Jump to content

Burnt a costly drivers coil with a pricey amp../Followup and prevention of costly mistakes..


LSC

Recommended Posts

Sooo here we go..
As the title suggests I've somehow managed to fry my first pro audio speaker...
Not really proud of it but I thought I'll make it a new topic so me and others can learn from my mistake and maybe some of the more experienced members here can give us a lil rundown on how to set the limiters right. :)

So let me describe what happened:
I was having a lil snooze around 4 at night when someone came up to me and said that the dj requests some help...
I went and had a look at the amp rack to find my NX6000 turning on and off because of protection mode..
I couldn't believe it but after sticking my nose into the vents of my skram I had to face the reality that my beloved 21ds115 which seemed so indestructible had burned it's coil x(

I fully deemed this impossible as the NX6000 puts out 1.8-2kW sines and it was a techno night, so no prolonged sinewaves like you'd have with dubstep or bass music in general...
Butttt; now I know better and I suppose it was just getting hotter and hotter as the night passed by and thermal build-up was getting too much at some point?!
Or could it be that steep filters are just better than the 12db BW I was using as high pass filter?
Could it be that the DJ made the signal clip and the DSP somehow produced a harmful signal?

Now to the prevention of this happening again..

Does it make sense to add a temperature sensor to the magnet of the driver which can be read by a handheld thermometer or perhaps even an arduino to monitor the temperature and adjust accordingly?

Can someone please explain how to set limiters in the right way so nothing can get hurt but no bass resources gets wasted?
 
...I already thought of running sines, measuring them at full amp throttle and set the limiters according to the RMS rating of the driver but that seems (seemed haha) like overkill..

Please share your thoughts and tips fellow bass lovers so I can assess what went wrong and how to avoid it next time as this was a quite traumatizing thing to happen on the second occasion I had my skram out and about..

Thanks in advance, Nathan
 

Link to comment
Share on other sites

I'm sorry to hear that.

DJ's are notorious for pushing subs as hard as they will go and blowing speakers. This means often pushing the signal DEEP into bass amp clipping and leaving it running like that for hours. What this does is effectively chop the peaks off of the signal and increase the RMS power into the drivers. The drivers are no longer getting "breaks". If the 6K was effectively clipping constantly it possible that the driver received an average power of 1000w or perhaps even more. If that happens at frequencies with little driver excursion it will result in a cooked coil very shortly. 

I wasn't there and there is no way of knowing what the content was, what spectral makeup, or whether the above is true, but I'd bet that something similar to the above happened. The signals used and the duration are big variables. 

Despite having an AES power handling of 1700w in no way will the 21DS115-4 handle that amount of power for any significant amount of time. None of the drivers will. A 100% duty cycle signal with that amount of power will destroy the coil in under a minute. Was a limiter in use? The harder and longer that the subs will be played the more tightly set the limiter will need to be. Also the type of music is a big factor. A sub for HT or sporadic music listening in a home can probably get away with no limiter. The content is very dynamic and the amp/sub isn't constantly pushed to its limits. Live sound with rock type music would be another area where the limiter may not need set very hard as it would mostly just be the kick drum peaks with bass guitar or other bass instruments much lower in level. Once you get into electronically generated bass heavy music with lower crest factor and prolonged maximum output use you have to be much more conservative. 

As an example of how conservative... The 21IPAL driver is rated as a 2500w AES driver with a huge 6" coil, very high efficiency and tons of motor venting etc. In the IPAL amplifier module manual the long term power limiter setting is recommended at 700w maximum for that driver. About 1/4 of the AES rating and that's one of the highest power handling drivers on the market and that's the bleeding edge maximum that they recommend for it. The default setting for it is 400w...

Based on that I'd say that starting with a long term average power limit of 400w maximum with a 3 or 4 second attack and release seems a reasonable start for the 21DS115 and it may need adjusted down a bit from there. Your peak limiter can be set much higher with a much faster attack/release. Some limiters are voltage based, some are current, and others power. If yours is voltage based something like 40 volts rms equates to about 400w into 4ohms. Unfortunately the NX Behringers only seem to have a single "peak" voltage limiter which is really not much to work with in this case. The problem is not peaks but sustained power over long periods. Perhaps someone can confirm the available limiting settings on the Behringers. I don't own or use them. The online information is surprisingly vague

Limiting is a complex subject and there are so many variables involved. If you are letting others "pilot" your system they will invariably bang it as loud as it can go all night no matter how much rig you bring and how loud it gets, they will find the limits. You will have to plan for worst case scenarios if handing the controls over to others. Basically the same as most companies have to do with their active speakers. 

