A Morel CAT378 will be dealing with about 8 watts power when reproducing 103 dB. This shouldn't be an issue at all for an hour or so continuously crossed above Fs at fairly compressed audio levels. I've pushed my CAT378s harder than that for several hours and they still measure / perform the same. I have them "protected" with a 0.75A polyswitch, which will trip at about double that rating after several seconds. So far however it has been tripped once, and only because of my curiosity if the polyswitch was sized correctly. The CAT378 is a tough little tweeter and it remains very composed at higher drive levels which would normally distress your typical 25 - 28 mm non-ferrofluid cooled fabric dome, especially with a lower HP filter.
You can safely get up to 106 dB continuous (pink noise) level with a CAT378 (depending on crest factor and type of music), which would easily integrate into a 110 dB capable 2 way system. The type of program material will determine the power density to the tweeter.
The Audax titanium (and gold) dome will also cope well with these levels. It won't like the lower HP filter though. It will require a HP of at least 3500 Hz 2nd order.
You can safely get up to 106 dB continuous (pink noise) level with a CAT378 (depending on crest factor and type of music), which would easily integrate into a 110 dB capable 2 way system. The type of program material will determine the power density to the tweeter.
The Audax titanium (and gold) dome will also cope well with these levels. It won't like the lower HP filter though. It will require a HP of at least 3500 Hz 2nd order.
Last edited:
The ATC 1 inch fabric dome is also a candidate with its double suspension and greater xmax. The S-version found in most of ATCs monitors is crossed around 3500 Hz 3rd order and keeps up with their massive 3 inch mid driver. Unlike the mid, the tweeter is available to buy separately for diy purpose, but i wouldn't outright tell them its being used for that.
Ok. Thanks for all the responses.
Of course I will test what I have on hand and report back.
I have been a fan of Yevgeniy’s HIficompass website for sometime (donate if you can!) but, apart from the compression driver, no other has been taken to the 110dB level.
The T34A/B have a good chance, as well as the waveguided Satori- both of which was tested up to 108dB at 10KHz.
I might have to review the physics of whether it’s possible to even reach 110dB for a dome tweeter…
Of course I will test what I have on hand and report back.
I have been a fan of Yevgeniy’s HIficompass website for sometime (donate if you can!) but, apart from the compression driver, no other has been taken to the 110dB level.
The T34A/B have a good chance, as well as the waveguided Satori- both of which was tested up to 108dB at 10KHz.
I might have to review the physics of whether it’s possible to even reach 110dB for a dome tweeter…
Last edited:
This doesn't mean they were tested at 110dB. This is just the SPL that the microphone is recording when audioxpress tests them. They have the microphone closer to the driver than 1 meter. The actual SPL for the distortion tests is always mentioned in the text. In this case 94dB.
Thanks Headshake. The problem with Voice Coil is that sometimes the details are hidden in the text, but not in the legend of the graph. As Vance wrote:
"For the last objective test, I used a noise stimulus to set the 1 m SPL to 94 dB (3.20 V) and measured the second (red curve) and third (green curve) harmonic distortion at 10 cm (see Figure 9)."
"For the last objective test, I used a noise stimulus to set the 1 m SPL to 94 dB (3.20 V) and measured the second (red curve) and third (green curve) harmonic distortion at 10 cm (see Figure 9)."
Ok.
Upon investigating this further, my system CAN do 110dB@1m.
the dynamic compression is due to the measurement technique. REW does not use MLS for measurements; but log sine sweep, the length of which is adjustable from 128K samples all the way to 4M samples.
Using the 128 or 256K samples; sweeps are short and there is NO loss in output at the top octave.
I had the measurement set at 512K samples from my precious setting when measureing at 2.83V.
Do NOT use 1M+ samples for stress testing small transducers.
The test duration is very long (20+ seconds) of continuous power at 28.3V.
No dome tweeter is able that much heat- don’t let out the magic smoke.
Upon investigating this further, my system CAN do 110dB@1m.
the dynamic compression is due to the measurement technique. REW does not use MLS for measurements; but log sine sweep, the length of which is adjustable from 128K samples all the way to 4M samples.
Using the 128 or 256K samples; sweeps are short and there is NO loss in output at the top octave.
I had the measurement set at 512K samples from my precious setting when measureing at 2.83V.
Do NOT use 1M+ samples for stress testing small transducers.
The test duration is very long (20+ seconds) of continuous power at 28.3V.
No dome tweeter is able that much heat- don’t let out the magic smoke.
Last edited:
That makes sense. When I saw your compression test comparisons in post # 14, I suspected there might be some measurement error, but I had no ideas on how you should proceed.
j.
j.
Can your ears do 110dB by the way?
Doesn't REW do any kind of MLS, periodic noise or pink noise? Or just not with this type of measurement? I'm an ARTA guy, so just curious here.
Doesn't REW do any kind of MLS, periodic noise or pink noise? Or just not with this type of measurement? I'm an ARTA guy, so just curious here.
Log sweep or stepped sine for distortion measurements only.
Ears can do 120dB or 0 dB, depends on how far away you are from the source. 😛
Ears can do 120dB or 0 dB, depends on how far away you are from the source. 😛
Totally. And I hear what you are saying and the critique is welcome.
This testing is part of QC testing to verify that my design meets modelling expectations. In this case I am looking to see if my HF or LF driver limits the useful output of my system; and when to apply the limiters to protect the drivers from misuse.
This testing is part of QC testing to verify that my design meets modelling expectations. In this case I am looking to see if my HF or LF driver limits the useful output of my system; and when to apply the limiters to protect the drivers from misuse.
Bliesma T34 as has been mentioned. If you want a waveguide Augerpro already has a couple versions that could be 3d printed etc for the T34B in his thread here.
