What usually happens if you drop a ferrite magnet speaker, it will just chip if designed & constructed properly. YMMV
Don't try this at home especially if you have a vintage JBL Alnico treasure
Today, the cheapest magnet for a treble unit is Neo. (Probably for a bass unit too but only the most evil & cheapskate designers would do this. I haven't done detailed design for a Neo bass but I have for treble units.) You have to consider the cost of whole assembly.
You can measure noticeable compression effects in a complete speaker using Prof Farina's Log Sweep method. More importantly, you can pick this up in Blind Listening Tests to L&V standards bla bla
The "Alnico is more linear" myth is particularly persistent./QUOTE]Chapter & Verse is in the JBL paper. I'm sorry I can't trace it.
Hi Harry,
I must admit that I have not seen the material on the Hypex NCore, but I think Class D still has a long way to go, especially if you believe that high-end audio is not just pure measurements. I will be the first to agree that class D has come a very long way in the last 5+ years. Bruno has made a tremendous contribution to this advancement.
In fact, the class D technology is moving quite fast, and one of the things that will be updated in my second edition is the 5 chapters in my book dealing with class D amplifiers.
Class D will indeed continue to take over a larger and larger part of the market.
One thing that is very important to recognize for class D amplifier measurements is that THD-20 is virtually meaningless for most designs because of the output filters. 1kHz THD does not separate the men from the boys. It is very important that class D amplifier distortion measurements include CCIF 19+20kHz with spectrum analysis. John Atkinson wisely shows this test for both conventional and class D amplifiers. In-band intermodulation distortion (CCIF or multitone, not just SMPTE-IM) is really where it is at.
Some class D amplifiers also suffer from crossover-like distortion at low levels. This must be carefully measured, sometimes underneath the noise.
Cheers,
Bob
Hi Bob,
One factor that plays in the very high performance quoted by Harry is that in the nCore amps the output filter is enclosed in the feedback loop. It's a 5th order loop, actually.
That allows the milliohms Zout over the entire audio band.
I agree there's more than measurements, but it's hard not to be impressed.
jan
One factor that plays in the very high performance quoted by Harry is that in the nCore amps the output filter is enclosed in the feedback loop. It's a 5th order loop, actually.
That allows the milliohms Zout over the entire audio band.
I agree there's more than measurements, but it's hard not to be impressed.
jan
Absolutely! CCIF distortion of the Hypex modules is similarly low.
The NCore uses quite a low filter inductance value (10 uH). This results in high ripple current in the filter inductor and eliminates this crossover-like distortion.
Hi Jan,
Yes, I believe that all of the Hypex class D amplifiers enclose the output filter in the feedback loop. That is a key genious of the design.
Cheers,
Bob
Neo prices
You'll know when this changes cos inexpensive 'mini-monitors' like the Infinity Primus 150 and KEF Coda will stop using Neo treble units. These are VERY price sensitive products.
Also the market will still pay a premium for Neo cos they think its better. That's why the silly fad for expensive Neo bass units.
My apologies to Bob for hijacking this thread.
Yes. I know that. But my friends in the business tell me it is still the cheapest for a treble unit when you consider the whole design.Maybe it was whilst you went walkabout
You'll know when this changes cos inexpensive 'mini-monitors' like the Infinity Primus 150 and KEF Coda will stop using Neo treble units. These are VERY price sensitive products.
Also the market will still pay a premium for Neo cos they think its better. That's why the silly fad for expensive Neo bass units.
My apologies to Bob for hijacking this thread.
Bob
It's a nCore thread here somewhere at the forum, I read only the first tens of pages (like all other interesting threads it became diluted and uninteresting).
You can find the nCore white paper and product sheets at the Hypex site, Bruno's newest patents is also worth reading.
It would be interesting to see a class A or class AB design that can beat the nCore.
Stein
It's a nCore thread here somewhere at the forum, I read only the first tens of pages (like all other interesting threads it became diluted and uninteresting).
You can find the nCore white paper and product sheets at the Hypex site, Bruno's newest patents is also worth reading.
It would be interesting to see a class A or class AB design that can beat the nCore.
Stein
What about Bob's 1983 design?
The NCore has lower distortion. I dare say Bob's amplifier will also clip sooner into low-impedance loads. No disrespect to Bob - I've never made an amplifier with as low measured distortion as his.
Don't know about those.
The only linear amplifier I'm aware of with lower measured distortion than the NCore is Edmond and Ovidiu's PGP amp (but the NCore has higher output power).
Not lower than Bob's design; have you read his paper?
Last I checked, 0.0007% is lower than 0.001%.
I thought about doing that a couple of years ago, as i expected it would be beneficial. They beat me to it
Still there's always other things As far as THD you mean ? There are a number that do.
Plus i've been working on an Add on circuit for Amplifiers etc, that reduces THD quite remarkedly ! For eg,
With Circuit @ 1khz = 0.001% - Without = 0.006%
With Circuit @ 10kHz = 0.007% - Without = 0.039%
With Circuit @ 100Hz = 0.000% - Without = 0.001%
My simulator doesn't go lower than 0.001%