  • Like 2
Link to comment
Share on other sites

Wow thanks for your thorough reply!

Very eye opening.. 

It gives me something to work with and puts my (can't really blame the dj) abuse into perspective..

I assumed that the RMS rating states how powerful a sine wave can be for 2 hours continually ☹️ 

Silly me.. Oh well there's always a first... and more to learn haha..

Is there any literature on that? 

Also if there's any more input on how to set this combination up for bass heavy content then that'd be very appreciated!

Will go from ricci's recommendations but maybe someone else who has experience in limiting right for low crest factor signals wants to chime in ;)

Also does anyone like the idea of temperature sensors? Apart from being impractical they would allow us to really know how the driver is doing regardless of how sophisticated the limiting works.. 

Link to comment
Share on other sites

22 minutes ago, LSC said:

Wow thanks for your thorough reply!

Very eye opening.. 

It gives me something to work with and puts my (can't really blame the dj) abuse into perspective..

I assumed that the RMS rating states how powerful a sine wave can be for 2 hours continually ☹️ 

Silly me.. Oh well there's always a first... and more to learn haha..

Is there any literature on that? 

Also if there's any more input on how to set this combination up for bass heavy content then that'd be very appreciated!

Will go from ricci's recommendations but maybe someone else who has experience in limiting right for low crest factor signals wants to chime in ;)

Also does anyone like the idea of temperature sensors? Apart from being impractical they would allow us to really know how the driver is doing regardless of how sophisticated the limiting works.. 

I don't consider myself an expert on setting limiters and everyone seems to have a little bit of a different take on how to set them. 

With the amount of power you have with the Behringer amp you shouldn't have to worry about short term peaks or excursion except below tuning. You are concerned with long term heat saturation only. 

A few links. Like I said it's a deep hole and I don't claim to be an expert in this field. A lot of it seems to be trial and error depending on the settings available in your amp or DSP. Perhaps some others with hands on experience will chime in.  

https://www.powersoft-audio.com/en/downloads/documentation/application-notes-white-papers/quick-guides/804-how-to-setup-limiters/file

https://www.rane.com/note127.html

https://www.bassboss.com/limiter-subwoofers/

Link to comment
Share on other sites

Cascading limiters is only possible with some of the more costly amps I know (Powersoft, the new Crowns, Linea, Lab PLM etc.).

My Sanway D10 has a variable attack time, you'll be able to limit the signal to say 50% of the RMS rating after it has been anywhere above that level. By saying anyware, you'll already know that you might be sending 2KW into your sub for the duration of the attack time, which might already burn some smaller drivers.

If you have different limiters available, I'd set one with a quick attack (say 5ms) to cut off peaks (somewhere in the ballpark of 2-4x the rated AES power), one with a moderate attack time of around a second or two to cut off anything above 1-2x the rated AES power and one long term average which gets active after 5s and limits the driver to less than half the rated AES power.

I'm not an expert on this either, but I'm sure you get the point. Ideally you'd set up limiters to be transparent, and that's more or less possible with a peak limiter, but the long term average limiter might be quite noticable if you have a DJ pushing the absolute max with sine waves. You'll just notice it drop off at some but, but you will know that your limiters work as well 😉

As Ricci already said, the worst cooling a driver can get is at its excursion minimum. And when the amp is heavily clipping that's basically every frequency since the clipped since wave (aka square wave) means no movement at the peaks.

If you take a 1KW sine wave and clip it to the point of getting a square wave with the same peaks, you'll end up with a 2KW signal (the peak voltage of a sine wave is the rms voltage multiplied with the square root of two and every voltage increase squared is the power increase, so that's twice the power).

