March Audio Sointuva

[RickS]

Remember to click " I'm not a robot"

Wg version looks best to me https://hificompass.com/en/projects/2-way-systems/puri-bliss-bewg

Yes, already past that, as did get a registration processing confirmation in Russian.

 

[Phorize]

Me too, but the site admin must have bigger problems to address as (despite trying) never seem to be able to get a registration to go through.

There is more info posted on the waveguide version, but seems to be concealing design details as part of IP protection.

Just dropped them a line asking if there are plans to release the plans or a kit.
Fingers crossed.

 

[Phorize]

Yes, already past that, as did get a registration processing confirmation in Russian.

Me too, but nothing so far.

 

[Hemi-Demon]

Rick Craig passed last weekend.

RIP, that's very sad news. Hope the son can keep the business going in honor of the father. Maybe someone can purchase the IP and keep the projects going. Maybe ASR audio - forum group funded

 

[Mikakimi]

The tweeter is not rated at "80 watts thermal" with a 2.5kHz 2nd order. That is the system power, when used with a 2.5kHz 2nd order. And that rating is the 28minute "long term". The tweeter may see just a few watts of that after padding down. But at a genuine 250W, there's >5-10W (even after padding down) to get rid of and you and I know a tweeter of that construction will be unlikely to survive that. We are talking a CCA VC of 1.4mm in two layers and 4R. Not a lot of thermal mass in copper wire to absorb and disipate indirectly to the magnet structure.

As a company like Scanspeak specifically articulates in their data on power handling:
View attachment 150171

And this:
View attachment 150172

Just as the Purifi woofer has a 100hr rating (IEC) of just 80watts. The 250W is a 28minute IEC (33% duty cycle test).

Anyway, unless someone actually tests these speakers in accordance with the IEC spec advertised, nobody will know. That is, until they vaporize a tweeter on an otherwise benign transient or dynamic passage.

Typically tweeters on TOTL tweeters (think berrylium, titanium, diamond, alumina etc) had ratings of ~10W. Having dismantled literally many hundreds of tweeters over the years, 10W is pushing it. The VC wire can be thinner than a human hair. Or the flat, edge wound wire can only be seen under a microscope.

Here's a TOTL JVC Laboratory Series Diamond tweeter voice coil which has an enormous 1.5kg magnet structure. It is flat edge wound 6R and rated at 10W. Even 10W is pushing it as the lead in wire is 0.2mm total diameter litz copper- before it terminates to this single layer coil of only 1.6mm in total windings height.

View attachment 150175

The calculations are pretty simple to follow and have already be presented. They are basic and do not take into account all factors, but demonstrate how little power is going to the tweeter. There is *no risk at all* of the tweeter being damaged at the speaker max rating of 250 watts. However it should help you to understand if the calculations are run through again.

The T34B tweeter power handling is 80 watts iec 268. Its sensitivity is 97.5dB/2.83v/m. Nom impedance is 4 ohms
Woofer sensitivity is 88.7db/2.83v/m. Nom impedance is 4 ohms.

Power is voltage * current.
Therefore to run the woofer at 250 watts the input voltage from the amplifier will need to be approx 32 volts.
32v / 4 ohms = 8 amps
32v * 8amps = 256 watts

As shown above the tweeter is significantly more sensitive than the woofer. As such (when you include baffle step compensation) its input signal voltage needs to be reduced by approximately 12dB to bring its output volume in line with the woofer.

So at the woofers max rating of 250 watts (32 volts input) the voltage applied to the tweeter is reduced by 12dB.
voltage gain in dB = 20 × log (V2 / V1).

So what power is being applied to the tweeter?

8v / 4ohms= 2 amps
8 volts * 2 amps = 16 watts

There is zero risk of this tweeter being damaged even with a low crossover. Its running nowhere near its capacity of 80 watts.

The above is assuming a consistent energy level between low and high frequencies. A secondary consideration is that music does not have a consistent energy level across the frequency range. The energy falls significantly with increasing frequency.

