Much appreciated, good stuff. Some clarification on the circuit breaker comments; All circuit breakers can sustain their full rated amount indefinitely, period. The 80% is related to electrical code. Your comments on a typical circuit breaker capacity handling massive high kw PA loads are spot on, yet so misunderstood by many. To avoid nuisance tripping, a 20 amp circuit can pass 7-8 times the rated 20amp trip amount, .. for up to a second or more. It will allow up to 3x the rated amount for up to 10sec or so. Amazingly, the same 20amp circuit, can allow up to 2 times the rated amount for a period extending as long as 30 seconds. 𝙏𝙝𝙖𝙩'𝙨 𝙖𝙣 𝙚𝙭𝙩𝙧𝙖𝙤𝙧𝙙𝙞𝙣𝙖𝙧𝙮; 100amps for around 1-2 secs 60amps for around 10 secs 30-40amps for up to 30 secs It's always prudent to upsize any cabling or extensions to avoid voltage drop, and ensure all relevant connections are properly executed and tight. I've been involved deeply on both sides, a lifetime in audio engineering, decades as an FOH engineer, as well as over 10yrs on the house side, electrical dept, powering stadium tours etc. In addition to all the live tours, our facility hosted the Superbowl, multiple Final Four tournaments, and every type of event big/small. You spoke of driver failure, well I've seen every kind of power delivery failure imaginable! From catastrophic fireball to a puddle of nylon where an extension cord cap once was... (without tripping!) You mentioned the Bi-Metal strip in a breaker ... heating up etc. Typically breakers have two methods for sensing overcurrent. Akin to a soft limiter, and a hard limiter. The soft limit is thermal ... just as you described. However there's also a hard limit instantaneous trip based off a magnetic element. All the best
Very good in depth talk. The question still remains why manufactures don't provide these details in the datasheets, especially if they are known by the manufacturer? With enough experience you sometimes can derive (or at least have a decent starting point) the behavior from the T/S, looking at the mechanical structure (voice coil) itself as well as looking at the VA graph (expected power per frequency as function of the impedance). Taking things like power compression/factor into consideration of course. It only makes the whole designing process more complicated and time consuming than it needs to be and quite often needs more practical testing (which even takes more time).
Thanks, b_force! We could include pages and pages of supplemental material, but in the end it wouldn't really matter because all we'd be telling you about is our test in our enclosure. More and more the advantage of one particular woofer comes down to durability in a very specific application, which is defined by the customer. So we can give some useful data (and make a lot of hidden design choices) that lets designers make relative comparisons among our products, but then the customer needs to do thorough testing in their box with their amp and DSP. Our OEM customers have an advantage here, because they have the budgets to do objective testing and then the volumes to convince us to make revisions so the loudspeaker works better in their application.
Is it true that a sub with say 33 hz fs. "Cant" play much lower hz than that. ? I have a DLS nobellium 12i with a mega magnet. (BL 12.5) (33 fs) (You can find online pdf specs) Mr. Bennett say that BL is key. So - I just need to give it some EQ under 33 hz ? How low hz can i expect. I want to build a closed box. For home use. 1. How low hz can it go 2. How big a closed box must i build Thanks
you posted a klippel data set for 18DS115 a few years ago in avs forum, that was my decision to build a 2,3m long backloaded horn with it. Its the most powerful driver i have ever tested. Thx
i have sold 24 so far, building a better cnc and will continue. I guess its the loudest single sub 35Hz-80Hz i have heard in 20 years - had a rough time finding a good amp but i will try X4L. Thanks for the good videos, subscribed!
Did I understand correctly that you can set the voltage limit depending on frequency as long as I don't exceed AES power at that frequency or excursion? (Whatever comes first?) And the thermal limiter can use the same principle as long as it it's about 1/3 of then AES? 21sw152-4 is about 100ohm at 30hz (free air) theoretically I can put about 400v in to it at that frequency? Thank you for the great videos, please keep them coming!
It's important to always consider how long you're going to be applying the voltage. I'm solely concerned with the long term thermal capabilities of the woofer which have 30 second + integration times. You also need limiters that consider multi-second attack and tens-of-seconds release. You are trying to catch the average (heat) energy of the waveform here, anything short term or peak responding isn't important since it just gets soaked into the long term power. The nice thing about amplifiers with built in power limiters is they do the work for you, integrating across frequency and time and measuring the real delivered power which you can then limit for. In my experience DSPs outside of amplifiers, or simpler amplifiers with DSP which doesn't sense current but only volts, usually are useless for power limiting since they have too short integration times and are often peak sensing (!) instead of RMS.
@@bennettprescott i'm currently designing a 400v supply class d amplifier with onboard DSP. After seeing your talk i'm thinking of adding a currentshunt to feed back to the DSP. The math and integration is no problem there. On "normal" speaker prossesors i have usually used nominal impedance with a tonegen and a multimeter to set the limiter at AES power with fast attack and 5-10 second release. By the way, is M-noise something you think is good for measure real life SPL? Thank you
@@modalen2 I like M-Noise at least in concept, whether it's better in practice I'll leave to someone who has really studied it. It's closer to a real music crest factor, but it's complicated to have that crest factor actually arrive at the loudspeaker terminals. One advantage of an amplifier that measures delivered power is that it is crest factor agnostic, so the setting that works for rock and roll will also work for EDM.
Much appreciated, good stuff.
Some clarification on the circuit breaker comments;
All circuit breakers can sustain their full rated amount indefinitely, period.
The 80% is related to electrical code.
Your comments on a typical circuit breaker capacity handling massive high kw PA loads are spot on, yet so misunderstood by many.
