of high-frequency.
This make sense, the high frequency response is primarily determined by
the mass of the diaphragm and the diaphragm to phase plug spacing. The mass
will determine the first pole in the response, not knowing the relative mass
differences between aluminum and symbiotik it is hard to comment. The phase
plug spacing produces another pole ( kind of ) at a higher frequency and would
have to be greater for the 808 to compensate for the larger excursions. There
is definitely a trade off here.
Greg
tom brennan
All high performence compression drivers start rolling off power
response in the 4-6khz region, the use of a horn with collapsing pattern
control as frequency rises can give extended response <on axis>, this
why CD horns need EQ.