NC252MP (class D) vs. A250W4R (classAB) burst measurements into 4ohm//2.2uF load

[LTig]

I am starting to wonder if you do not understand the advises on purpose. "the transformer has sufficient leakage inductance and series resistance"

0.3 Ohm is probably the resistance of the air core inductor. We don't know whether this is a realistic value, but it certainly is far below the specified minimum load (2 Ohm, AFAIR) of the amp.

 

[IAtaman]

I am starting to wonder if you do not understand the advises on purpose. "the transformer has sufficient leakage inductance and series resistance"

Same here. Do you just want to make a point of these results or are you after an actual case, I don't know anymore.

Again Ncore is excellent with pure resistors, good with "usual" speaker load but hardly usable with speakers that have high capacitive impedance, even if in series with small inductance and resistance.

Which speakers are those?

 

[Deleted member 48726]

Same here. Do you just want to make a point of these results or are you after an actual case, I don't know anymore.


Which speakers are those?

Same. Looked promising to begin with but I can't really see the point in these tests if not based on real scenarios.. Shame really. Feel like missed opportunity to, at least for me, to expand knowledge on amplifier behavior with complex loads.

 

[pma]

On a positive note

Previous tests with 4R//2.2uF load and 4R7//(2.2uF+1.6uH+0.3ohm) loads have shown that Ncore does not like and does not work well with high capacitive load, even if there is a small inductance and small resistance added to the capacitor. It was the so called "worst case" test and I understand than not many users here will connect the Ncore to Martin Logan Montis or to Sound Lab A-1. Still, it does not mean we should stick with tests with load resistors only.

For the purpose to simulate "usual" 2-way passive speaker, I have the dummy load with this impedance:

The test setup looks like this

The test frequencies chosen for THD+N vs. level
63Hz: 13.74ohm/-51°
500Hz: 13.52ohm/+44°
2310Hz: 7.105ohm/-9.2°
10kHz: 10.35ohm/-43°
This is quite "light" load

1. Noise
Noise (with NC252MP connected to Topping D10s) is very good 72.91 uV(A)

2. THD+N vs. output voltage

At 63Hz, we can see the nonlinearity of the big 18mH inductor that simulates woofer resonance, above 10V. Real speaker behaviour is similar.

3. THD+N vs. frequency at 9.5V output voltage and 2 measuring BW

Below 60Hz, the influence of 18mH inductor nonlinearity. Real speaker behaviour is similar.

4. Conclusion

With resistive load or with complex load of most usual speakers with dynamic drivers, the Ncore behaves very well.

 

[Davide]

I wonder, as a technically layman, at the cost of a finished NC252MP (about 400 euros), is there on the market a balanced class AB with similar power/performance ratio, which doesn't exhibit the anomalous behavior with this load, nor the DC protection issue highlighted by pma in the other thread?
Just to see if it makes sense to put money into nCore...

 

[ahofer]

Not that the low frequency distortion is audible, but it is interesting to consider the rising distortion at low frequencies and the fact that class D has been adopted for lower frequency drivers much more readily than for high frequencies (Obv power, heat, and space are big considerations)

 

[Sokel]

Not that the low frequency distortion is audible, but it is interesting to consider the rising distortion at low frequencies and the fact that class D has been adopted for lower frequency drivers much more readily than for high frequencies (Obv power, heat, and space are big considerations)

There are some strange things happening to mine at least,have a look here (acoustical measurement at 85db),i have replicated this 5874168456 times:

Topping LA90 Review (Integrated Amplifier)

I just did a sloppy test just for fun.The measurements are taken in the listening position 2.8 meters far all other things been equal,the only thing I switched was the amps.Have a look at the distortion (all the chain plus room,thats why I said sloppy).I think power has a lot to do with the...

www.audiosciencereview.com

It's driving a 10" scanspeak in a 110 litters ported box.
Power?I don't know.

 

[pma]

There are some strange things happening to mine at least,have a look here (acoustical measurement at 85db),i have replicated this 5874168456 times:

Topping LA90 Review (Integrated Amplifier)

I just did a sloppy test just for fun.The measurements are taken in the listening position 2.8 meters far all other things been equal,the only thing I switched was the amps.Have a look at the distortion (all the chain plus room,thats why I said sloppy).I think power has a lot to do with the...

www.audiosciencereview.com

It's driving a 10" scanspeak in a 110 litters ported box.
Power?I don't know.

As I replied in LA90 thread, it looks like an interaction between the woofer impedance near resonance with output circuits of the amplifiers used. I have seen it several times even at acoustical side, as you did.

Just for fun, here is the comparison of THD+N vs. frequency of NC252MP and my small PM-AB2, the amplifier that measures worse, into resistor load, than the NC252MP. The differences are not big, but they exist. Again, the interaction of output stages with the complex load.

 

[Matias]

Interesting rise in distortion below 60 Hz. I wonder how the higher power modules behave, a pity you only have NC252MP with you and not say NC502MP or NC500/NCx500 or 1ET400A.