This is an example

The signal level at 700Hz is about -9dB while at 1500Hz its -19dB. So clearly the power being applied to tweeters is significantly less than that being applied to woofers, even with a low crossover.

A third consideration is that people dont listen to sine waves. Music has a high RMS to peak ratio and is not "continuous" power. This, and the above, is why speaker power ratings are derived using IEC standard test procedures and not continuous sine waves. From the above example the difference between the RMS (which note is the equivilent "heating" power) and the peak power is 14dB Thats a ratio of about 6:1. In basic terms you will never get anywhere near an amplifiers max RMS power output with music. It will hit its peak power output long before.

The only way this tweeter will get damaged within the advertised power range is if the amplifier is clipping and distorting, in which case all bets are off as they would be for *any* speaker.

I hope that helps clear things up for everyone.

 

[ctrl]

What @restorer-john meant to say, at least as I understood him, is that the Sointuva's power rating of 250W long term maximum power

may be formally correct, but lulls the consumer into a false sense of security, as the test conditions have little in common with reality:

The real acid test of the specification is the following:

This is pink noise with a crest factor of 2.

The Purifi woofer used in the Sointuva speaker has a rated noise power handling of just 80W:

Likewise, the Sointuva's Bliesma tweeter only has a rated noise power of 80W with a crossover frequency at 2.5kHz and BU12 filter in this test.


For the rated noise test, as already mentioned, pink noise is used which slopes down at about 3dB per octave (not unlike a music signal in terms of intensity distribution).

But since the tweeter of the Sointuva is not crossed over at 2.5kHz, but almost an octave lower at 1.5kHz (we ignore the acoustic filter order of the Sointuva, which we do not know), the tweeter "sees" in the worst case 3dB higher sound pressure.
Which in the worst case means 1.5 times higher voltage. Therefore, the rated noise power of the Sointuva tweeter and thus the entire speaker should be below 80W.

With this rated noise power, the speaker can be operated safely, for days when it must be.

With the specified "long term power handling" the speaker can actually only be operated safely for the one minute used in the test - everything else is not tested and therefore speculative.

 

[Mikakimi]

What @restorer-john
View attachment 150372

may be formally correct, but lulls the consumer into a false sense of security, as the test conditions have little in common with reality:

View attachment 150373

The real acid test of the specification is the following:
View attachment 150377
This is pink noise with a crest factor of 2.

The Purifi woofer used in the Sointuva speaker has a rated noise power handling of just 80W:
View attachment 150379

Exactly as stated, the long term power handling of the Purifi woofer, as per internationally accepted testing standard IEC 268-5 18.2, is 250 watts. No manufacturer rates their speakers based on the 100 hour test as that would actually be misleading to the general consumer. Music does not have the same RMS (heating) potential as that test (see below).

The tweeter is being driven nowhere near its maximum capability as per that standard as demonstrated above. Not even close - its a factor of 5 lower. It is not at risk of damage as John asserted.

Also as mentioned above the crest factor of music is significantly greater than 2 as used in the IEC test. As such the "heating" (rms) power, which is the important factor, is significantly less with music.

In the example above the peak levels are 14dB above the RMS level, a factor of 6. So even if you were only running at the 80 watts rms cited for the 100 hour test, that means the peak power levels with that music is going to be in excess of 800 watts! Are you doing this at home 100 hours at a stretch? Or even short term?

So it doesnt lull consumers into a false sense of security simply because the real world operation of speakers with actual music means they are not being driven in the way you imply. A typical consumer who doesnt have an understanding of the complexities we are discussing will just compare the rms power of an amp with that of the speaker rating. A 250 watt amp being used normally with *music* and up to a level before clipping will not break the speakers even if run long term. The actual RMS levels with music will be quite low.

 

[ctrl]

The tweeter is being driven nowhere near its maximum capability as per that standard as demonstrated above. Not even close - its a factor of 5 lower. It is not at risk of damage as John asserted.

Yes, you are right about the power handling of the tweeter.

