I have played around with a pair of G3's for years now in various combos of drivers, boxes, & no boxes. the trick I have found most challenging is to find drivers of equal"ish" sensitivity or all you hear is the tweeters overpowering in their frequency range the other drivers.
you can buy a pair and test it yourself but don't forget replacement Aluminum foil which is not that expensive though
btw, about the xos point of the TPL150-H, what's the lowest frequency that does't compromise the performance of TPLH? If we wanna make the best sound out of it, what frequency you would choose? (or at least we say, it won't sound its best before achieving that frequency?)
That's about exactly where I found the Arum Cantus G1 limit was - Also 8th order high pass at 1.2k sounds harsh and ripped the aluminum on transients. 2k or higher is best.
I still prefer a good 1" compression driver in a good horn. Ribbons can sound great but never seem to do :live" sounding transients right. I don't have the Faital 1" mentioned. The 1.4" HF144 is great in the midrange but the upper treble is pretty close to terrible.
The AMT (big Heils) always sound fake to me even with the 2K bump removed.
I have to mention I revisited the HF144 and in a big horn (HR9040) with some passive eq the treble is better than terrible. Actually quite good. 😎
My 1st choice about sound quality would be SB Acoustics TW29DN-B-8 Satori, 2nd Mundorf AMT 29CM1.1-R & 3ed Faital Pro HF108.
Mundorf AMT29CM1.1-R
Hi koutrou,
I am developing a 3-way high-efficieny with the AMT29CM1.1-R and the result does not meet my expectations, it is sometimes harsh sounding, at least violins. Voices and Jazz are ok, other instruments too.
To complement the midrange, the tweeter got a 12dB crossover. The filter is a 4,7uF and a 0,68mH. Acouctical crossover point is around 2k or little higher.
Meassurent is excellent. Phases and addition of both curves are top.
My fear is, I will never be totally happy with this tweeter. Next step is not clear to me, should I try a different one? Ribbon, beryllium, horn?
How do you use the AMT29CM1.1-R?
reg.,
Bernhard
Hi koutrou,
I am developing a 3-way high-efficieny with the AMT29CM1.1-R and the result does not meet my expectations, it is sometimes harsh sounding, at least violins. Voices and Jazz are ok, other instruments too.
To complement the midrange, the tweeter got a 12dB crossover. The filter is a 4,7uF and a 0,68mH. Acouctical crossover point is around 2k or little higher.
Meassurent is excellent. Phases and addition of both curves are top.
My fear is, I will never be totally happy with this tweeter. Next step is not clear to me, should I try a different one? Ribbon, beryllium, horn?
How do you use the AMT29CM1.1-R?
reg.,
Bernhard
Last edited:
the 8 inch handing off to the Arum at around 3 k is probably not going to be a good dispersion match. Off axis will likely be far from uniform.
Better to cross at 1.5-1.8 k with a wave guided driver ( ribbon or other wise) if you have to use an 8 inch mid
or
Put a 4 inch driver between the 8 and the Arum
Better to cross at 1.5-1.8 k with a wave guided driver ( ribbon or other wise) if you have to use an 8 inch mid
or
Put a 4 inch driver between the 8 and the Arum
Last edited:
This may be it...
The amt as well as many "ribbons" can be extremely revealing and unforgiving of source. In almost every single design I have done using this type of driver I had to have a downward tilted on axis response to get it right. Flat always seemed to sound harsh.
Approx -2 down by 10Khz ( relative to 1 Khz) often seems to be about right. I use a simple "rake" circuit or sometimes a RLC trap to get this response. And watch out for any hump between 2 and 4 Khz. A dip is ok but even 1 db up through here can be harsh.
Also I dont know what size mid or woof u have below the amt, at your cross point, IF cone is larger than about 5 inches then u can have great on axis response with an ugly off axis response due to non uniform dispersion. Sometimes this off axis response non uniformity in some rooms will always make the overall sound harsh
Last edited:
Midrange is a Fostex 8" FE208EZ with approx 97dB. Crossover is little higher than 2kHz, electrically 12dB. The horizontal off-axis response at crossover point is "excellent" at 15° and even at 30°. Vertical is not, of course. In general the sound is very stable on every place of the room, much better than initially expected due to the large drivers - and better than well known brand speakers in the 20k€ range did.
Woofers are Scanspeak 10" sealed with Hypex DSP modules.
Amp is a 300B Single Ended DHT and DAC is a Rockna R2R. Cables and component placement is good quality.
Don't get me wrong: It sounds damn good with many albums, but especially violins do not how I'd like them to do.
Tweeter level is always matched by ear. Final confirmation is done by measurement, it shows a decent downwards direction, as you described. The AMT is very large and from 8kHz upwards just few degrees off-axis result in a downwards direction anyway - more than I like.
My understanding is, if violins sound on the harsh side, it is not the range from 10-20kHz, but lower.
I am not sure, if:
- the large AMT in general is very unforgiving (more than I like), then I have to change the tweeter. I heard small AMTs many time, they sounded really good.
- the root cause is too cheap crossover-parts during development ("extremely revealing and unforgiving of source")
- the harsh sound is coming from crossing to low. I expect higher uneven distortion lower than 2kHz
- "bad" frequency response is the reason, even if I am in a +/-2dB range from 800Hz upwards (with very smooth downwards direction to the right), then I have to work on the crossover again
reg.,
Bernhard
Greetings ! Is there some sort of mathematical equation to determine dispersion angle with respect to driver size ?
Driver diameter is the most relevant factor. But in the end every driver is different.
Last edited:
as euro says diameter is usually the issue and an 8 inch is a bit large to cross at 2 k
However it all can change IF the 8 inch cone behaves more as a lossy cone where at higher freq only the center of the cone is moving ( this is an approximation). Anyway such a cone looks like a smaller diameteter source at higher freqs and thus can disperse better.
The opposite approch is a very stiff cone like a metal type where they may try to get the cone to operate with much less flex and loss
However it all can change IF the 8 inch cone behaves more as a lossy cone where at higher freq only the center of the cone is moving ( this is an approximation). Anyway such a cone looks like a smaller diameteter source at higher freqs and thus can disperse better.
The opposite approch is a very stiff cone like a metal type where they may try to get the cone to operate with much less flex and loss
Well assuming the AMT doesn't have any pulse issues ( I dont know would be good to see pulse and CSD)
I lean a bit towards the idea that some recordings simply dont sound all that good with a revealing driver. U can play with putting a dip of a few db in critical areas like the BBC dip that some talk about.
Having said that are we sure the problem isnt in the midrange driver? If there is a peak in response in that driver anywhere under 4 k it could be causing all sorts of issue with a 2 k cross to the very clean tweeter. U may need a trap on the midrange cone around 3-4 khz
Last edited: