Hi
I read somewhere that it would be beneficial to make lowpass for the tweeter.
I think the idea is to make it smoother sounding by reducing intermodulation distortion that can affect the frequency response down hearable range, and reduce tweeter range it has to play.
Any experience, good idea? Yes, no?
If one were to use it, how high it should be used to have effect on hearable range?
(reduce distortion)
I read somewhere that it would be beneficial to make lowpass for the tweeter.
I think the idea is to make it smoother sounding by reducing intermodulation distortion that can affect the frequency response down hearable range, and reduce tweeter range it has to play.
Any experience, good idea? Yes, no?
If one were to use it, how high it should be used to have effect on hearable range?
(reduce distortion)
Not necessary for a good tweeter but for one with a lot of ultrasonic ringing, who knows. But it would be better in that case to quench the peak with an elliptical filter, imo.
All the commercial speakers I've owned had a high pass on the tweeter. A 6 db/octave single capacitor is cheap. If you don't have current to burn from the amp, among other things, not sending low frequency current to the tweeter keeps the speaker impedance from being below 4 ohms or less. I set fire to a single output transistor pair amp in a 4.5 hour rehearsal driving a speaker that had 4 ohm bass drivers (two 8's parallel) instead of original parallel two 16 ohms. One simple cap series the tweeter was the plan in the market leader speakers I bought before 2010 when I found SP2-XT.
If the tweeter has a power limit (most do) than not having a low pass filter on the tweeter reduces the watt rating on the speaker total. For example the 70 w (continous RMS) RX22 compression driver tweeter. Pink noise rating (AES) for the 1980? SP2 (RX22 + 15' woofer) with 800 hz crossover, 175 watts. For the 1995? SP2-XT with 1200 hz crossover, 300 watts. For the 2004 SP2 with 1800 hz crossover, 500 watts the package.
Furthermore, driving the tweeter below the recommended frequency limit, can damage the suspension or diaphragm. Movements get uncontrolled out of range, xmax can be exceeded. Both the KLH-23s I bought in 1976, and the used Peavey 1210's I bought in 2005, had caps series the tweeter. First was a 3" paper cone, second was three parallel mallory piezo tweeters aimed in 3 directions. My 1970 LWE-III I blew a 3" cone tweeter with a 35 watt/ch dynaco ST70 amp. Don't know what was in there, my father gave them to Goodwill while I was at Ordnance Officer basic course.
Last edited:
You mean High Pass...right? A low pass filter passes low frequencies, and filters out the high frequencies above Fc.
Mike
Indeed! I had a number of young females that heard quite well above 20 kHz, which was a problem with the Altec 1" drivers that has/had a ~22 kHz spike plus had to add a galv. steel shield behind our then 25" CRT TV to boot!
Ultrasonic peaks can be filtered out if wanted. It however is not a lowpass but a contour notch.
I have experimented with low pass (or high cut) filters on tweeters based on the premise that there is a lot of noise above 10 or 12 KHz. None of my more successful designs have had one however.
I notch out the 25k+ peak on Ti domes. It makes a difference when listening to sources which have content above 20k. That can excite that breakup mode and down modulate into the upper treble. Things like cymbal strikes, triangles and acoustic guitars can provoke it.
If you hear sibilance on digital sources on speakers that have Ti domes and it doesn't sound that way listening with headphones, especially at higher volume levels, it can be caused by UHF breakup. It can also sound like gritty or raspy noise on some sharp treble transients (sort of like a poorly tracking phono cartridge). Alot of people wouldn't hear it, but if you know what it sounds like, your ears will always be drawn towards it in the future.
One way to know if you're Ti domes have this ringing breakup peak is by looking at the impedance sweep of the driver. There will be a slight blip in the curve between 24 - 30k. A simple LCR circuit will get rid of it.
If you hear sibilance on digital sources on speakers that have Ti domes and it doesn't sound that way listening with headphones, especially at higher volume levels, it can be caused by UHF breakup. It can also sound like gritty or raspy noise on some sharp treble transients (sort of like a poorly tracking phono cartridge). Alot of people wouldn't hear it, but if you know what it sounds like, your ears will always be drawn towards it in the future.
One way to know if you're Ti domes have this ringing breakup peak is by looking at the impedance sweep of the driver. There will be a slight blip in the curve between 24 - 30k. A simple LCR circuit will get rid of it.
You're talking about filtering out ultrasonic frequencies, to prevent the tweeter from attempting to reproduce them. I suppose this could be a good idea, depending on how the tweeter reacts to ultrasonic frequencies and if the program material actually has any ultrasonic frequencies.
But it's going the opposite direction of how much of "hifi" or "high end audio" has been going the last couple of decades, with the advent of 24/96 and 24/192 converters and the idea of reproducing whatever musical signal may be above 20kHz. All these different fads seem ironic to me.
But it's going the opposite direction of how much of "hifi" or "high end audio" has been going the last couple of decades, with the advent of 24/96 and 24/192 converters and the idea of reproducing whatever musical signal may be above 20kHz. All these different fads seem ironic to me.
Last edited:
Possibly only if the concentration of HF energy is significantly above the LP corner frequency. You'll be killing the openness and air, defeating the purpose of a good tweeters performance just to claim a margin of safety that won't really matter in the end.
How is it going the opposite direction if you're removing a significant distortion mechanism from the tweeter passband, improving sound? You wouldn't run a metal cone driver in its breakup region either, so why would it be smart to do similar with a metal dome? Just about any decent design using a metal cone driver will have a notch on the breakup frequency,,even if its not directly in the drivers desired range of operation ie. less than 24 dB of attenuation by the crossover LP itself.
Killing the openness and air might be less of a design objective to the warranty accounting dept. I agree it wouldn't sound as good...
Hi, thanks for answers very interesting read. In this specific case, tweeters im goin to use are:
-T25s-6 bliesma (silkdome)
-Tw29txn-b 8textreme)
Are these well behaved on ultrasonic end?
-T25s-6 bliesma (silkdome)
-Tw29txn-b 8textreme)
Are these well behaved on ultrasonic end?
No clue, TTBOMK the only ultrasonics I had available way back when were piezo horn drivers that gave us guys a sense of top end 'air' and sent the young females making quick retreats if I didn't EQ roll them off centered at 16 kHz.
These tweeters are fine without any further measures on the ultrasonic side. Very well controlled break up peaks all of them.