PDA

View Full Version : adding a resistor to crossovers?



jaygeorge1979
11-07-2005, 09:35 PM
I have seen some DIY setups, such as

http://www.zaphaudio.com/audio-speaker16.html

which show on the crossover circuit a resistor added in series to both drivers...in the description it says something about brightness, adding a higher R makes it less bright...can someone clear this concept up for me?

PV Audio
11-07-2005, 09:36 PM
a resistor effectively dampens the impedance of the circuit of the driver being used, and thus attenuates that certain region to flatten it out and make it less peaky and bright.

jaygeorge1979
11-07-2005, 10:11 PM
so they arent mandatory in these crosovers...just personal preference for brightness?

thanks dave...where u been lately? you havent posted on my threads in days...

PV Audio
11-08-2005, 07:15 AM
i've been around :)

and no, they aren't mandatory, but are excellent for tweaking :)

jaygeorge1979
11-08-2005, 10:28 AM
anyone know if the behringer DCX2496 active crossover thing can throw in a resistor, or would i have to do that manually?

PV Audio
11-08-2005, 11:01 AM
i couldn't tell you, thylantyr would be the best to ask

thylantyr
11-08-2005, 11:28 AM
anyone know if the behringer DCX2496 active crossover thing can throw in a resistor, or would i have to do that manually?

The Behringer is a powerful processor.

*active crossover
*parametric EQ
*delays
*phase control
*gain control

If you want to change the sound of your speaker, having all those features
sure beats a silly ole resistor eh ? ;)

jaygeorge1979
11-08-2005, 01:17 PM
you make a good point bud...the thing is, i would use this processor to determine how i want to build passive crossovers for speakers that i want to build...how can i mimick all of that stuff in the crossover?

for the record, i do plan on using this for my home theatre setup, that is, once i build it...but for other speakers that i want to build, say, for my dad or something...i would mimick the crossover using a network of caps and inductors :)

thylantyr
11-08-2005, 04:09 PM
how can i mimick all of that stuff in the crossover?

With a digital crossover all of this is easy;
*active crossover
*parametric EQ
*delays
*phase control
*gain control

Making a passive crossover to mimick these;
*active crossover {easy}
*parametric EQ {hard}
*delays {almost impossible}
*phase control {almost impossible}
*gain control {you can only attenuate with passive, ie LPAD the tweeters,
LPAD the midranges, etc., but for higher powered designs the LPAD introduces
power losses}.

plus, you need a zobel for those odd impedance curves that need to be stabilized, don't need one for active.

Sooo... I only see one easy item on that list, the generic crossover. The trick is
to pick drivers that will work good with a generic crossover design. If you can
find a combination that works without resorting to a steep slope, you reduce
crossover parts and cost.

If you know you will convert a design to passive then you need to find
drivers with less problems 'out of the box' otherwise you will have to
add more circuits to fix problems in your passive crossover design, that is where
the electronic skill is required. You can also buy software and study how to use
it to do this for you. You also might want to measure the drivers if you want
more accuracy because if a parameter changes, it will skew the whole design, ie
no zobel in a design will cause your crossover frequency to shift and other gremlins.

Hypothetical. You like Seas Excel 8" in a MTM design. This design requires a
1.5khz - 2khz crossover point {low pass} with a 24dB slope to keep the driver
from excitation as the cone break up modes are nasty. Pull a chart on that
driver, Seas W22EX001;
http://www.madisound.com/pdf/seas/e022.pdf

Some people won't even consider this driver unless the crossover is 1.5khz
with a steep slope to 'filter' out that high end crud. Look at the nasty spike
at ~5khz.

If you use a digital crossover you can use a higher crossover frequency in
conjunction with the parametric EQ to fix these problems but translating this
'fix' into a passive network would require more skill on the designers part,
so your choices would narrow and you might only consider a lower crossover
point.

There is good news, you can use drivers that have less problems to keep
the passive crossover design simpler and those designs can work well. The trick
is to find these 'good sounding' drivers.

There is alot of comedy in the driver manufacturing business. For instance,
Seas Magnesium cones are rated high by audio-geeks in spite that their exotic
cone materials are full of gremlins, yet they are always recommended as elite
drivers -- are they ?

Focal Kevlar, another highly recommended driver. Same thing, nasty cone breakup modes that have to be dealt with. Out of the box, they aren't elite
drivers at all unless you tame it's problems.

Focal W cones, I have a 6" WM driver with phase plug, this driver has a pretty
nice chart but oddly enough, it failed SQ tests all the time during audition. It uses
another exotic cone material.

Which drivers have I found that sound great in the midrange? treated paper cones. I'm fond of a particular brand, not cheap, but you can email me for details.

So... charts tell a tale but not all. You have to audition the drivers to find a winner,
trust your ears.

jaygeorge1979
11-09-2005, 12:11 AM
yea....i figured most of the features that the DCX gives me would not be mimickable with passive circuitry....oh well...thats why i am gonna keep that bad boy for myself :)