• 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!

New Purifi amplifiers coming

boXem

Major Contributor
Audio Company
Joined
Jun 19, 2019
Messages
2,018
Likes
4,901
Location
Europe
I don't believe it. 4uV residual is seriously at the level of a half decent D/A converter.

And exactly what are we looking at outside their brick-wall 20kHz? A weighting means -10dB down at 20kHz and -10 to -50dB below 1kHz down to mains harmonics.

UNWTD is where we are at. ASR is all un-weighted.
From the THD+N vs power curve:
THD+N -110 dB @ 0.4W 8R gives 5.6 uV UWT
THD+N -120 dB @ 4W 8R gives 5.8 uV UWT
THD+N -120 dB @ 9W 4R gives 6 uV UWT
 
OP
M

mocenigo

Major Contributor
Joined
Dec 8, 2018
Messages
1,288
Likes
1,052
From the THD+N vs power curve:
THD+N -110 dB @ 0.4W 8R gives 5.6 uV UWT
THD+N -120 dB @ 4W 8R gives 5.8 uV UWT
THD+N -120 dB @ 9W 4R gives 6 uV UWT

And at those power levels THD+N is dominated by noise, not distortion, as distortion starts to pick up only after 10-20W depending on the load. They also told me “10Db better than the 1ET400A” and this fits with the latter’s 12μV (A) residual noise value.

W O W.
 

theREALdotnet

Major Contributor
Joined
Mar 11, 2022
Messages
1,202
Likes
2,075
I wonder whether they have tested it with a 4Ω || 2.2µF load…
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,706
Likes
38,861
Location
Gold Coast, Queensland, Australia
And at those power levels THD+N is dominated by noise, not distortion, as distortion starts to pick up only after 10-20W depending on the load. They also told me “10Db better than the 1ET400A” and this fits with the latter’s 12μV (A) residual noise value.

W O W.

Incredible achievement at 1kHz. What about 5kHz, 10kHz and 20kHz? You know, full audible bandwidth stuff....
 
D

Deleted member 48726

Guest
From what I have seen, it's very far from bad
1685026528080.png
 

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,007
Location
Albuquerque, NM USA
Sorry I do not understand what you are saying. We were talking about the “1” in the module name referring to the number of channels.
All I meant is that one reason to make a monoblock version is heat dissipation. In spite of class D's efficiency, I simply wondered if this might be a factor.
 

Matias

Master Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
5,085
Likes
10,943
Location
São Paulo, Brazil
To be honest, I don't know why Bruno keeps improving what was arguably the best class D tech in the market (Eigentakt gen 1) instead of branching out to power supplies and mains powered all in one modules. These would greatly increase sales, while a high power state of the art Eigentakt gen 2 would sell how many? But I suppose the challenge of pushing the limits and pride of accomplishment is a lot more fulfilling than selling tons of cheaper "good enough" modules.
 
OP
M

mocenigo

Major Contributor
Joined
Dec 8, 2018
Messages
1,288
Likes
1,052
To be honest, I don't know why Bruno keeps improving what was arguably the best class D tech in the market (Eigentakt gen 1) instead of branching out to power supplies and mains powered all in one modules. These would greatly increase sales, while a high power state of the art Eigentakt gen 2 would sell how many? But I suppose the challenge of pushing the limits and pride of accomplishment is a lot more fulfilling than selling tons of cheaper "good enough" modules.

Power supplies: they are doing that. I held a Purifi power supply in my hand a few days ago. However, it is a new design, and it takes AGES for a new design to be approved by trade commissions and safety bodies in all relevant markets. So they are bringing it to the market, but the date is not under their control. In fact, they were powering their amplifier modules with Hypex power supplies at their Open House event.

All-in-one: once they have their power supply design approved for sale, they can build the all in ones.

DSP: they have the in-house expertise to do DSP development, in fact they are using it to put a DSP also in the power supply (do not ask) and it is actually to save money on the BOM. So, once they have the power supply approved everything else can follow quickly.
 
Last edited:

ethanhallbeyer

Active Member
Joined
Feb 24, 2019
Messages
175
Likes
50
Until last week I was wondering if we would do anything with it. Then I saw some performance curves which were just insane. Now I feel that we have no other choice than doing something around this module. What, when and at which cost are questions remaining to be answered.
Are these modules going to be that different from the 7040?
 

boXem

Major Contributor
Audio Company
Joined
Jun 19, 2019
Messages
2,018
Likes
4,901
Location
Europe
Are these modules going to be that different from the 7040?
The one and only advantage of the 7040 over the 1ET400A is the power in very difficult loads. All the other metrics are equal or not as good. The 9040 is better than the 1ET400A for all the metrics (at the exception of price).
 

fredoamigo

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2018
Messages
638
Likes
1,123
Location
South East France
I'm not the type of person to change amplifiers every 3 years, but my venerable amp, designed in the last century and which I've had with me for 15 years, is starting to show some signs of weakness, and I think it's now time to move on to 21st century amplification, and while we're at it, we might as well do it with a SOTA amp capable of powering anything presented to it without any weakness, whatever the impedance.
I'm now ready to pull the trigger on a pair of mono blocks with this module ...

