It's OK. I'm learning as I go along.
Most of the passive circuits in that article are known to me already, but it's still interesting for sure. What was new was the idea of using a resistive L pad to also dampen the tweeter resonance. It worked almost as well as a series RLC notch filter set at the tweeter's resonance.
Since I use high quality SS amps I don't think it's useful to simulate something with a high source impedance.
I have done the simulations with the crossovers installed and of course the Zobel filter works as intended on the woofer. The idea of a notch filter at the woofer resonance doesn't seem to do anything regardless of the presence of a low pass crossover at a frequency much higher than the woofer's resonance. (for example, notch is at 60Hz and LP crossover is at 500Hz or higher)
I find it curious that this kind of woofer RLC filter doesn't do anything to any of the other woofer parameters like cone excursion, maximum SPL, etc. This appears to be true for both sealed and vented boxes. I'm working with a Dayton 3" full range driver that I intend to HP cross over @ 500Hz and both a Zobel and a notch filter affect the SPL response of the driver, according to Basta! so it's surprising to me that it doesn't do anything to any of the woofers I tried it on. I would have thought that damping the resonant peak of the woofer would have some effect but I'm apparently mistaken. Maybe it has something to do with the woofer's alignment and thus system Q. I wonder if I simulated an alignment with a really high Q and installed the RLC network what would happen? Maybe later we find out...
Last edited:
Ha, I sort of reinvented the wheel, apparently:
http://www.davidsaudio.com/Infinity_Renaissance_Technical_Summary0001.pdf
Infinity used a "Watkins" woofer with dual voice coil. They claim that the woofer low end response is extended by their technique. It would not really be possible to implement their idea as far as I know because the second voice coil is much lower impedance than the other one. Most DVC woofers have equal impedance coils.
http://www.davidsaudio.com/Infinity_Renaissance_Technical_Summary0001.pdf
Infinity used a "Watkins" woofer with dual voice coil. They claim that the woofer low end response is extended by their technique. It would not really be possible to implement their idea as far as I know because the second voice coil is much lower impedance than the other one. Most DVC woofers have equal impedance coils.
Attachments
I would not be too worried about some phase shift; no passive Xover get away without it (except Urei and Tannoy, who did correct phase with all pass filters decades ago). The reason you don't see them anymore AFAIK is that the ear is not sensitive to absolute phase shift (it is to phase shifts that are different between the two ears, but that is a different story).
vac
vac
Last edited:
Thanks. Yeah, I was wondering if any of this was really audible anyway. The Watkins woofer looks interesting though.
Wayne doesn't use (or like) RLCs that much and he is very much against it. He primes and prefers for the use of very good quality materials/drivers/horns and cut the problem of peaks/resonances at its source. I'm not the right person to speak about it because my knowledge is very limited. He doesn't use it like in the diyAudio (members) forum use them so profusely. Some Pi speakers/crossovers are a pseudo first order where you can find some attenuation (with RC parallel to the driver, you can find this in his papers). Maybe this helps.
Many engineers use a resistor (or Lpad as you mention) for the attenuation of resonances at high frequencies in conjunction with the cap (or crossover). In this case it's obvious the CD drivers are of high sensitivity and need some form of attenuation. I see this in 1.order examples.
Hi,
So its also not useful simulating RLC woofer impedance correction,
or any parallel impedance correction, as all it does is draw more
current from the amplifier and nothing else, for prettier graphs.
rgds, sreten.
I'm not sure I understand what you mean, but obviously some kind of impedance correction is required for almost all drivers in order for passive crossovers to work as intended.
Well, I wouldn't use them if they didn't perform a useful function, which is the point of this thread regard a particular type of RLC filter.
Hi,
Not true at all. Impedance correction is needed for "textbook" or
"cookbook" crossover to "work" as intended, but such an approach
is completely missing the plot, because its practically not true.
e.g. checkout the designs at Zaph|Audio ,
impedance correction is hardly used, it is in some cases,
when needed. (And the designs in the links given earlier).
With low impedance SS amps there is no point whatsoever of
correcting for bass impedance peaks, it makes no difference,
and just draws more current from the amplfier, as it has to.
rgds, sreten.
Last edited:
A different way to say it is the impedance correction is needed for standard textbook crossovers to work. But with the advent of widespread used of CAD back in the '80s (I still use CALSOD...), one can achieve the target function without impedance correction- it's just that the crossovers won't be textbook, which is thoroughly unimportant.
Yes. If you are talking about <100Hz impedance spikes being corrected for, we're talking about a filter which is much more easily achieved in or before the amplifier. Not too many serious speakers use a passive highpass filter on the mains. Those that do would be sub/sat systems where there's a passive XO in the sub. Those don't get much love around here- 'cause it's a dog for wiring and performance. Easier to just do it the smart way.
Now, cap tuning the bass has been done to change the bass profile, which could be a vaid reason to tweak the impedance near resonance. But it's again more easily done with LL EQ.
So, you're saying that Mr. Watkins and the Infinity loudspeaker company are wrong? See the white paper I attached to a previous post. Granted, this is a special woofer, but still the RLC circuit was used by them for many years to cancel the bass resonance and apparently lower the low frequency limit of their driver.
Attachments
Not everyone builds there own electronics so much of this is a moot point. Not everyone has or wants a subwoofer with a plate amp and the associated electronic HP crossover for the satellite speakers.
How else are you going to put in a HP subsonic filter in a vented loudspeaker in order to control subsonic woofer excursion if you have no choice but to do it passively? Your passive "textbook" crossover would not work without an RLC filter to cancel the bass impedance resonance.
I've checked some of the passive crossover software and they say "does not work for compression drivers" which I like to use, so it's no use to me. Maybe you didn't read close enough about the "steep learning curve" I mentioned either. I don't have time to spend weeks and weeks mastering arcane electronic simulation software.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.