Link to comment
Share on other sites

The Behringer NX shouldn't clip the signal as its limiter is always in place and has zero attack.  What happens if if you request a signal output beyond what can be produced at the output stage is that the gain is reduced.  Of course the signal can be clipped earlier in the signal chain whereupon what has already been said about low crest factor waveforms is true.  Its also true that it can output 2 kW per channel continuously and the limiter could be heavily compressing the input waveform still.  Considering the power of the amp and the rating of the driver running the amplifier on its limiter for an extended period of time would result in driver failure, without any signal clipping,  -6dB from 2kW is still 500W which would eventually kill the driver.  

If you recone the driver yourself you could add in a temperature sensor or two to the voice coil former, or perhaps even a non contact thermometer if you can see the voice coil?  There are also a lot of papers on modelling voice coil temperature but they would tend to be time consuming to implement.    

Link to comment
Share on other sites

I've got a pair of keystrones with 18sw115-8 drivers, I'm generally running them with a driverack pa+ for filters/limiting and a sae pxm 1450 amp which, when paired with a robust power source, is quite a performer. I've been running very demanding techno music (Enrico Sangiuliano, Joseph Capriati and the likes, low crestfactor and continuous basslines) for hours to an end with appropriate filters, 30-100hz for that box, and a limited setting to 87 volts. I've even run them at about 95 volts long term limit for two hours during peak hour at full throttle. That would be equivalent of about 1150 respectively 1388 watts into that driver(when using the Volt x Volt / impedance minima formula) if I understand it correctly the 700w for the ipal21 Josh mentioned above are based on a much more conservative calculation method so these figures won't necessarily clash, there are many ways to calculate powerhandling, and they never complained or sounded bad, quite the opposite actually, and LOUD. I've run bfm t30 cabs very hard but within recommended limits for a few thousand hours aswell never blown a vc in them either..

The 21ds115-4 or 21ds115-8?

They have the same vc diameter as mine, and powerhandling on paper. The 8 ohm has the same impedance minima as my drivers(6.5 ohm) so I would use these long term voltage limiter settings with them that I use, but I've not been confident running them at 95 volts 99% of the time I set them to 87. The limiter resolution of my drpa with that amp is in 8 volts steps if I recall correctly.

That would be equivalent of approx 60 respectively 65 volts for the same power applied to the 3.1 ohm impedance minima of the 21ds115-4. Take it for what it's worth, I've always had the intention of extracting every bit of performance from my drivers and they never failed me:) but actually I'm a bit surprised I've never blown anything.

Not to say these settings will work for you, I guess limiters can work differently in different LMS for example, just giving my perspective of things.

Link to comment
Share on other sites

Well well..

Thank you all so much for your input!

I see that this is a complex topic which I really want to master one day.. 

Lots of different opinions and ways on this one.. 

I will try and implement limiting based on your suggestions/DSP capabilities ;)

Wondering how to set it though.. 400w long term seems so little and >1000w seems like a bit much but thats also what I thought the driver should be able to handle 

Also nice to hear a feedback on the temperature sensor idea, I thought of mounting a contact sensor onto the magnet to read out the temperature from time to time..

Maybe that'll be the safest bet for me as my DSP is limited in it's limiting capabilities and the opinions vary quite a bit ;( 

But I guess that's guesswork also as the temperature inside of the magnet is probably higher and I don't wanna find out how hot it gets before failing again.. 

I think I will try 3 different measures:

1. Never letting the signal be clipped by the DSP

2. Trying different limiters (short attack for kicks that are too loud, long attack for prolonged sinewaves) 

3. Adding a temperature sensor to the magnet that let's me read out the temperature from outside the box

 

Please keep on giving suggestions, I'm really soaking up input on this as it's very sad to not hear the beautiful skram bass in my living room anymore.. don't want it to ever happen again haha 

Also..

If there's anyone from the bass music scene that has experience in limiting, please let us know how to set a limiter without having a major drop of power during long notes.. if that's even possible? 

 

 

Link to comment
Share on other sites

Let's assume for a second that the 21ipal is as robust of a driver as the 18sw115 within it's given specifications (I'm not sure whether it is or not, it's an older design).

