jamesgangnc wrote:
> "bear" <bearlabs@[EMAIL PROTECTED]
> wrote in message
> news:fqi18a0aiv@[EMAIL PROTECTED]
>> jamesgangnc wrote:
>>> "bear" <bearlabs@[EMAIL PROTECTED]
> wrote in message
>>> news:fqc0ig01jv6@[EMAIL PROTECTED]
>>>> jamesgangnc wrote:
>>>>> I updated my setup to bi-amp last year. I was wondering if any
others
>>>>> in the group is doing this now and what your set up and experiences
>>>>> were?
>>>>>
>>>>> I'm using a dbx 223 crossover usually at 700hz. I play around with
it
>>>>> a bit but keep it between 500 and 1k. The lower end I'm running to
an
>>>>> adcom 555 connected to 4 ten inch moderately priced woofers in
closed
>>>>> cabinets, two woofers to a side wired in parallel. The high side I
>>>>> have going to an adcom 535 with a pair of 5 1/4 midwoofers and one
>>>>> ribbon tweeter on each side also in closed cabinets. The passive
>>>>> crossover on the high side is just a simple 6db centered at 5k. Of
>>>>> course it is subjective, but I feel like moving the bass out of the
>>>>> rest of the signal improved overall clarity. The two adcoms, when I
>>>>> experimented with them independently also seemed better suited to
the
>>>>> separate tasks. Besides the obvious power difference the 535 seemed
>>>>> to have a better sound at mid and highs than the 555 when listening
to
>>>>> them with a full signal range at about the same sound levels. Of
>>>>> course the 555 is obviously the choice for the low end anyway due to
>>>>> the power difference.
>>>>>
>>>>> For a signal source I'm using a technics sh-ac500d as a preamp and
>>>>> either xm radio or itunes acc ripped at 256k via a squeezebox. The
>>>>> setup doubles as part of my surround sound as well hence the ac500d.
>>>>> I mostly listen to old and new rock but occasionally anything else
as
>>>>> well, except country. I'd like to take some measurements as well.
>>>>> I've got a signal generator and a scope but don't have a good high
>>>>> quality mic.
>>>>>
>>>>> As I suspect most of us are, I'm trying to decide what to change
next
>>>>> as well :-)
>>>> I'd get a higher order filter on your ribbons first.
>>>>
>>>> The distortion increases in typical ribbons when the excursion below
the
>>>> useful rolloff does not decrease substantially. A first order filter
in
>>>> practical terms tends to make the excursion *remain the same* below
the
>>>> inflection point of the "rolloff". That means that you've got excess
>>>> deflection/excursion. I'd consider a 24db/oct filter there... most
>>>> ribbons
>>>> can not handle a first order filter.
>>>>
>>>> Then use one of the various freeware FFT programs and a halfway
decent
>>>> mic to measure the freq response of the mid/HF combo. Take note of
the
>>>> off axis response and the waterfall as well. Look at the impulse
>>>> response of the two drivers together.
>>>>
>>>> You may need to adjust the xovers and do other things to arrive at a
>>>> best compromise response for the mid/tweeter combo.
>>>>
>>>> Then you can look at the relation****p between the woofs and the upper
>>>> range.
>>>>
>>>> After that I'd say that with the "average" sort of ripped music, it
>>>> probably won't make much difference what else you do. BUT, once you
get
>>>> the speakers optimized you may start to hear larger differences in
terms
>>>> of source material - then it may make sense to think about upgrading
>>>> other
>>>> parts of your system.
>>>>
>>>> Btw, the difference you are now hearing is likely due to two things:
>>>> - different amplifier & load to the amp
>>>> - different crossover slope & Q
>>>>
>>>> Until then, enjoy the experimenting! :_)
>>>>
>>>> _-_-bear
>>> Thanks. Interesting. The low end on the ribbons is around 2k so I
was
>>> thinking that I would get way with the 6db crossover because I was so
far
>>> above it. Given that would you still recomend the 24db crossover or
>>> maybe a 12? Some of what I have read on the topic suggested keeping
the
>>> mid/high crossover simpler. In all honesty I'd probably think a
little
>>> bit about switching to a three way active to get a 24db crossover
rather
>>> than using passive.
>>>
>>> I'm clueless in the mic world. But I would like to be able to really
>>> test rather than rely on subjective listening. Can I hook the mic
>>> directly to my scope or should I have a preamp on it? I realize I
will
>>> have to power the mic if I do not have a pre-amp involved. What mic
>>> brands tend to be decent and what kind of dollar level should I be
>>> shopping at? tia
>> What Sonnova said...
>>
>> But you can get the Behringer "test" mic for $35 from most internet and
>> retail
>> places that sell it. Almost any mic will get you some good information,
>> even if it is "relative" and not absolute.
>>
>> Any soundcard and any freeware - no scope - will tell you enough
>> information
>> to make good decisions.
>>
>> As far as the xover? I'd try to make an xover that gets things below 2k
>> out of
>> the way a bit faster... You're only about an octave down at 2.5k,
assuming
>> you
>> are really at the -3dB point at 5khz. 6dB down isn't all that much. So
the
>> issue still remains.
>>
>> I doubt that you would get better fidelity, or put another way, reach
the
>> full
>> potential of a good ribbon tweeter using an electronic xover. Just my
>> opinion.
>>
>> Also getting a mid up to 5khz. creates some problems in terms of
response
>> and
>> polar response...
>>
>> This is a good example of why speaker design is a complex problem with
>> many
>> variables, and many compromises...
>>
>> _-_-bear
Thanks to DP for clarifying an im****tant point in his post...
>
> I think you make another good point about the midbass. It's response
curve
> goes to 6 but it not as flat between 5 and 6 and rolls off sharply at 6.
I
> had choosen 5k because of the ribbon. Originally I was trying to make
it a
> bit high to keep the vocals in the mids but your comments make me
realize
> that with a 6db crossover that's not happening anyway.
I'd be most concerned about keeping the 'vocals' in the mids between
<300Hz and
~3kHz. That's the nominal 'telephone response' frequency range where most
of the
energy that we perceive as 'vocal' is taking place.
As you are probably noticing, there is a real balancing act going on in a
real
world speaker system. What *sounds best* ends up being *best*. You
probably
can't get everything you want, no one does! There is going to be some sort
of
compromise, give and take.
You may well get a better sounding result with a higher order filter along
with
a lower rolloff frequency for the ribbon tweeter... you certainly can run
the
ribbon tweeter much closer to the resonant frequency/manufacturer's
recommended
rolloff frequency.
You might ask this same question on http://www.diyaudio.com
- there is a
speaker
design section in that forum, where there may be some others with specific
experience with the same tweeter or similar designs... the ratio of
actually
useful and valid suggestions to not so good ones is pretty high.
_-_-bear
http://www.bearlabs.com
>
> With a more effective crossover would you also recomend lowering the
> crossover point as well?
>
|
