likely different measurement bandwidth.
Thank you @hipoagu, Yes, good observation :=)
As noted, the bandwidth for the dashboard is half as much so integrates less noise and distortion into THD+N computation (22.4 kHz vs 45 kHz).
Seeing this makes me regret my purchase even more. What with the somewhat
LOL. I am old enough to actually have started my engineering degree using slide rules. Sometimes, old ways die hard.
Keep in mind that for use as a desktop amplifier in a simple system, the A20 is OK. I can hear a reduction in distortion with small speakers when I switch the high pass filter in. If your goal is to put together a high fidelity system, then no, the A20 misses the mark but at current Black Friday pricing it is competitive. When priced closer to $200, I agree it is not a great choice. I regret buying mine from that standpoint but I also have fun measuring and testing this equipment so there is some value in that for me.
I mean yeah, if all you needed was a single pole highpass, you could just as well use a single inline capacitor between devices, and calculate its value by desired corner frequency and input and output impedances of your gear.
import numpy as np
from scipy import signal as signal
import matplotlib.pyplot as plt
norder = 1
hpf = signal.butter(norder, 1.0, btype='high', analog=True) # For the filter, cutoff frequency is normalized to 1
wc = 80.0 # Actual required filter cutoff frequency
w = np.logspace(np.log10(10.0), np.log10(100000), 200) / wc
w, h = signal.freqs(*hpf, worN=w)
ymin, ymax, ystep = (-5.0, 5.0, 1.0)
fig, ax = plt.subplots(figsize=(8, 6))
ax.semilogx(wc*w, 20 * np.log10(abs(h)), '-', markersize=2, markeredgecolor='k')
ax.set_title('Butterworth order {:d} HPF frequency response'.format(norder))
ax.set_xlabel('Frequency, Hz')
ax.set_ylabel('Amplitude, dB')
ax.set_yticks(np.linspace(ymin, ymax, int((ymax-ymin)/ystep)+1))
ax.set_ylim(ymin, ymax)
ax.margins(0, 0.1)
ax.grid(which='both', axis='both')
ax.axvline(wc, color='green', lw=1, ls='--')
ax.axhline(0, color='k', lw=1)
fig.savefig('hpf_fr.png')
plt.show()
I'm feeling like Amir should start putting a disclaimer on his reviews as this needs to be noted in almost every amplifier review it seems: the way Amir rates power is not how any manufacturer (or any other reviewer to my knowledge) rates power and you can't compare Amir's power rating to anyone else's. It's also somewhat iffy to compare between Amir's reviews as his method of taking power at the "knee", where the amp's distortion starts to increase rapidly, is inherently subjective and done by eyeball. To be clear, I have no issue with Amir's method as I understand the reasoning and it's consistent enough between reviews, but it does cause this issue where Amir's ratings are often very different from the manufacturer's specifications and we get these comments.
Aiyima claimed 150W per channel into 8ohms with 1% distortion using the 48V/10A power supply. I cannot see that this is the case. Perhaps my reading of Amir’s graphs is incorrect.I'm feeling like Amir should start putting a disclaimer on his reviews as this needs to be noted in almost every amplifier review it seems: the way Amir rates power is not how any manufacturer (or any other reviewer to my knowledge) rates power and you can't compare Amir's power rating to anyone else's. It's also somewhat iffy to compare between Amir's reviews as his method of taking power at the "knee", where the amp's distortion starts to increase rapidly, is inherently subjective and done by eyeball. To be clear, I have no issue with Amir's method as I understand the reasoning and it's consistent enough between reviews, but it does cause this issue where Amir's ratings are often very different from the manufacturer's specifications and we get these comments.
Let's look at the actual graphs shall we:
I'm unclear why there isn't agreement between the max power bar chart and the power versus distortion chart, as we can plainly see that the amp reaches about 200W before hitting 1% distortion (-40dB THD+N). Rating power at 1% distortion is common. We are still short of the 250W rating, but I don't see in Aiyama's specs how they did the rating so it may have been taking at an even higher distortion figure.
While not specifying the conditions for the power rating is bad practice, based on the graph this amp almost certainly can put out 250W into 4ohms as long as you allow for a fair amount of distortion.
Yes, I tried to make clear that I have no issue with how Amir does his testing. Its more about people seeing Amir's power ratings and noting (correctly) that they are often very different than the manufacturer's specs without understanding that they aren't comparable.
Ah yes, double measurement bandwidth will add up noise and distorsion and no signal, i understand.
The high-pass filter (HPF) is 12 dB/octave, and the low-pass filter (LPF) is 24 dB/octave. Initially, we planned to make the HPF 24 dB/octave, but due to the difficulty of ensuring potentiometer consistency in a fully balanced circuit, after multiple tests and adjustments, we ultimately changed the HPF to 12 dB/octave. We sincerely apologize for any confusion this may have caused.
Hi, sorry ... what Low Pass Filter? I missed that, how is this implemented? Thanks