I'm considering limiting the 18sw115 to 67% of the nominal powerhandling to be safe for very heavy techno, as I've been running them with these settings for many hundreds of hours. I've even limited them to 81% of the nominal powerhandling for two hours and they did not sound strained during peak hour, would it survive that for more extended periods of use? I don't know. One of the reasons I'm not very keen on trying is that some aspects of the cabs are not very good built by me and they already dance around abit on the floor already at the 67% setting. 67% of 21ipal nominal handling is 1675w, or about 43 volts into it's 1.1 ohm minimum impedance. This it should be able to handle like a stroll in the park if it's as robust as 18sw115.

My guess is that this is a bit lower than the 700w rating Josh is talking about above, what the amp is permitted to give the driver and how much juice the amp draw from the outlet are two very different things

Link to comment
Share on other sites

The fact that Josh ran full power (23s long) sine wave sweeps of a SP-6000 into two IPALs and said that they were not complaining shows how bad a worst case scenario can really be. Run several tracks with focus on the excursion minimum and voilà, melted VCs.
That being said, I've run my FP13000 in bridged configuration at full power into two paralleled 21DS115-8 for several minutes to see how quickly they'd give out, but they didn't even complain. I had to clip the amp by a few db to actually get them to smell somewhat unpleasant. I turned it down and both drivers are still in excellent working condition. During these tests the amp was drawing a solid 3KW average power from the wall.

Link to comment
Share on other sites

1 hour ago, Murmel said:

My guess is that this is a bit lower than the 700w rating Josh is talking about above, what the amp is permitted to give the driver and how much juice the amp draw from the outlet are two very different things

I'm not sure if I don't understand but the figures you calculated are way higher (not lower) than 700w?

Am I missing something? Are we talking about sines coming from the amp?

Sry for being confused and thanks for caring and elaborating! 

Link to comment
Share on other sites

31 minutes ago, LSC said:

I'm not sure if I don't understand but the figures you calculated are way higher (not lower) than 700w?

Am I missing something? Are we talking about sines coming from the amp?

Sry for being confused and thanks for caring and elaborating! 

I'm not overly confident in explaining the differences but my guess is has to do with duty cycle and crest factor. When testing amps and voicecoils many use a test signal that somewhat reassemble music which is 33% duty cycle, where 100% duty cycle has much more power density hence the long term tolerable VC wattage is much lower. If you use VERY conservative calculations you use a very high duty cycle signal to measure tolerable long term limiting but the output will suffer and subs not reach their max potential is my take on it.

I'm not sure under what conditions b&c came up with the 400w figure for the ipal, what I do know is that by simply measuring the output voltage of the amp with a sinewave and dial the limiter accordingly I use four times the wattage probably due to much lower duty cycle signals.

Link to comment
Share on other sites

Quote

 

4.5 Power-Handling
4.5.1 Test Conditions and Equipment.

The lf driver shall be mounted in free air so that the direction of motion of the diaphragm is in a horizontal
plane and so that there is no appreciable air loading from adjacent structures. The driver shall be excited with a
band of pink noise extending one decade upward from the manufacturer’s stated lf limit of the device. The
noise shall be bandpass filtered at 12dB per octave with Butterworth filter reponse characteristics, and the peakto-
rms voltage ratio of the noise signal supplied to the lf driver shall be 2:1 (6 dB). Refer to Appendix C for
recommended method. The manufacturer shall state the upper and lower cutoff frequencies (– 3 dB) of the
noise signal.


4.5.2 Test Procedure.
The device under test shall be subjected to successively higher powers and allowed to reach thermal equilibrium
at each increment (approximately 2 h). Power shall be determined as the square of applied rms voltage, as
measured with a “true rms” voltmeter, divided by Zmin. The rated power of the device shall be that power the
device can withstand for 2 h without permanent change in acoustical, mechanical, or electrical characteristics
greater than 10%.

 

 

AES_Recommended_Practice_Specification_o.pdf

  • Like 1
Link to comment
Share on other sites

Oh, I see.. ..different test signals vs. Real life conditions.. 😐

I guess I need to work out a limiting scheme for music with snappy transients in bass and one for music with continuous low notes..

And the power consumption of the amp is a big one for me as well as I like to keep things efficient.. 

Such a headache, I wish there'd be a limiter that measures the changing resistance of the coil as it gets warmer and then adjusts accordingly.. 