 

[SIY]

Interesting rise in distortion below 60 Hz. I wonder how the higher power modules behave, a pity you only have NC252MP with you and not say NC502MP or NC500/NCx500 or 1ET400A.

I see that with THD curves obtained with Farina sweeps. It's likely a measurement artifact, but I haven't thought about it deeply enough to explain or cure it.

 

[pma]

Interesting rise in distortion below 60 Hz. I wonder how the higher power modules behave, a pity you only have NC252MP with you and not say NC502MP or NC500/NCx500 or 1ET400A.

I see that with THD curves obtained with Farina sweeps. It's likely a measurement artifact, but I haven't thought about it deeply enough to explain or cure it.

Already explained several times, like here.

Below 60Hz, the influence of 18mH inductor nonlinearity. Real speaker behaviour is similar.

Stuart likes to speak about measurement artifacts. There is no such rise with resistor load or R//C load.

The distortion rise is with level, see the 63Hz plot.

As an example, measurement of PM-AB2 into 4ohm//2.2uF load. Yes, the load that NC252MP cannot handle. Comparison with speaker dummy load with nonlinear inductor.

 

[pma]

0.3 Ohm is probably the resistance of the air core inductor. We don't know whether this is a realistic value, but it certainly is far below the specified minimum load (2 Ohm, AFAIR) of the amp.

Sir, you need to have a look at the impedance plot.

At 20kHz, the impedance magnitude |Z| is 3 ohm. The NC252MP specs say

All the specs are within audio band. 0.3 ohm of this 4R7 // (2.2uF+1.6uH+0.3ohm) load is at 75kHz. Who puts 75kHz specs into the datasheets? Speaker or amplifier manufacturers? Haven't seen any. You guys are just reluctant to admit the fact that the Ncore cannot handle the higher capacitive load, and it is because of the circuit design used. It is easy to show it in loopgain simulation. But this task should have been done by somebody else.

 

[LTig]

Sir, you need to have a look at the impedance plot.

At 20kHz, the impedance magnitude |Z| is 3 ohm. The NC252MP specs say

View attachment 276145

All the specs are within audio band. 0.3 ohm of this 4R7 // (2.2uF+1.6uH+0.3ohm) load is at 75kHz. Who puts 75kHz specs into the datasheets? Speaker or amplifier manufacturers? Haven't seen any.

As an EE i learned to read specs and especially to notice what's missing in the specs. When the designer does not specify a frequency range for which the minimum load is specified (empty Note) I must assume it to be at any frequency. Life is easier this way.

 

[NTK]

As an EE i learned to read specs and especially to notice what's missing in the specs. When the designer does not specify a frequency range for which the minimum load is specified (empty Note) I must assume it to be at any frequency. Life is easier this way.

The Hypex spec didn't specify frequency range. It is an assumption that PMA made. A conservative interpretation would be applied to all frequencies, instead of assuming the spec limits are applicable only from 20 -20k Hz, and can be a short elsewhere.

 

[mdsimon2]

I wonder, as a technically layman, at the cost of a finished NC252MP (about 400 euros), is there on the market a balanced class AB with similar power/performance ratio, which doesn't exhibit the anomalous behavior with this load, nor the DC protection issue highlighted by pma in the other thread?
Just to see if it makes sense to put money into nCore...

This is an interesting question. I haven't looked at class AB amps in some time but I really can't seem to find something that does:

100 W x 2 @ 8 ohm
200 W x 2 @ 4 ohm
400 W x 1 @ 8 ohm BTL

Preferably with gain <= 26 dB and 5W in to 4 ohm SNR > 90 dB. Anybody know of such a product (regardless of price)?

Michael

 

[rwortman]

This discussion reminds me of “diesel-gate”. I wonder how many amps are designed to generate great test bench performance with less regard paid to actual use cases.

 

[NTK]

This discussion reminds me of “diesel-gate”. I wonder how many amps are designed to generate great test bench performance with less regard paid to actual use cases.

Ummm ... This Hypex amp is designed to actual use cases. It is the test that wasn't representative of actual use cases.

 

[MaxwellsEq]

Ummm ... This Hypex amp is designed to actual use cases. It is the test that wasn't representative of actual use cases.

True, but the normal test case - 8 or 4 Ohm resistive loads - is equally unrepresentative of the real world.

 

[PeterOo]

True, but the normal test case - 8 or 4 Ohm resistive loads - is equally unrepresentative of the real world.

We would need measurements of actual speakers to support that statement. I have seen only measurements of “equivalent circuits” with little validation.
Did I miss those measurements?

 

[rwortman]

We would need measurements of actual speakers to support that statement. I have seen only measurements of “equivalent circuits” with little validation.
Did I miss those measurements?

Traditional cone speakers have wire wound coils with significant inductance and crossover networks are chock full of capacitors and more inductors. That speakers are not a resistive load is readily apparent.