The rising 3:rd order just below 2k - where is that from? Tweet or midrange?
The first measurements looked a bit hot upwards. There is a sort of a peak at 1k and one at 9... sort of ... that would be "bright". And the general FR is really "flat" - many points to a that a general drop towards the highs is preferred.
Did you do any listening position measurements?
//
The first measurements looked a bit hot upwards. There is a sort of a peak at 1k and one at 9... sort of ... that would be "bright". And the general FR is really "flat" - many points to a that a general drop towards the highs is preferred.
Did you do any listening position measurements?
//
The 3rd harmonic rise is from the midrange driver, but it's not get stronger as the volume increases.
Maybe that gated measurements looks slightly hot at the top, but around 1 kHz there is so much baffle influence on-axis, and around 9 kHz the tweeter gets directive. But look at the horizontal off-axis power average, it's smooth and sloping down.
The in-room response is now slightly sloping down towards the highs and sounds really good even at moderate high volume and longer listening. Tested yesterday evening.
Maybe that gated measurements looks slightly hot at the top, but around 1 kHz there is so much baffle influence on-axis, and around 9 kHz the tweeter gets directive. But look at the horizontal off-axis power average, it's smooth and sloping down.
The in-room response is now slightly sloping down towards the highs and sounds really good even at moderate high volume and longer listening. Tested yesterday evening.
Last edited:
It's not inconceivable if what I think is a Hypex Class D amplifier is getting into trouble with this 3 ohm tweeter.
Firstly from the low impedance, and secondly at high frequency where a tweeter goes high and inductive.
I like to fit a tiny Zobel to give the amp a lower high frequency impedance. As little as I can get away with really, because the full correction here, 7.5R and 0.68uF does too much high frequency rolloff to my ears.
I don't know what you've got in your parts box, but a 3.3R in series with the tweet, and maybe a 10R plus 0.33uF shunt might transform the top end. Cheap entertainment anyway.
Firstly from the low impedance, and secondly at high frequency where a tweeter goes high and inductive.
I like to fit a tiny Zobel to give the amp a lower high frequency impedance. As little as I can get away with really, because the full correction here, 7.5R and 0.68uF does too much high frequency rolloff to my ears.
I don't know what you've got in your parts box, but a 3.3R in series with the tweet, and maybe a 10R plus 0.33uF shunt might transform the top end. Cheap entertainment anyway.
I find that my ear always seems to want a down tilted response. About -2 db by 10K.
And I am very sensitive to 6-7k problems. I believe 6-7 is where those soft domes tend to break from piston to modal?? The shelf down of a couple db in freq response? Not sure
Anyway its the smooth straight line rake down to about -2 at 10K that seems to get it all right to my ears AND heres the thing, even 1/2 db off over an octave or more can upset it all in the 5-10k range. It can go from artificial to believable with just 1/2 db.
Also this careful tailoring is easily missed and frustrating because when I first level things right it will sound a bit dull. Extended listen needed to know.
And I am very sensitive to 6-7k problems. I believe 6-7 is where those soft domes tend to break from piston to modal?? The shelf down of a couple db in freq response? Not sure
Anyway its the smooth straight line rake down to about -2 at 10K that seems to get it all right to my ears AND heres the thing, even 1/2 db off over an octave or more can upset it all in the 5-10k range. It can go from artificial to believable with just 1/2 db.
Also this careful tailoring is easily missed and frustrating because when I first level things right it will sound a bit dull. Extended listen needed to know.
Good insight lowmass, thanks for that! I am start to feel similar as you described.
Yes, around 6-7 kHz for normal 1" soft domes where they begins to lose their pistonic movement I think.
I would be curious to try the Be version of that tweeter where everything is the same except the dome material which is Beryllium for the D3004/6640, but hey it's expensive as hell.
Yes, around 6-7 kHz for normal 1" soft domes where they begins to lose their pistonic movement I think.
I would be curious to try the Be version of that tweeter where everything is the same except the dome material which is Beryllium for the D3004/6640, but hey it's expensive as hell.