Some kind of a feedback loop

 

 

Link to comment
Share on other sites

You could inject a low amplitude sine below a the audible band (~10Hz) before the power amps (IE in your DSP).  Then use a sound card on a computer to measure the driver impedance at this frequency using the pilot tone.  It would work best to use 3 sound card channels so you can recover the carrier, current and voltage and use coherent IQ demodulation to recover the signal.  Given sufficient integration time you should be able to obtain good results up to almost the full dynamic range of the ADCs.  

  • Like 2
Link to comment
Share on other sites

Okay sweet, easy does it! 

No just kidding sounds complicated but if that's a DIY approach of a feedback system then it's worth digging into.. 

Didn't read anything about measuring impedance with a soundcard but my soundcard has lots of ins&outs so channel count won't be a problem. 

How does one accomplish coherent IQ demodulation? 

Also, radulescu_paul_mircea, if you're reading this, what's your approach for limiting when it comes to prolonged bass notes on a regular basis? 

Link to comment
Share on other sites

Here is some MATLAB code in this case the exact frequency was known in relation to the sample rate and so the carrier is locally generated and didn't need to be recorded as the same sample clock is used for the ADCs as the the DAC that generated the carrier (thus only needing two ADC channels).  This was used in another impedance measurement device that used a similar method.  As you can see its quite a simple process and could be continuous unlike if you used Fourier methods, so doesn't need a lot of memory for long integration times.

function [ Z ] = CalcZIQ(V, I, Fs, F)
%CalcZIQ calulates impedance from the voltage and current results
%   by increasing the length of the voltage and current recording length
%   less energy is integrated into the result not from the test stimulous
%James Lawson 2018
%Inputs:
%V: voltage waveform at constant sample rate [V]
%I: current waveform at constant sample rate [A]
%FS: sample rate [Hz]
%F: frequency of interest [Hz]
%Outputs:
%Z: complex impedance [ohm]

%Trim off trailing zero values (caused by result storage method sometimes)
%do this by finding first non zero value
index = length(V);
while V(index) == 0
    index = index-1;
    if index == 1
        break;
    end;
end;
if index ~= 1
    V = V(1:index);
    I = I(1:index);
end;

%Generate sine waves at the sample rate to demodulate the data
TimePoints = (0:index-1)/Fs; %Sample times
demod = sqrt(2)*sin(2*pi*F*TimePoints); %Demodulation waveforms
demod90 = sqrt(2)*sin(2*pi*F*TimePoints+pi/2);
    
%IQ modulators
ReVmult = demod.*V';
ImVmult = demod90.*V';
ReImult = demod.*I';
ImImult = demod90.*I';

%Low pass filter to remove 2*f component 
PassbandFreq = Fs/length(V)*10;
lpFilt = designfilt('lowpassiir','FilterOrder',8, ...
         'PassbandFrequency',PassbandFreq,'PassbandRipple',3, ...
         'SampleRate',Fs);
%fvtool(lpFilt)
ReVmult_filt = filter(lpFilt,ReVmult);
ImVmult_filt = filter(lpFilt,ImVmult);
ReImult_filt = filter(lpFilt,ReImult);
ImImult_filt = filter(lpFilt,ImImult);

%integrate, there is not need to do division step as all parts equaly
%effected
ReVacum = sum(ReVmult_filt);
ImVacum = sum(ImVmult_filt);
ReIacum = sum(ReImult_filt); 
ImIacum = sum(ImImult_filt);

%Calculate final impedance
Z = (ReVacum+1i*ImVacum)./(ReIacum+1i*ImIacum);

end
 

Link to comment
Share on other sites

On ‎8‎/‎3‎/‎2019 at 9:24 AM, Murmel said:

I'm not overly confident in explaining the differences but my guess is has to do with duty cycle and crest factor. When testing amps and voicecoils many use a test signal that somewhat reassemble music which is 33% duty cycle, where 100% duty cycle has much more power density hence the long term tolerable VC wattage is much lower. If you use VERY conservative calculations you use a very high duty cycle signal to measure tolerable long term limiting but the output will suffer and subs not reach their max potential is my take on it.

I'm not sure under what conditions b&c came up with the 400w figure for the ipal, what I do know is that by simply measuring the output voltage of the amp with a sinewave and dial the limiter accordingly I use four times the wattage probably due to much lower duty cycle signals.

