• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Buckeye Purifi Eigentakt 1ET9040BA monoblock power amplifier Stereophile Measurements

Status
Not open for further replies.
Agree... something seems off.


JSmith
Tried to find a patern with other class D amps that he reviewed and it seems that similar amps show the same.
NAD C 298 for example (way lower of course) :



1734701372655.jpeg



It must be the measuring protocol,or something.
 
Last edited:
Tried to find a patern with other class D amps that he reviewed and it seems that similar amps show the same.
NAD C 298 for example (way lower of course) :






It must be the measuring protocol,or something.
I do not think so. It is 100W(peak) 8 ohm and it looks realistic at more than -100dBr. It is definitely not that “similar amps show the same”. I would not disqualify JA, he is doing his job very well and to me is credible. Remember that his Buckeye measurement is only -70dBr. That is a huge difference. I do not suspect the measurement “protocol”. Something is just wrong. We need to wait for the answer regarding the amp under test case material. And we would need one more CCIF independent measurement. Not at 5W, as Amir is doing, but at about 50W.
 
A guess would be because a heat sink at room temperature won't generate airflow, but once preheated, it allows convection to continuously remove heat from the components.
That has since occurred to me and it would explain why John thinks there's no need to do preconditioning - although there are slots in the top, there's not much chimneying going on, unlike a vertically mounted heatsink with a large surface area.
 
I do not think so. It is 100W(peak) 8 ohm and it looks realistic at more than -100dBr. It is definitely not that “similar amps show the same”. I would not disqualify JA, he is doing his job very well and to me is credible. Remember that his Buckeye measurement is only -70dBr. That is a huge difference. I do not suspect the measurement “protocol”. Something is just wrong. We need to wait for the answer regarding the amp under test case material. And we would need one more CCIF independent measurement. Not at 5W, as Amir is doing, but at about 50W.
I am puzzled because even the Rogue Audio DragoN with its tube front stage (followed by Hypex) and the humongous distortion only shows the weakness of tubes at 1kHz or so and not the huge sidebands of the Purifi.

That's at 100Wp at 4 Ohm:

1734705309321.jpeg

It's strange,something is way off.
 
That is a huge difference. I do not suspect the measurement “protocol”.
Maybe that's the difference between engineers and scientists. :D
 
Our cases are definitely aluminum.
OK, thanks for the explanation. Then - what is your explanation of such huge difference in CCIF IMD compared to the datasheet parameters? What is wrong? Will you blame it on the reviewer?
 
Then - what is your explanation of such huge difference in CCIF IMD compared to the datasheet parameters?
You're making a massive assumption here.
 
OK, thanks for the explanation. Then - what is your explanation of such huge difference in CCIF IMD compared to the datasheet parameters?
Will be getting back to this question once we do some new measurements (after the holidays) since it is not what we observed during our own measurements during initial prototype/final production builds earlier in the year.
 
Maybe that's the difference between engineers and scientists. :D
Your typical comment. Saying absolutely nothing, just an ironic meaningless kick. So, your qualified scientific opinion, if you have any, why so high CCIF 3rd order distortion in the Stereophile test? Why worse at least 30x than the datasheet specs?
 
You're making a massive assumption here.
What assumption? Have you ever seen the datasheet data? Have you ever measured any Purifi module for CCIF at higher power? What assumption, man? The Stereophile result is 30-40dB worse than the datasheet plots, it is not huge difference? It is you who are making meaningless pointless comments.
 
Will be getting back to this question once we do some new measurements (after the holidays) since it is not what we observed during our own measurements during initial prototype/final production builds earlier in the year.
It may be an error in the individual measurement as well, so it should be clarified at both parties. I would expect that before JA published the data he should have consulted that discrepancy with you. If I was not living that far from you, I would suggest to re-measure the amp here.
 
It may be an error in the individual measurement as well, so it should be clarified at both parties. I would expect that before JA published the data he should have consulted that discrepancy with you. If I was not living that far from you, I would suggest to re-measure the amp here.
I was not contacted by JA regarding any measurement results, only found out about them during final review/edit phase of the article.
 
What assumption?
As you correctly followed up with after my comment, that the measurement reflected reality. It looks screwy to me (-120dB at 1kHz but -70dB sidebands?), so before I would publish it, I'd dig in a bit to see if I or my equipment were doing something wrong.
 
So, your qualified scientific opinion, if you have any, why so high CCIF 3rd order distortion in the Stereophile test? Why worse at least 30x than the datasheet specs?
I can't remotely and without data troubleshoot the measurement. But I also don't assume that it must be correct.
 
With a 1kHz IMD product down -120dB, this is more of what an expected spectrum should look like, just for reference. And because the newer APx series have fewer issues with slewing distortion at their inputs, I did not use the external filter to make this measurement.

1734714203293.png
 
As you correctly followed up with after my comment, that the measurement reflected reality. It looks screwy to me (-120dB at 1kHz but -70dB sidebands?), so before I would publish it, I'd dig in a bit to see if I or my equipment were doing something wrong.
But it was the Stereophile who published it (and I quoted and linked the source in the post #1), with that huge IMD, not me. I have hoped you did not assign that measurement to me. My measurement of (yes, 1ET400A, not 1ET9040BA) is in the first post of the thread and is close to datasheet plots.

Regarding your repetitive comments on engineers vs. scientist, let me please post a link


to clarify some misled views on my/our career. Yes, not audio.

Last, why my assumption on iron steel case - because of the plots below. Taken with another DUT, but clearly showing the impact of case material to filter inductors non-linearity within such case.

LC_Al_case_ccif.png LC_iron_case_ccif.png
 
I don't know what "50W peak" means to describe test tones.

I use that phraseology because each of the 19kHz and 20kHz lies 6dB below the peak waveform level. (See attached 'scope capture.) The levels of the intermodulation products are referenced to that peak level, not the RMS level.

John Atkinson
Technical Editor, Stereophile
 

Attachments

  • 19-20k Waveform.pdf
    29.3 KB · Views: 166
I use that phraseology because each of the 19kHz and 20kHz lies 6dB below the peak waveform level. (See attached 'scope capture.) The levels of the intermodulation products are referenced to that peak level, not the RMS level.
That graph shows Volts, not Watts. With Volts (and Amperes) peak-to-peak is often a useful way to measure. But with Watts, given that they don't go negative, ... I'm confused.
 
Status
Not open for further replies.
Back
Top Bottom