Last edited:
Having messed around with tweeters for years, my next experiment is a fullranger as tweeter:
CHN-50 full range driver | KJF Audio
planet10 says the Markaudio CHN-50 is excellent, and can be crossed much lower than most regular tweeters. Needs a 1L enclosure, and is dirt cheap.
My experience on cone tweeter dispersion suggests a supertweeter might help too. I am even thinking of a TST array, which is tweeter,supertweeter,tweeter since they are so cheap.
That's the hobby!
CHN-50 full range driver | KJF Audio
planet10 says the Markaudio CHN-50 is excellent, and can be crossed much lower than most regular tweeters. Needs a 1L enclosure, and is dirt cheap.
My experience on cone tweeter dispersion suggests a supertweeter might help too. I am even thinking of a TST array, which is tweeter,supertweeter,tweeter since they are so cheap.
That's the hobby!
I think it's a CSD problem with this tweeter. It's very similar to the SB 29 rdc I have measured and in many respects sounds very similar. I have not measured the Scanspeak tweeter but I've heard it and it sounds just as you describe. Unfortunately there is no notch or moving around the x-over point to fix it since it in the 7-12khz range.
Well, maybe a CSD problem, but how can we see that on the CSD graph? I don't see that, at least in my measurements.
Anyway if it's a CSD (or like this) problem, it can be a breakup of the dome? The diaphragm is very soft (I mean, really), hard to imagine how can she produce normal sound, even up to 30 kHz.
On the other side, metal domes if stronger and stiffer (but relative well damped) the better.
Anyway if it's a CSD (or like this) problem, it can be a breakup of the dome? The diaphragm is very soft (I mean, really), hard to imagine how can she produce normal sound, even up to 30 kHz.
On the other side, metal domes if stronger and stiffer (but relative well damped) the better.
I firmly believe that a driver can have a "sound" all its own, regardless of how it measures. I don't think we yet have all the analytic tools to disect sound in a way to fully characterize it... i.e sound from an instrument, a voice, or a driver.
It is possible that the tweeter you have simply sounds a certain way. If you are going to switch, I have recently become quite fond of the SB Accoustics SB26-CDC, which has a ceramic diaphram formed by deeply anodizing an aluminum dome. The deep anodize layer helps damp the break-up, which happens at ~ 24kHz. Even at twice its price it would be a bargain.
It is possible that the tweeter you have simply sounds a certain way. If you are going to switch, I have recently become quite fond of the SB Accoustics SB26-CDC, which has a ceramic diaphram formed by deeply anodizing an aluminum dome. The deep anodize layer helps damp the break-up, which happens at ~ 24kHz. Even at twice its price it would be a bargain.
Yeah, the SB26 series contains very nice tweeters. The only complain for me is their low sensitivity. It's low around 88 dB and they are not even 8 Ohm but 4. They nominally need 4x power than an average 91 dB 8 Ohm tweeter to produce the same SPL. Maybe the weak motor is one reason why they produce so low harmonic distortion, I don't know just speculating.
Although maybe the low efficiency is not a big problem in domestic situations.
The CDC anodizing doesn't do much to the (ultrasonic) break-up resonance according to measurements compared to the ADC version, although the CDC is not much more expensive either.
That tweeter may worth a try because it's not expensive and the distortion is less than much more expensive drivers.
Although maybe the low efficiency is not a big problem in domestic situations.
The CDC anodizing doesn't do much to the (ultrasonic) break-up resonance according to measurements compared to the ADC version, although the CDC is not much more expensive either.
That tweeter may worth a try because it's not expensive and the distortion is less than much more expensive drivers.
Have a look at Zaphaudio site at tweeter mishmash. He is a perfectionist so i trust his measurements. Looks pretty hashy above 9khz. The problem is its not frequency specific but broad band in it nature. Harmonic distortion in mid bass is easily noticeable but after 35 years of building and measuring speakers I have not found much of a consensus with others about what causes sibilance or similar artifacts. I'll take ultra low CSD vs HD
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.