RF7-II tweeter issue

[Scrappydue]

I think the excuse they gave is bogus

that is funny. there is another member i am in contact with right now and he blew both rd-82 tweeters and called klipsch. they are sending him new units as well but told him the same thing.

OP. how loud were you listening when you blew them?

[Deang]

"Clipping" means you've exceeded the specified operating limits of the amplifier. Instead of clean power, you are now delivering distorted energy to your drivers. What is "clipped" are the waveforms, which are now more like square waves instead of sine waves. This is hell on voice coils, especially the kind wound with 40 gauge magnet wire.

I used to run my RF-7s with everything from 8 wpc (300B tubes) to 200 wpc (Bryston). 125 wpc should drive them to insanely loud levels, unless the room is huge, or the user has hearing loss.

Try to remember that doubling the power only gives you 3dB of extra loudness.

lowpower.pdf

Edited May 9, 2014 by DeanG

[derrickdj1]

This sounds like what I said. I didn't mention the square wave which is where the DC current comes in and leads to overheating of the voice coil. Thanks Dean!

Scrappy, I have turned my avr up to 0, the avr does go to +12 so I guess you could blow a tweeter.

[pzannucci]

The clipping issue might be an old argument but why would you suppose cranking up the bass to amplifier exhaustion and cranking the volume will toast tweeters? I've seen it happen multiple times.

[Guest Steven1963]

I just read that article from JBL and it sounds like they are saying harmonics are the culprit when/if bass is overdriven in an amp. i.e., if the overdriven amp is producing distortion from a base guitar, it can damage the tweeter, because of the harmonics. If I understood it correctly. Very interesting.

Thanks Dean for the PDF.

Edited May 9, 2014 by Steven1963

[garyrc]

This one is an old argument that always lead to differing opinions.

Indeed.

Someone should run a controlled experiment to see if clipping (due to using a low powered amp) will take out a vulnerable tweeter with a relatively low power handling capacity (like the K77), but not take out the same make & model tweeter that is part of a speaker system operating at the same SPL, but without clipping (due to using a high power amp). Of course, since tweeters of the same make and model vary, this should be tried with several exemplars in both the low powered amp & the high powered amp group.

Any volunteers?

Until I know more, I'll try to avoid clipping, square waves, and very loud kazoos.

Edited May 9, 2014 by Garyrc

[Deang]

I'm pretty sure DC doesn't enter into the picture here, we're in the realm of AC.

http://www.learnabout-electronics.org/ac_theory/ac_waves01.php

Bass harmonics will be blocked by the network, which won't allow those frequencies to pass.

All I know is that if you send distortion to your tweeter, the VC will burn up.

[pzannucci]

I'm pretty sure DC doesn't enter into the picture here, we're in the realm of AC.

http://www.learnabout-electronics.org/ac_theory/ac_waves01.php

Bass harmonics will be blocked by the network, which won't allow those frequencies to pass.

All I know is that if you send distortion to your tweeter, the VC will burn up.

I would believe so but I am still trying to figure out why my ribbons are flapping in the wind when I get feedback. I have cranked them to high heaven and they still are corrugated (Aurum Cantus G1). If I let feedback occur which seems to be no louder than when I crank them (minor mishaps when doing karake), the ribbons are stretched out and just flapping. With a passive crossover, you would think DC wasn't an issue but this makes me think it could very well be. This puts me into the tweeter moves and over heats in that position such as square wave or DC(which doesn't 100% make sense). In other words, you got me....

[djk]

"I would believe so but I am still trying to figure out why my ribbons are flapping in the wind when I get feedback. "

Mechanical damage, too much low frequency.

" With a passive crossover, you would think DC wasn't an issue but this makes me think it could very well be. "

How does the DC get past the capacitors, and then through the built-in transformer?

" This puts me into the tweeter moves and over heats in that position such as square wave or DC(which doesn't 100% make sense). In other words, you got me....""

Install a lightbulb.

[derrickdj1]

In my earlier statement, I should of said the DC component of the sinusoidal wave.

The DC component would be the constant voltage added to a pure AC waveform. For example, the true average voltage a pure AC waveform would be zero. When the AC waveform has a DC component, the average voltage would be equal to the DC voltage instead, because oposite peaks cancel each other leaving only the DC component. http://forum.allaboutcircuits.com/showthread.php?t=7210

The is the part of the wave that if clipped provides high power to the signal in even and odd ordered harmonic that lead to little movement of a voice coil and the build-up of heat. This can lead to tweeter distruction.

Edited May 10, 2014 by derrickdj1

[ricktate]

Did you get your tweeter fixed yet ? How did you get it out another poster here has same problem. Rick

[djk]

"In my earlier statement, I should of said the DC component of the sinusoidal wave."

BS.

"The is the part of the wave that if clipped provides high power to the signal in even and odd ordered harmonic that lead to little movement of a voice coil and the build-up of heat. This can lead to tweeter distruction."

BS.

" when I get feedback"

Sorry, you don't seem to have any common sense. Even if you had an amplifier with a hard limiter that could never clip, this would still destroy a tweeter in less than one second.

Install a lightbulb (or buy a bushel basket of replacement ribbons).

Music waveforms are asymmetrical, that would be DC off-set as per your theory.

Loudspeaker off-set can destroy woofers (with a poorly designed amplifier), not HF drivers.

[derrickdj1]

Let's here a detailed answer in what exactly damages a tweeter. If possible, keep it a civil discussion. I will not restort to putting you down because we disagree. At the start of this I said there would be differing views.

[djk]

"I am interested from the academic point of view. Also i might require to make for 2" compression driver. This will also increase my knowledge about the subject. "

"I am also waiting 4 ur details/new thread about the klipsch protection circuit"

OK, let's keep on with it here for a while.

Drivers fail for two reasons:

Excess average power
Mechanical damage

Clipping will not hurt anything if it does not cause the above problems.

After larger amplifiers (above 50W) became common, problems with high frequency units self-destructing became a problem.

A tweeter like the EV T35/Klipsch K77 could handle 5W continuous, 50W for 10mS peaks, so normal program material wasn't the problem.

The tweeters were dieing from mechanical failure. The voice-coil was wound with aluminum wire that ran out to the terminals on the frame (self-termination). Shallow slopes in typical crossovers (6dB) combined with higher drive levels produced failures. Copper wire was tried instad of aluminum (aluminum work-hardens in a very short time).

Still, the failures continued.

Klipsch switched to the 18dB crossovers in the very early 70s. This helped with the excursion failures.

Now, 100W+ amplifiers became common.

Klipsch went to cathode-to-cathode connected zener diodes. These clipped off the peaks that were mechanically destroying the tweeters. Things went well for about ten years.

In the early 80s it was time to try and get rid of the expensive band-aid (the zener diodes). EV changed the lead-out wire from the self-terminated copper voice-coil wire to a flat BeCu wire like used on expensive JBL and Altec type compression drivers. A new network was designed with an elliptical filter with 50dB of attenuation only a half-octave away from the crossover point. A fast- acting instrumentation fuse was added.

Problem solved?

The new version of the tweeter used a UV cure adhesive vs the old thermo-set adhesive. After the fuses blew from modern program material (about the same time as the introduction of the CD), they got replaced with fast-blow types (which offered reduced protection). The special instrumentation types were very expensive, and very hard to find.

The new UV cure adhesive got soft, bubbled, and failed quite easily.

A PolySwitch was tried. Too slow.

The old thermo-set adhesive came back.

The combination of the super-steep crossover, the flat BeCu lead-out wire, and the PolySwitch seemed to work (with the old adhesive).

The zener diodes were retired for mainly two reasons: cost, and limited dynamic range. The AA networks used a pair of 5.1V 10W zeners. These only allowed about 2W RMS through before they started clipping off the peaks (a 4W peak square-wave) .

With the advent of digital program material, 2W of undistorted program material no longer seemed adequate (about 97dB at 10 foot). The zeners limited the maximum distorted output to about 100dB at 10 foot.

Removal of the zeners allowed exploitation of the 50W/10mS rating of the tweeter, about 14dB more output capability (referenced to the 2W RMS zener clamp).

What did EV do for products sold under their brand?

The STR tweeter protector was developed for this use. Later it was modified by adding a lightbulb in parallel with the relay contacts.

I hope this give a little insight into what is needed for tweeter protection, and how we got to where we are.

Vifa, Dynaudio, and others offer a choice between self-terminated tweeter lead-outs and a braided (or tinsel) type lead-out.
The difference in cost for the braided type is worth it in my book. Most manufacturers of lower-priced product do not spend the money for this, or better crossovers either (although I am seeing lighbulbs and/or PolySwitches in some inexpensive product).

Crossovers and tweeters must be designed to avoid mechanical damage in normal use (correct slope, frequency, and lead-out wire for the intended use).

Long-term thermal protection is worthwhile. Some sort of switch device (relay, PolySwitch, fuse) in conjuction with a lightbulb seems to be the most cost-effective. Due to the long time constant of lightbulbs, most use does not seem to demand the switch (which shorts out the lightbulb in normal use).

Suggested current levels:

For 1" coils (tweeters), about 1A. The 561 or 211-2 automotive lamp has worked well in this application.

For 1-3/4" coils (1" compression drivers), about 2A. The 1156 type automotive lamp has worked well in this application.

For 3" (kapton) to 4" (nomex) coils (2" throat 16 ohm compression drivers), about 2A. I have found a 1.5A AGC (or 3AG) type fast-blow fuse will pass 400W of program material in normal use, and blow almost instantly if bad feedback is encountered. A pair of 1156 automotive lamps wired in series with each other, and then wired in parallel with the fuse will allow the show to go on when the fuse blows, and still offer some protection. If the fuse blows in the course of normal use, you need additional HF drivers, horns, and amplifiers.
__________________
Candidates for the Darwin Award should not read this author.

"As for using a L pad attenuator the idea of this is to maintain the same Resistance - Impedance as the Driver ie 8R0 to allow the crossover components to correctly filter the frequency. So how does this increase the loading? "

Is this a trick question?

It would seem obvious that the size of the lightbulb would need to be changed for every different attenuation value (if it's to be of any use).

"So a 500W amplifier will deliver over 700W power to the driver if clipped hard enough."

Really? Power goes with the square of the voltage, so it goes from 500W to 1KW, but that has little or nothing to do with why the drivers blow.

Clipping can damage speakers, but not for the 'urban legend' reasons.

"If you’re getting the idea I don’t believe in the clipping/
harmonic theory, you’re right. So let’s investigate the
phenomena further.
WHEN SINE WAVES CLIP
When sine waves clip severely they resemble square
waves in shape, introducing massive distortion. In the
extreme case, a perfect square wave has the highest level of
harmonic components (See Figure 1). A less clipped sine
wave has components at the same frequencies but at lower
levels.
Let’s look at the square wave example shown in Table 1
(at left). Fourier analysis shows the harmonic structure.
As you can see, the total amount of instananeous power
left to make it through an ideal 1kHz crossover (and on to the
tweeter) is less than two watts (0.83 + 0.589 = 1.419W).
Hardly a problem. And remember, this simulates severe
overdrive of a 100 watt amplifier with a sine wave to make an
ideal square wave. Driving it harder will not increase the
harmonics.
This analysis shows if a small tweeter that only handles 5
or 10 watts is used in a 100 watt speaker system it would not
blow out, even under square wave conditions. Yet it does.
It takes a lot more than this to cause major failure. "

Consider a worst case situation, an amplifier overdriven by 10dB or so.

The bass content of the program material causes the amp to clip, the excess HF content from clipping harmonics fed to the tweeter is trivial.

However, the average level fed to the speaker at mid and high frequencies is elevated as much as 10dB (due to the input overdrive) during non-clipped passages.

WOOFER MECHANICAL DAMAGE
(assume the design has sufficient x-max for in-band signals)

It's not the clipping per se, but the poor design of the amplifier causing this. The pole in the feedback loop is generally poorly chosen and lower in frequency than the input pole, and above the power supply/load pole, so the poorly designed amplifier pumps out rail-to-rail signals at the feedback pole every power supply time constant. Good design reduces or eliminates this problem (although it seems like 99% of amplifiers have this problem).

The easiest way to fix this is to change the poles so the feedback pole is at a higher frequency than the power supply pole, and the input pole is higher than the feedback pole. It also helps to put clamp diodes across the feedback cap.

Another way to fix this is to put an input clipper on the amplifier. NAD amplifiers have this kind of a circuit (they call it 'soft clipping') and it tracks the supply voltages so you get the maximum power the amplifier is capable of. Since the input clips before the feedback loop clips, it's always stable. I used to put 50W NAD amplifiers in clubs with four Klipsch Heresy on them, you wouldn't believe how loud this combo will play driven a few dB into clipping.

"2) Many crossovers barely handle the rated power of a speaker system, let alone the effect of a clipping amplifier. When inductors saturate, the tweeter gets not only the normal signal at high power, plus the distortion products from the clipping, but also back EMF from the saturating inductors and signals below the normal crossover frequency because the filter system in overload."

Only the cheapest of speakers use low quality parts these days. Back-in-the-day a network needed the tweeter cap replaced (if it was an electrolytic) or the tweeter inductor (if the tweeter cap was a film type) after the tweeter burned out and the operator continued to play the system. In 30 years I can pretty much count on one hand the woofer inductors that failed in hi-fi use, these all developed shorted turns (causing damaging infrasonic woofer excursions). In pro sound I saw quite a few woofer inductors fail from too high of a DC resistance (causing the insulation to burn off the wire and the inductor to short out).

The last KW is the most important, how it behaves at the limit.
A well behaved amplifier with no dynamic off-set when driven 6dB into clipping sounds much better than a wimp amp with a built in limiter, even if it is twice the rated power.

(This is tiresome as I have explained it many times before. The above was a compilation of many old postings of mine, I hope it is intelligible)
__________________
Candidates for the Darwin Award should not read this author.

[derrickdj1]

Causes Of Tweeter Damage

Most loudspeakers contain a HF and LF driver. Under normal condition the speaker functions as intended. But, there are cases where drivers fail, notably the tweeter for this discussion. Why.

Some possibilities are: 1.) clipping 2.) to high of average power to the tweeter 3.) Leakage of LF through the XO resulting in higher than normal power to the tweeter 4.) Mechanical damage from tweeter components overheating leading to increase displacement and damage.

Unlike woofers, tweeter are smaller, lighter and can reacat faster to reproduce the HF of a loudspeaker. Due to their physical properties, they are more easily damaged than woofers in general term. Music, movies and game played on loudspeakers have a more complicated wave form that is not sinusoidal and is asymmetrical in nature.

There are different types of clipping: signal exceeds rail voltage-voltage clipping, current exceeds current limit-current clipping, Any form of clipping means that the output is not the level of the intended input. This results in: 1.) increase THD 2.) increased average power 3.) ineffective heat dissapation across the voice coil 4.) unintended harmonic. Most speaker cannot survive a long peroid driven by a sinusoidal wave at their rated power, let alone a square wave.

A square wave dose not necessarily cause a speaker to be damaged. It depends on what spl the square wave in imposed on the loudspeaker. In examining a square wave of a clipped signal, the area underneath the clipped portion is larger than the non-clipped portion of the wave and has more power and energy. This extra power can lead to tweeter damage. This damage can occur in the tweeter or woofer or, XO due to overheating. It is therefore, plausible that clipping can lead to HF driver failure.

LF power leak through the XO may also go to the HF driver increasing the total power that the driver has to handle. This can result in excessive heat leading to increase displacement and lead to mechanical damage to the driver.

It is likely that tweeter damage if multifactorial and not limited to one cause.

In general, therefore, it seems quite plausible to assume that damage may sometimes occur due to any one of the following mechanisms, or various combination of them:

• HF creation by clipping of the LF.

• Rise in the HF level of the original signal which – although limited by clipping by an amount similar to that which occurs at LF – rises to a high level.

• Leakage of LF power.

• Non-themal effects due to an increase in the peak displacement or accelerations.

The main conclusion is that there is no obvious reason indicated by the analysis why we should either rule out any of the above factors entirely, or to assume that only one of them is always (or usually) a sole or primary cause. Which is in practice a more common cause, or whether they tend usually to act together, would be hard to say without more information on typical systems and musical waveforms in order to do a relevant statistical analysis, etc. http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/clipping/page6.html

In conculsion, lack of an open mind can lead one to get the Darwin award. A bumptious, gail attitude does not lead to a healty disccusion of the topic and does not promote learning which is an integral part of this forum. Being respectfull of others is one of the things that make this a great forum and is a quality that most forum members display.

Futher disccusion can be done via PM as not to derail or clip this thread.

Edited May 15, 2014 by derrickdj1

[djk]

"In conculsion, lack of an open mind can lead one to get the Darwin award"

Incorrect.

However, the D-K Effect may apply to you.

The Dunning–Kruger effect is a cognitive bias in which unskilled people make poor decisions and reach erroneous conclusions, but their incompetence denies them the metacognitive ability to realize their mistakes.[1] The unskilled therefore suffer from illusory superiority, rating their own ability as above average, much higher than it actually is, while the highly skilled underrate their abilities, suffering from illusory inferiority. This leads to the situation in which less competent people rate their own ability higher than more competent people. It also explains why actual competence may weaken self-confidence. Competent individuals falsely assume that others have an equivalent understanding. "Thus, the miscalibration of the incompetent stems from an error about the self, whereas the miscalibration of the highly competent stems from an error about others."[2]

Kruger and Dunning proposed that, for a given skill, incompetent people will:

A)tend to overestimate their own level of skill;

B)fail to recognize genuine skill in others;

C)fail to recognize the extremity of their inadequacy;

D)recognize and acknowledge their own previous lack of skill, if they can be trained to substantially improve.

Dunning and Kruger were awarded the Nobel Prizes in Psychology in 2000 for their report, "Unskilled and Unaware of It: How Difficulties in Recognizing One's Own Incompetence Lead to Inflated Self-Assessments".[16]

__________________

Edited May 16, 2014 by djk

[djk]

I see your copy-and-paste agrees with my original statement of:

"Drivers fail for two reasons:

Excess average power
Mechanical damage

Clipping will not hurt anything if it does not cause the above problems."

Do you even understand what your cut-and-past says?

Did you understand the cut-and-paste from Rane (about the FFT of the power going to the HF) that I used?

[derrickdj1]

Your original post stated that clipping could not cause driver failure. You have been educated. I never said anything about the rest of your post. It has been a pleasure helping wipe out ignorance. You know, I like you! You have spunk. Have a great day Darwin!!!

Edited May 16, 2014 by derrickdj1

[djk]

Your total lack of comprehension is amazing!!!

[Zen Traveler]

Fwiw, I don't understand why original internet sources aren't linked to when trying to defend ones position...