Correct. Duty cycle has everything to do with it. What might be fine with light amp clipping on 3hrs of sporadic kick drum or HT playback, will probably not be with bass test tones which some of the EDM and other similar genres are pretty much using in some cases.

The 400w recommendation is not the B&C rating. That's from the Powersoft Ipal module literature. The list it as the default setting and somewhere else in their literature for it there is a 700w max recommendation for the 21. That rating is way different from AES testing or pretty much any other that companies are using to rate drivers. That is what we would call a WCS rating. Powersoft offers an actual power limiter that is based on the real average power output of the amp in addition to voltage and current limiters. Most other units do not have this type of limiter.  

Most of us probably know this stuff but it bears repeating for anyone reading along. A sine wave has a 3dB crest factor, but music often has a crest factor over time of 12dB or more. If the average power of the sine wave is limited to 400w and the 12dB crest factor music signal is limited to the same average power, the peak output requirement from the amp is 9dB higher for the music signal vs the sine. While you may be clipping a 3200w amplifier on the signal peaks the true average power to the driver could be under 400w at the same time. In fact most of the time this is how things operate and how drivers survive connected to amps capable of thousands of watts of power. It's all about how much time.

The 23 second long measurement sweeps I use do occasionally kill a driver due to excursion or a burnt coil. A 23sec sine wave at the maximum sweep level I use is very tough but at the end of the day it's only 23 seconds and most drivers have no problem with it once or twice and are allowed to cool off immediately afterward. If I were to loop that same measurement signal 13 times in a row it would burn up the drivers in under 5 min. Once we are talking about hours of playback at high volume the drivers have a chance to fully heat soak the motors and coils and all it takes a couple of low crest factor or higher than average level signals to toast the already hot coils.

With all that said I've found most of the modern pro drivers to be quite resilient and durable including the 21DS115, so I'd have to assume the OP's driver was taking quite the beating when it burnt up.

  • Like 1
Link to comment
Share on other sites

I am reposting a line of comments I made about this subject on a group on Facebook

The limiters are very safe but you have to take care! I use the clip limiter as a 0 time attack, infinite to one ratio peak limiter, sometimes I let soft clipping disabled. This is a voltage limiter and it should be set a bit higher than the peak power of the speaker, or about 6-8 dB over AES RMS power. You will have to use V=✓(P*I ) to find what voltage you need. Next, the peak limiter is set with a time constant high enough to let the transients pass without problems, but low enough to protect the drivers from over excursion. Take notice that this is working as follows: if a signal is stronger than the Threshold, the limiter is engaged instantaneous and the attack time is the time it passes until the signal is reduced from it's original value to the Threshold value. I am setting this one with a time constant =1/2 of a single oscillation at the frequency where the excursion is the greatest in the bandpass. So if the excursion is max at 50 hz, I put 1/50/2=10 ms. Some are using the frequency at the loads but sometimes this is not good because for example in a reflex enclosure, the peak excursion I'm the octave above port tuning is exactly one octave from High pass frequency and this means it will let it go over once before engaging. The release is a multiple of the attack , 10-20x for lows, 4-8x for mid etc but you'll have to play with it cause it will sound good on some types of music but not on the other...  The voltage for this one is based on the way the driver is loaded in the enclosure and the maximum excursion allowed, but if you don't know that, you'll need to put it between 3-6 dB over AES power.. The TruePower setting is for thermal protection. Depending on the type of music you are playing the settings are different. For long duration bass lines ( dubstep, trap etc) , put 1/4 of the AES power with a longer time constant, soft knee disabled (3-6 seconds atk, 12-15 s release for 4.5 " 1700 Watt voice coil)  .it will be louder for longer but then it will limit harder. For techno, trance , dnb etc, you need to put a higher threshold 1/3 - 1/2,5  but with a lower attack time (0.5-2 s atk, 2-4 s release). It is important to know that the limiters are safety nets,  not meant to be engaged all the time. The are there to save your equipment when a part of a song is a tad too loud, or you need a moment where you increase the level for Dynamics. But if they get engaged all the time, you need more equipment. The gain topology you choose has to be set so the limiters are not interfering with the music too often.