@boXem

I hope you'll be my target. :)
 

SkipperGW

Member
Joined
Jun 2, 2023
Messages
8
Likes
1
i really hope the first purifi 1et9040ba amps will be available soon — can't wait…
 

SkipperGW

Member
Joined
Jun 2, 2023
Messages
8
Likes
1
found this:

Purifi 1ET9040BA

This is a new product with little information. Although it is a rumor level, it is said that the maximum output has been greatly enhanced with BTL, up to 1500W, and the noise is kept lower than 1ET7040SA and 1ET400A. If this is true, the 1ET7040SA seems to have little reason to exist, but looking at the board size and output limit specs, it seems that they can live together. For example, the 1ET7040SA is suitable for stereo power, and the 1ET9040BA is suitable for monaural power amplifiers, judging from the maximum output of the power supply.

However, whether such a lineup will be developed depends on the sound quality of 1ET9040BA and future trends of Hypex . If Hypex unexpectedly comes out with a high output module as well, you'll have to compare the two and choose one over the other.

It seems that the power supply voltage is 90V as a concern, but from my experience so far, the higher the power supply voltage, the more likely the high frequencies will become rough. The reason is that the switching voltage itself rises. Both Hypex and Purifi have very good EMC performance, and both have little leakage noise and residual noise, but I think there is a subtle difference in texture. 1ET400A and 1ET7040SA have a lower power supply voltage, but 1ET9040BA has a higher power supply voltage, so there is a possibility that it is not all good in that part.

However, when it comes to BTL, this problem tends to be alleviated, so it is possible that everything is improving at the same time. In this area, the actual output sound is the most important, so I can't draw a conclusion until I evaluate the actual product.

The other is how the pre-buffer should be. As far as I can see from the board, the buffer is virtually integrated, and I often don't know how far I can go with the sound quality tuning. Like the NC1200, it may be a method that assumes an onboard buffer and switches to an external buffer as needed. The all-in-one specification has only advantages as long as the on-board buffer is used, but if you want to use an external buffer, the bond between the boards will be weak and there will be a problem in terms of sound quality.

For us, the worst case scenario is when the 1ET9040BA is a very nice and underwhelming design that gets worse with tweaks. In that case, it is best to use it as a standard, and the sound is almost the same regardless of which company's 1ET9040BA product you buy, so I don't feel the significance of our company's work. (In that case, cheaper foreign-made ones are fine)

In any case, the development after 1ET7040SA will be decided after checking the sound quality trends of Nilai500, 1ET7040SA, and 1ET9040BA. If the 1ET9040BA does not have a special superiority after actually checking it, we will not make a new product.

IMG_5432-1024x778.jpeg
 

EdW

Senior Member
Joined
Jul 5, 2020
Messages
337
Likes
428
Location
Cambridge, UK
Solid dose of complete nonsense
. . . One stand out misunderstanding is of course the 1ET9040BA is not running with 90V power rails since it is a bridged output stage so more like +/-45V. I think that there is a hypex supply, SMPS1200A180, around this voltage rated at 1500W which some implementors may use.
 
OP
M

mocenigo

Major Contributor
Joined
Dec 8, 2018
Messages
1,288
Likes
1,052
Solid dose of complete nonsense

Yup. They say “there may be new modules and who knows how they sound so we wait and they decide” spiced with a bit of wrong information.
 
OP
M

mocenigo

Major Contributor
Joined
Dec 8, 2018
Messages
1,288
Likes
1,052
Solid dose of complete nonsense

Yup. They say “there may be new modules and who knows how they sound so we wait and they decide” spiced with a bit of wrong information.
 

RandomEar

Senior Member
Joined
Feb 14, 2022
Messages
335
Likes
775
Following a private request, I gathered the (preliminary?) THD+N data from the image provided by @mocenigo earlier in the thread and updated my plot data. The resulting comparison looks like this:
THD+N Comparison_1ET9040BA.png

The usual caveats apply: The data is collected from different sources. The amps are using different gains which I did not track. Some are just not specified, some may be listed in the sources below.

Sources:
 
Top Bottom