Pardon my ignorance. I have learnt a lot on this forum over the past few weeks, but why doesn't anybody use variable resistors in crossovers, as this will alow one to fine tune the speakers from outside the box? Is it because of the power of the circuits or what other reasons are there?
Hi Jeanrick, variable resistors as such aren't used, however variable lpads certainly can be used. They look like a big potentiometer but they are actually different in their construction.
Tony.
Tony.
The problem is power.
Most crossovers will use quite high power resistors. It is not easy to obtain high power variable resistors.
Most crossovers will use quite high power resistors. It is not easy to obtain high power variable resistors.
When I adjust crossovers, I use a soldering iron. There is just as much chance that I will need to adjust a capacitor, or a notch filter, as a padding resistor. Having a variable L-pad would actually be inconvenient as the values aren't marked and an L-pad alters the impedance curve.
Many thanks guys - I didn't think it would affect the other components or curve too. It just seems a lot of work to play around with resistors and soldering them every time.
It really depends on where you are with it, but once you get a crossover zoned in the changes become small and specific...nothing just an L-pad can fix.
I would suggest you design your crossover to exclude woofer cone breakup, tweeter resonance, roughly match levels and phase at the listening position, as well as your other system requirements. Then use an equaliser to develop the tone. It is much simpler and more accurate when you can make virtually instant changes. Then later you can translate that back into soldered components.
I would suggest you design your crossover to exclude woofer cone breakup, tweeter resonance, roughly match levels and phase at the listening position, as well as your other system requirements. Then use an equaliser to develop the tone. It is much simpler and more accurate when you can make virtually instant changes. Then later you can translate that back into soldered components.
If you are able to take good measurements of the SPL over frequency you can dial in the tweeter level very quickly.
L pads in circuit are infrequently used, they oxidise & become scratchy, so they're better replaced with fixed resistors. I have one with the cover removed and alligator clips attached; this allows me to measure the values of it's arms when tweaking, and then replace it with fixed resistors
A crossover network is a filter network whose characteristics including crossover frequency depend on the resistance of the load. A potentiometer used for level control changes the crossover frequency besides changing the driver output level. When an L-pad is used instead, the crossover network sees the same load impedence so the crossover frequency does not change but the output level still changes. An L-pad is just two potentiometers, one in parallel and one in series. As you turn it, the resistance of one increases while the other decreases by the same amount. This is how the impedence on the crossover network is kept constant while adjusting the level of the driver.
Neither potentiometers nor L-pads should ever be used with woofers. Any resistor in series with a woofer reduces the effective electrical damping factor of the amplifier to surpress spurious resonances. Therefore, in designing a speaker, the woofer should be the least efficient (sensitive) driver in the group and the midrange and tweeters padded down with resistors or L-pads to match the level of the woofer to bring the system into acoustic balance.
Neither potentiometers nor L-pads should ever be used with woofers. Any resistor in series with a woofer reduces the effective electrical damping factor of the amplifier to surpress spurious resonances. Therefore, in designing a speaker, the woofer should be the least efficient (sensitive) driver in the group and the midrange and tweeters padded down with resistors or L-pads to match the level of the woofer to bring the system into acoustic balance.
Thanks guys, but then I need more help. Instead of spending around $500 on a high end passive crossover for the Seas Small Thor build, would it not be a better option to buy an active digital crossover?
I know little about these things and any help would be highly appreciated.
I know little about these things and any help would be highly appreciated.
An active crossover will require multiple amplifiers, one for each driver or group of drivers. Whether analog or digital, technically it is a far superior idea to a passive crossover network for many reasons however it is a much more expensive route. One caution, if you use an active crossover, I'd still put a capacitor in series with the midrange and especially with the tweeter driver. The capacitor will prevent any sudden unexpected transient, even a turn on thump from accidentally destroying the driver in a way a fuse can't. (You can also use fuses for added protection as well.)
Why destroy the perfect signal path with an in-line capacitor?
In an active crossover that is excatly what you are trying to remove.
Just use well designed amplifiers that have NO TURN ON THUMP.
I have a NAD T775, which has 7 seperate channels at 100W RMS each. I was thinking of using seperate channels, one for the two woofers and one for the tweeter, or can this not be done? The manual says that it can be used for bi-amping and I currently run it on a 5.1 basis, leaving two spare channels. Aparently each channel has its own "amp".
With a tweeter connected to an amplifier with no blocking capacitor, the slightest low frequency disturbance will destroy even the most expensive driver. Use one that is appropriate for a first order filter below but not too far from the desired crossover frequency to prevent low frequency transients from ever reaching the tweeter. They can't handle the excursion. On thump or moderate 60 hz hum for any reason will be all it takes to wipe it out. Anyway, that's my recommendation.
It could. Passive equalizers are not only expensive and inefficient, they are very hard to adjust. You literally have to rewire them with substitute components which are often selected in developing a speaker for manufacure by trial and error, a long and tedious process. Even if you were going to manufacture a speaker in quantity, it would be much faster and more efficient to determine the optimal crossover characteristics using an active crossover and then design a passive crossover that is equivalent. Bi-amplification allows optimization in selecting each amplifier for its purpose, say for example a high powered solid state amplfier for the woofer, a lower powered tube amplifier for the midrange and tweeter if tubes are your preference. It also prevents back emf from the woofer from exciting the midrange although a steep crossover slope of 12db per octave in a passive network should prevent this problem.
Try using variable Linkwitz/Reily band-pass filters (12 or 18 db slopes) and bi amp or tri amp. Of course you bypass the original crossovers. It's more complex (And costly), but the results will astound you. Sounds like nothing I've heard for five times the cost, and it is also much more efficient ie. less power needed..
Any required apologies for dragging this one out of the dust bin as well as the naïveté of the question...
Here’s the problem. I have a very carefully crafted ($$) OEM speaker level input time/phase coherent/level/slope passive low pass sub crossover intended to integrate a known main and sub. Unfortunately the line level output is too low when there is very low room gain. And the gain on the sub amp is fixed. Since the room gain will be variable, it would be nice to be able to vary the crossover level output relative to the speaker level input. Since the speaker level reduction section of the circuit involves - as I understand it - just a simple resistor or two, would a variable pot or lpad be possible in such an application?
Here’s the problem. I have a very carefully crafted ($$) OEM speaker level input time/phase coherent/level/slope passive low pass sub crossover intended to integrate a known main and sub. Unfortunately the line level output is too low when there is very low room gain. And the gain on the sub amp is fixed. Since the room gain will be variable, it would be nice to be able to vary the crossover level output relative to the speaker level input. Since the speaker level reduction section of the circuit involves - as I understand it - just a simple resistor or two, would a variable pot or lpad be possible in such an application?