The TruePower limiter is a setting which is using the real power transferred to the driver by measuring the real voltage and current ( disregarding the reactive part) so it has to be set in relation to the number of drivers linked in parallel on the same channel. If you have a 1 KW AES driver you will set it at 250 W, if you put two of them on the same channel, you set it at 500. The peak limiter is a voltage limiter and it is independent of the number of drivers linked in parallel. The same applies to Power vs V@ 8Ohms.
When I am setting the TruePower Threshold, I am usually also measuring the magnet  temperature and voicecoil reference impedance and I try to keep the temperature rise under 150°C and the magnet under 80 degree after 2 hours of low crest factor music. Another thing I am doing sometimes , especially when I have the time , space and conditions (so not many times now) is to measure the acoustic power a subwoofer is generating. I am putting pink noise in band with all filters and processing activated  and I turn the gain until the TruePower is measuring 10-20 Watts on the output. Then I measure the acoustic power of the device with a method , I compare it to simulations in Hornresp and Akabak and I calculate the Input/Output mean efficiency. If it has a mean efficiency of let's say 10%, I'll add that power to the True power Threshold , because that energy will not be transformed into heat ,but in Sound. Another thing is that when you couple more subs , they increase in efficiency so less power is turned into heat. This means I can increase the threshold once more.
I got my hands on a some Klippel LSI reports from some drivers made by some well known manufacturers and there Delta Tv is the voice coil temperature rise from base which can get as high as +180 °c . So the drivers voice coil temperature can easily get to 200° C but not the magnet's temp. That one I like to keep it under 80° . If the temp is of the coil is kept under 170 ° at all times , the Re of the driver is kept under a certain variation and the power compression stays under 3 dB .


There is something I would love to see more in the processing units: a limiter with a frequency dependant Threshold. A limiter with an output Threshold and given attack and release constants that are not engaged ,even if passed, if the frequency is not in the right band. Powersoft does have this,  I think Linea Research has something similar and I've seen this behavior in Eminence D-Fend.

This type of limiting is letting one to better protect the drivers from over excursion in subwoofers. If you know the impedance graph (which correlates to the excursion vs frequency) one can set the limiter to trigger earlier in a certain band. In the previous example, a reflex enclosure has its second impedance peak at 50 hz, where the excursion is the greatest, the FS is at 38 and at 33 is where we need a crossover to keep the excursion under FS at the same maximum level as it is at 50 hz. If there is this type of limiter, we can increase the crossover frequency and we can put a higher sensitivity/lower threshold in the 50 hz band. If the amp allows it and the driver doesn't melt, one could increase the level in the rest of the band a lot. Off course, the frequency response equilibrium would be a problem but that would be another topic. This same type of limiter helps the Powersoft amps to be louder and safer than anything else on speakers with passive crossovers.
 

Link to comment
Share on other sites

This is my first post to this thread, so let me first offer condolences to the OP.  That's a major bummer!

As is apparent by the posts here, this subject is very complicated.  Something I don't think is emphasized enough in this discussion (at least directly) is the importance of impedance, which varies with frequency.  Impedance describes the relationship (both magnitude and phase) between voltage and current in the driver.  The amount of power turned into heat in the coil depends on the coincidence of voltage and current.  That is, instantaneous power equals instantaneous voltage multiplied by instantaneous current, but to the extent voltage (V) and current (I) are out of phase and don't peak at the same time, the real power consumed (or turned to heat) by the coil will be less than Vrms * Irms.

A reason why impedance is so important is that it varies substantially vs. frequency in a way that depends on the particular driver and cabinet design.  Frequency dependence means that actual power may be very different when playing two different music or test signals, even if they have the same duty cycle.  For example, an efficient vented cabinet may consume very little power throughout its range *except* near its tuning frequency where impedance is very low and power consumption may rocket way up.  This means that such a cabinet might do just fine being overloaded with music for many hours until you come across a track that has a strong sustained drone right at the tuning frequency.  This is a doubly-bad scenario.  It's bad because the coil is taking way more power than it was with other content, and it's bad because (as mentioned earlier here) the tuning frequency is also an excursion minimum which prevents the driver from cooling itself.

Now consider that all of the above also applies with AES testing, which is performed with the driver in free air.  In fact, the reported "AES power" is not really power at all.  It should perhaps be called *nominal power* because it's calculated directly from the *RMS average voltage* used for pink noise test signal by dividing it by the manufacturer specified *nominal impedance*.  The pink noise test signal is band-limited to the range 1X to 10X the manufacturer's specified low frequency limit.  To estimate the *actual power* consumed by the driver using the AES signal, one can sample the *real part* of impedance curve at *equal octave intervals* (i.e. 1/3rd octave or 1/6th octave, the finer the better the estimate --- because *pink noise has equal energy per octave*) within the test signal bandwidth.  Each sample should be divided into (Vrms^2) and the results averaged.  This works in principle, but in practice, the *real part* (or "in-phase" part) of the impedance is rarely plotted separately on data sheets.  (Perhaps it should be available to plot on Data Bass?)  The real part is always less-than-or-equal to what's normally plotted.  An alternative method of calculation would be to simulate the driver in a very large sealed enclosure in Hornresp (with proper semi-inductance parameters) and sample and average the "driver power" over the test signal bandwidth.

As an example, consider the DS115-4 (4 ohm nominal).  The AES power is 1700 W, which simply means that the driver passed the long-term test when operating in free air using a pink noise signal at around 82.5 Vrms average (calculated from Pnom = Vrms_avg^2 / Znom).  The specified frequency response lower limit is 30 Hz, so the test signal bandwidth is 30 to 300 Hz.  I used Hornresp as described above and sampled at 17 different points.  The average I calculated is around 950 W, which is nowhere near the AES power.  Keep in mind also that this test signal will cause a fair amount of excursion, which will help push air through the driver vents to cool it, and the free-air mounting allows much more ventilation than is typical inside of a cabinet.  Unfortunately, these variables change a lot between drivers and cabinets, so even this number (950W) doesn't tell us much about real world worst-case-scenario power handling.  Surely it will be quite a bit lower still.

Either way, one must consider that vented and BP6 cabinets (like Skhorn and Skram) will have frequency regions of particularly low impedance, which if hit with a high duty cycle narrow-band signal (as may occur in various EDM), could result in the driver seeing something like 2700W (!!) with the same 82.5Vrms average input used in the AES test.  So if one wishes to avoid putting more than 950W average actual power into the driver in a WCS for signal input, one must limit output to a merely 48.7V, which is only 600W nominal power.  And as noted above with regard to cooling capability, that 600W nominal may still be too much power.

 

  • Like 1
Link to comment
Share on other sites

I'm pretty sure I could make a device that could plug into the inputs and outputs of most sub amps and measure the voice coil temperature while the drivers in use.  Sounds like there would would be some interest in the device.  I have a spare voice coil and have ordered some other bits to test the method so should be able to report back next month on my progress.  I don't quite understand why amplifiers don't implement voice coil resistance sensing though as it would be very easy to inject a small DC offset and measure using that, DC protect could just be set to a higher level than the small offset. 

Link to comment
Share on other sites

One risk of high long-term average power that I don't really see discussed in speaker forums is the fact that Nd magnets lose magnetic strength at a rate that depends on temperature.  While they have to get pretty hot (like 350 deg C) to demagnetize rapidly, slower losses can occur with much lower temperatures.  I believe details depend on the particular grade of material and the shape, but I believe there's reason to be concern about wear if operating with magnet temperatures above 150 deg C for sustained periods.

Link to comment
Share on other sites

On 8/7/2019 at 3:06 AM, radulescu_paul_mircea said:

There is something I would love to see more in the processing units: a limiter with a frequency dependant Threshold. A limiter with an output Threshold and given attack and release constants that are not engaged ,even if passed, if the frequency is not in the right band. Powersoft does have this,  I think Linea Research has something similar and I've seen this behavior in Eminence D-Fend.

Would be pretty easy if you happen to have a mixing console with a WSG server. The Waves F6 could act as your frequency dependent limiter, say a quick attack limiter for over excursion set at the 2 frequencies you'll need it at using a BR enclosure, and a slow attack long release limiter at the excursion minimum which sits a little lower than the amp's long term power limiter to prevent heat buildup from missing ventilation.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...