To avoid nuisance tripping,
a 20 amp circuit can pass 7-8 times the rated 20amp trip amount, .. for up to a second or more. It will allow up to 3x the rated amount for up to 10sec or so.
Amazingly, the same 20amp circuit, can allow up to 2 times the rated amount for a period extending as long as 30 seconds.
𝙏𝙝𝙖𝙩'𝙨 𝙖𝙣 𝙚𝙭𝙩𝙧𝙖𝙤𝙧𝙙𝙞𝙣𝙖𝙧𝙮;
100amps for around 1-2 secs
60amps for around 10 secs 30-40amps for up to 30 secs
It's always prudent to upsize any cabling or extensions to avoid voltage drop, and ensure all relevant connections are properly executed and tight.
I've been involved deeply on both sides, a lifetime in audio engineering, decades as an FOH engineer, as well as over 10yrs on the house side, electrical dept, powering stadium tours etc.
In addition to all the live tours, our facility hosted the Superbowl, multiple Final Four tournaments, and every type of event big/small.
You spoke of driver failure, well I've seen every kind of power delivery failure imaginable!
From catastrophic fireball to a puddle of nylon where an extension cord cap once was... (without tripping!)
You mentioned the Bi-Metal strip in a breaker ... heating up etc.
Typically breakers have two methods for sensing overcurrent.
Akin to a soft limiter, and a hard limiter. The soft limit is thermal ... just as you described. However there's also a hard limit instantaneous trip based off a magnetic element.
All the best
Very good in depth talk.
The question still remains why manufactures don't provide these details in the datasheets, especially if they are known by the manufacturer?
With enough experience you sometimes can derive (or at least have a decent starting point) the behavior from the T/S, looking at the mechanical structure (voice coil) itself as well as looking at the VA graph (expected power per frequency as function of the impedance).
Taking things like power compression/factor into consideration of course.
It only makes the whole designing process more complicated and time consuming than it needs to be and quite often needs more practical testing (which even takes more time).
Thanks, b_force! We could include pages and pages of supplemental material, but in the end it wouldn't really matter because all we'd be telling you about is our test in our enclosure. More and more the advantage of one particular woofer comes down to durability in a very specific application, which is defined by the customer. So we can give some useful data (and make a lot of hidden design choices) that lets designers make relative comparisons among our products, but then the customer needs to do thorough testing in their box with their amp and DSP. Our OEM customers have an advantage here, because they have the budgets to do objective testing and then the volumes to convince us to make revisions so the loudspeaker works better in their application.
Is it true that a sub with say 33 hz fs.
"Cant" play much lower hz than that. ?
I have a DLS nobellium 12i with a mega magnet. (BL 12.5) (33 fs)
(You can find online pdf specs)
Mr. Bennett say that BL is key. So - I just need to give it some EQ under 33 hz ?
How low hz can i expect. I want to build a closed box. For home use.
1. How low hz can it go
2. How big a closed box must i build
Thanks
you posted a klippel data set for 18DS115 a few years ago in avs forum, that was my decision to build a 2,3m long backloaded horn with it. Its the most powerful driver i have ever tested. Thx
i have sold 24 so far, building a better cnc and will continue. I guess its the loudest single sub 35Hz-80Hz i have heard in 20 years - had a rough time finding a good amp but i will try X4L. Thanks for the good videos, subscribed!
We will all become youtubers eventually:) I haven't been watching yet but Im sure it will be awsome:) Getting my popcorn ready!
What do you think about using M-noise for finding power handling?
Did I understand correctly that you can set the voltage limit depending on frequency as long as I don't exceed AES power at that frequency or excursion? (Whatever comes first?)
And the thermal limiter can use the same principle as long as it it's about 1/3 of then AES?
21sw152-4 is about 100ohm at 30hz (free air) theoretically I can put about 400v in to it at that frequency?
Thank you for the great videos, please keep them coming!
It's important to always consider how long you're going to be applying the voltage. I'm solely concerned with the long term thermal capabilities of the woofer which have 30 second + integration times. You also need limiters that consider multi-second attack and tens-of-seconds release. You are trying to catch the average (heat) energy of the waveform here, anything short term or peak responding isn't important since it just gets soaked into the long term power. The nice thing about amplifiers with built in power limiters is they do the work for you, integrating across frequency and time and measuring the real delivered power which you can then limit for. In my experience DSPs outside of amplifiers, or simpler amplifiers with DSP which doesn't sense current but only volts, usually are useless for power limiting since they have too short integration times and are often peak sensing (!) instead of RMS.
@@bennettprescott i'm currently designing a 400v supply class d amplifier with onboard DSP.
After seeing your talk i'm thinking of adding a currentshunt to feed back to the DSP.
The math and integration is no problem there.
On "normal" speaker prossesors i have usually used nominal impedance with a tonegen and a multimeter to set the limiter at AES power with fast attack and 5-10 second release.
By the way, is M-noise something you think is good for measure real life SPL?
Thank you
@@modalen2 I like M-Noise at least in concept, whether it's better in practice I'll leave to someone who has really studied it. It's closer to a real music crest factor, but it's complicated to have that crest factor actually arrive at the loudspeaker terminals.
One advantage of an amplifier that measures delivered power is that it is crest factor agnostic, so the setting that works for rock and roll will also work for EDM.
22:55 i want to see this live messurement from blackbox in bootshaus cologne. I bet it would drop below -6 on the sub amps :D
Thank you
You're welcome! Next time I'll try and keep it to half an hour... but there was some interesting Q&A!
I'm going to put a 160k amp on 1 18ds
waauw. thanks!
Thank you for this educational talk.
B&C speakers Delhi me distibutior ka address milegaa ya contact number send me