Crossover slopes on RF 7

[v3spitfire]

Does anyone know what the crossover slopes are on the RF7. I'm thinking of getting an active crossover for bi-amping and am not sure which type would make most sense. Should they be the same as the RF7? I see some say 18db, 24db, 3rd order and 4th order... Thanks,

[v3spitfire]

Here is the xover I'm thinking of. Its 24db slope. Does any one see any issues with using this to biamp RF7's? Thanks,

dbx234 crossover.pdf

[Deang]

The main problem is that you will have no way to create a notch filter to control the horn's resonance peak.

[v3spitfire]

Thanks DeanG. I'll make sure I cut the freq high enough, and use to xovers to hopefully avoid this. If it doesn't sound right, I'll just throw them on ebay. Thanks for responding.

[sfogg]

"The main problem is that you will have no way to create a notch filter to control the horn's resonance peak."

Sure you can. Just put a notch filter directly on the driver you are trying to notch. That there isn't the rest of the speaker level crossover there isn't an issue.

With higher order slopes though the notch might not be needed.

Shawn

[hwatkins]

----------------

On 4/5/2004 7:59:39 AM sfogg wrote:

With higher order slopes though the notch might not be needed.

Shawn

----------------

Hello Shawn - I can't get modeling to give me comfort that a notch filter isn't needed in several variations. Since I am a novice, could you help me on the reason that it would possibly be excluded on higher order slpoes?

[sfogg]

Henry,

" Since I am a novice, could you help me on the reason that it would possibly be excluded on higher order slpoes?"

I'm basing this assuming the notch filter is outside the normal range of the driver in question.

For example say you have a two way speaker crossed over at 3kHz first order (6dB/octave) and that the upper frequency driver (tweeter/horn/whatever) has a resonance at 750hz that is say 10dB.

If the stock crossover was a first order slope then the tweeter will be receiving material at 750hz (2 octaves down) that is only attenuated by about 12dB compared against the material above 3kHz. If the tweeter has a resonance at 750hz then this could excite it enough such that you get a peak at 750hz from the tweeter which is almost the same loudness as what the tweeter is doing above 3kHz. That would be audible and probably pretty nasty sounding.

Now say that crossover was switched to say a fourth order (24dB/octave) crossover. At 750hz now the tweeter would be receiving material at 750hz that is 48dB down relative to the material above 3kHz. At 750hz the tweeter is now receiving a *very* small fraction of the power that it was with the first order crossover. As such it might be so low as to not trigger the resonance at all. And even if it does the resonance will still be at a much lower level then compared against the rest of the music.. in this case around 38 dB down relative to the rest of the music.

This is why when a speaker is designed with first order crossovers the drivers need to be very well behaved well outside of the frequency range they are being crossed over at.

Now if the tweeter had a peak at say 4kHz and it was being high passed at 3kHz then this doesn't apply of course.

Shawn

[hwatkins]

got it - cool

Thanks

[Trooper]

I'd say given the nature of the speaker drivers (metal) it's much more likely that there are potentially nasty resonant peaks at the higher end of each driver's response. Thus, the tweeter likely has a peak you'll need to address and the woofer probably has one an octave or so above it's pass band that should be addressed (even though it will already be rolled of somewhat by the crossover).

Back to the big picture...real speaker crossovers are not "cookbook" style filters. For example, even though the crossover for the RF-7 is listed as 2200Hz that doesn't necessarily (in fact it's not likely) that the HP for the tweeter and the LP for the woofer are actually at 2200Hz. There may be some overlap or actually a space between the two points - whatever it takes to make a flat ACOUSTICAL response when both drivers are operating. Also, the Q of each filter is unlikely to be the standard 0.7. Basically, unless your active crossover allows you to set the LP and HP independently, with control of order and Q AND you know what they should be or have a way to measure the acoustical response you will most likely not help anything.

Sorry, but speakers are more complicated than most anyone knows and you can't just jump in, set the crossover to 2200Hz and be done.

[Deang]

No doubt, and therein lies the weakness of electronic crossovers.

[sfogg]

" unless your active crossover allows you to set the LP and HP independently, with control of order and Q"

There are active crossovers that give you all these controls.

And unlike a passive when you make one change here it doesn't alter the rest of the crossover. And of course things like changing drivers is cake with an active crossover as is level matching between sections. Most passives require a lot of changes for changes like this.

I definitly agree that it can take some work to make it sound good though and measuring equipment certainly helps too. You of course need the same equipment if you make design changes (not as much for just part swapping) to passive crossovers as well.

Shawn

[v3spitfire]

I should have explained better what I'll be doing. I'll use 2 of these dbx crossovers. I'll have one set at about 2500, sending everything below this to the amp supplying the RF7 LF terminals (ie, 300 "extra" hz). I'll have the other crossover set to 1800, sending everything above this to the amp supplying the HF terminals (and extra 400 hz to the tweeter). The speaker's crossovers will get everything they need (or I'll adjust both up and down to be "safe". My main goal is to get the amps to work less so each only has to waste a few hz. I had asked a question that TBrennan answered for me in a previous thread, but the realized I kind of hijacked that one, so I started this one. I do understand the whole octave and slope thing a lot better now, so thanks everyone. Hope this is a little more clear. IN the end, I think the speaker will be getting everything it is intended, and nothing more.

[Trooper]

Whoa! If I understand your last post correctly that changes everything. So you're leaving the passive stock crossovers in the speaker? If so, then the active crossover definitely buys you nothing. As long as you take out the metal strips (or wires) between the woofer and tweeter inputs on the input cup and wire a separate amp to each set you've got what you want. Adding the active crossover will not cause the amps to "work" any less.

The detailed explanation as to why may be too complicated but here goes a try...

Above the woofers' crossover point the impedance keeps increasing way past the average 8 ohms (into the tens or hundreds of ohms). As this impedance increases with increasing frequency the amp is required to "work" less and less - enough as to be insignificant. Same goes for the tweeter.

[Frzninvt]

In addition, you are not going to have large gains in performance by doing so. Do they not sound good with a single amplifier driving each one? Bi-amping is only going to provide minimal gains if any at all. As trooper said to do what you are wanting to do you WILL need to bypass the internal crossover components and resolder the connections directly to the drivers. Why in the heck would you want to do that, you are adding more components to the system which creates another set of possible problems & issues.

[v3spitfire]

Thanks. Explaining that the impedence is causing the amp to work less anyway helps me understand why the active would not help. It sounds great the way it is, I just thought if I could get the amps working even less, I would have just that much more umph. I don't need it, but thought if it was an inexpensive mod, then I'de do it. Sounds like its done already, I have the jumpers removed. Thanks everyone who helped me finally understand (as much as I need to know). And, for helping me save money that I can put to the sub!