In my consideration I have not taken into account the sound pressure level reduction of the tweeter.
You said with the Sointuva this is -12dB to reach the level of the woofer. If one loses 3dB by the crossover frequency at 1.5kHz (because of intensity distribution of pink noise), then the tweeter with -9dB level reduction still has considerable room to maneuver.

Even with 250W pink noise RMS at the entire speaker, the tweeter would not yet be at its rated noise power limit.
The limiting factor in this case is the woofer.

 

[McFly]

And also that Erin would be have a sample pair in the near future to do this.

@hardisj can you confirm?

 

[McFly]

I dont know what the IEC 268-5 test conditions are, so this is my novice opinion/hypothesis only;

I have those drivers, and if I built a sointuva clone, I'd happily do that IEC 268-5 18.2 test if I had my own desert - to not piss anyone off. There is no way that tweeters gonna break, in that speaker, It's getting way less than the woofer from XO sensitivity matching. The tweeter sees way less signal that the woofer.

My moneys on the woofer bottoming out, and damaging the VC former with 250W clean signal. Depends on the test signal though.

 

[spacevector]

So at the woofers max rating of 250 watts (32 volts input) the voltage applied to the tweeter is reduced by 12dB.
voltage gain in dB = 20 × log (V2 / V1).

So what power is being applied to the tweeter?

8v / 4ohms= 2 amps
8 volts * 2 amps = 16 watts

I am calculating padding resistor to get 12dB reduction for a 4Ohm tweeter, Lpad = 12Ohms

This resistor will also see the 2A current flowing through the tweeter. Is the resistor rated for 50W then?

 

[witwald]

Of course, the rounding does not help with the directivity mismatch between tweeter and woofer, which must be tried to compensate via the crossover tuning.
View attachment 149856
Source

As you've noted, there is a pronounced directivity dip at about 3.5 kHz or so. That type of directivity dip around the crossover frequency would likely be present in many other similar two-way systems, as they essentially use similar driver technology.

What sort of crossover tuning would be needed to correct the directivity mismatch between the woofer and the tweeter? The tweeter has quite low directivity around 3.5 kHz, while the bass-mid driver is starting to beam quite noticeably.

 

[witwald]

As it is basically 6"+1" on flat baffle, crossover could be 2nd order midwoofer and 3rd order tweeter. That is standard solution, resulting quite often in wider directivity above crossover.

Are you referring to the acoustical crossover slopes or the electrical crossover slopes?

Front baffle tilt would allow for symmetrical LR2 and this would provide better directivity pattern, horizontally.

How would the LR2 acoustic filter provide better directivity horizontally? Isn't the purpose of an LR filter to provide a main lobe that points forward as horizontally as possible, so that being off axis (above or below the design listening axis) in the vertical plane doesn't affect the frequency response heard by the listener as they shift their head (ears) up or down?

 

[hardisj]

@hardisj can you confirm?

Well, I was contacted about it a few weeks ago and I said yes. But until it is in my hands, who knows.

 

[puppet]

That 3khz notch is very close to the overall diameter of the tweeter in wavelength. It's not showing up in the manufacturers measurements though. http://www.bliesma.de/Datasheet T34B-4.pdf

Some ripples are more apparent in the AluMag version however. http://www.bliesma.de/Datasheet T34A-4.pdf (not too bad however)
https://audioxpress.com/article/test-bench-the-t34a-4-tweeter-from-bliesma (of note: test baffle very similar in size to Sointuva)
http://www.troelsgravesen.dk/BlieSMa_T34A-4.htm

Seems to be a pretty clean response overall though. Wouldn't the maker just be using the natural response roll off @1.5khz here? So a 1st order response. Maybe the woofer is responsible for this. https://purifi-audio.com/ptt6-5w04-nfa-01/
Their datasheet shows a similar notch around 4khz which could vary within production.

 

[restorer-john]

Do you buy him a new tweeter if it fails catastrophically? For active speakers with limiters, I can understand, but for passive speakers, the manufacturer can always say "you pushed it too hard", as there's rarely a proof it broke without doing so.

Unfortunately, with reviewers (like @hardisj) borrowing speakers for testing, they are in the position of damned if you, damned if you don't. Absolutely they should test for adherence to advertised rated power. But, they don't want to destroy a speaker that doesn't belong to them. Then they are accused of not adequately testing a vital rated parameter. It's akin to a reviewer not testing the 0-60mph time, because you don't want to wear out the transmission.

In this case, a 250W test over a 28 minute period as per the IEC rating. At the end of that test, the speaker must be undamaged and perfectly functional. If not, the rating is BS and should be called out. That is more useful than simply not performing or mentioning the test.

Speaker reviewers would get around the potential destruction issues with review speakers by testing low duty cycle impulses, but you need a phenomenally powerful amplifier, hearing protection and you may upset the neighbours.

Rated power handling is one of the most important factors to buyers. I know, I sold a lot of speakers and inspected a huge amount for warranty claim purposes. Tweeters are always the most commonly damaged component, but most of the blame can be put at the feet of manufacturers who put ficticious or at least highly suspicious maximum power numbers on their speakers. Then to get out of a warranty claim, they will attempt to blame the customer, who, in most cases is fighting an uphill battle to get what they paid for.

It's ironic that old long departed companies like Radio Shack rated their speakers for continuous power, pink noise (24hr), offered a 5 year warranty and replaced literally hundreds of thousands of drivers under those warranties

 

[hardisj]

Unfortunately, with reviewers (like @hardisj) borrowing speakers for testing, they are in the position of damned if you, damned if you don't. Absolutely they should test for adherence to advertised rated power. But, they don't want to destroy a speaker that doesn't belong to them. Then they are accused of not adequately testing a vital rated parameter. It's akin to a reviewer not testing the 0-60mph time, because you don't want to wear out the transmission.

Yup.

I'm all for stress testing it if you guys will cover the cost of replacement parts, though.

 

[hardisj]

Alternatively, there is absolutely nothing stopping anyone from performing the test on their own. All you have to know is what the passive crossover filter on the tweeter is. You don't need an SPL meter. You just need the filter and a signal source and money to burn.

Seriously, after these days' worth of arguing the merit of the rating as spec'd, this test could have been conducted. You don't have to rely on myself or Amir or others for this kind of test. Order the tweeter, build the passive filter (or simulate it actively) and get to rockin'.

 

[restorer-john]

Alternatively, if the test is "so easy" then there is absolutely nothing stopping anyone from performing the test on their own.

You are the reviewer, not me. You are posting otherwise excellent content, apart for the most important part: Do they meet their rated specifications?

And, if a speaker fails when you review it, call it out, have it repaired/replaced by the manufacturer as they are required to do and write about the experience. That's the way proper reviewing has been done for many decades. How often have you read about an amplifier that blew up when being tested and a second sample was sought which blew up too? Quite often.

Be brave. Test your speakers in accordance with their rated power.

 

[restorer-john]

No manufacturer rates their speakers based on the 100 hour test as that would actually be misleading to the general consumer.

Really? No manufacturer eh? Misleading to the general consumer? I guess you'd better tell Andrew Jones (Elac) how to rate loudspeakers correctly, as he must be confused:

...When we establish power rating on a speaker, we test with IEC noise (pink noise filtered to a limited bandwidth to represent a music spectrum, with a crest factor of 6dB) . Let’s say that we want to test for a Max power of 100W. We adjust the long term average power to 25W (6dB crest factor) equivalent to a 100W amplifier just at clipping. We then run the speaker for 96 hours at this level. This is severe. Over such a long period the speaker gets very hot, the drivers being much too hot to touch. After cooling down, we retest the speaker. It must pass the standard test limits we put on a production speaker.

We then increase the level to 33W and run for a further 24 hours. it must survive and shift its specs just moderately. After that we turn up the level to destruction point. The speaker must not catch fire...


Regards

Andrew Jones

ELAC