2nd generation EIGENTAKT PURIFI 1ET6525SA (successor to the 1ET400A)

[boXem]

So you have higher noise, higher distortion and lower power output. You said it was 210watts 8ohms. Can you post some data for that? March audio measured 246 watts both channels driven.

You continue comparing apples to oranges. Measurements are done with different generators. So noise and distortion profiles are different.
210 W is what we can commit to customers as of today. That is the lowest value we regularly had when measuring the previous generation of amplifiers.
For the moment the data we have about the 1ET6525SA based amps are not statistically significant. The ones we measured were around 230 W. If the tendency becomes hard data, we will update our specs.
The only ones who have proper statistical data are Purifi and the value they published is below the one measured by March. Not saying that March is wrong, just that the information is not statistically relevant.

That said, most manufacturers do not measure anything and sell module specs. Do you really think that aggressive tone towards the ones who attempt to be honest with their customers is the smartest thing that can be done?

 

[AdamG]

That's fair enough, but he explains the reasons why. What is your noise level with the inputs shorted?
Alan syated the measurements were taken with 21.0dB gain. Your measurements were taken at 20.4dB gain. So as we discussed in the other thread this is too low to reach full power output of the module. At 21dB gain your SINAD figure would fall below 110dB.

View attachment 406099

So you have higher noise, higher distortion and lower power output. You said it was 210watts 8ohms. Can you post some data for that? March audio measured 246 watts both channels driven.

Being a little less aggressive and abrasive will result in greater community engagement and a wider range of responses that might be beneficial to you.

 

[boXem]

@boXem I'm sorry I have come across that way. Apologies. I'm just trying to compare the data that's out there. I do take your point about the different generators. So do I understand your position correctly in that the power outputs from the modules vary significantly? It would still be interesting to see your power sweep results and output noise with inputs shorted.

Apologies accepted, we all are passionate and do not always realize when some borders are crossed.
The SMPS1200 is not regulated and I always assumed that it's rail voltages and ripple performance were not strictly identical sample to sample. But I never took the time to hook up a scope to confirm this hypothesis. After the Arthur 4215/E2, we made some changes to our buffers design to cause less ripple at power supply level (channels working in opposition of phase). That could be one of the reasons why we have seen higher power for the few 1ET6525SA that we have measured.
Power sweep and input shorted FFT are in the todo list. For the moment, all the modules we have are allocated to customers eagerly waiting for their amp.

 

[pogo]

The SMPS1200 is not regulated and I always assumed that it's rail voltages and ripple performance were not strictly identical sample to sample. But I never took the time to hook up a scope to confirm this hypothesis.

Interesting. There should be a separate thread for this.

 

[miero]

From the Hypex SMPS1200 datasheet:

Note 1: Output voltage is proportional to the mains line voltage
(Min@180Vac, Typical@230Vac, Max@264Vac)

 

[Streamc]

I'm thinking of replacing my current amp with this amplifier, would SMSL SU-X be a good match as a DAC for it? It has 5.2 Vrms - does it mean I should match it with amplifier with low gain?

I answered once but reread qustion. Low gain is also suitable. High Gain is just my preference. I think impedance (reactive part of it) proportion is fine.
30 may of 2022 I asked directly Bruno Putzeys via email with this question

Should I use input buffer if without it sound is loud enough? I read that there should be high input impedance of amp.
But if I have dac should I consider DAC first cascade of amp system?
I do not like that I have opamp on DAC and on input buffer.
Like example DAC is SMSL M300.
What the pros and cons on input buffer with gain 13dB?

I got answer:

Actually the simplest solution is to try and compare by ear. On my system, the amp sounds best without buffer because my DAC can handle low impedance loads. But I have heard DACs that did not like the low impedance and in that case it sounded better with the buffer. It’s difficult to predict without knowing the exact output circuit of the DAC. So just try and hear what sounds best to you.

Before this also were question from me.

I have SMSL M300 MK2 with 4.2V RMS output XLR. Can it correspond as input signal to Eval 1 board?
Do I need some preamp. I have Topping A90 with 49vpp. But I think that direct connect give more clear sound as I listen.

And answer

Yes you should be able to connect your DAC directly to the EVAL1. 4.2V is quite sufficient (unless the input buffer is bypassed).

I thanked Bruno.

I used Eval1 with hg, lg both. I stay with hg now (buffer is on) because of color or something. But electrically there should be no problem.

 

[JohnnyNG]

I answered once but reread qustion. Low gain is also suitable. High Gain is just my preference. I think impedance (reactive part of it) proportion is fine.
30 may of 2022 I asked directly Bruno Putzeys via email with this question

I got answer:

Before this also were question from me.

And answer

I thanked Bruno.

I used Eval1 with hg, lg both. I stay with hg now (buffer is on) because of color or something. But electrically there should be no problem.

I have a Topping Pre90 and buffered (Audiophonics 1ET400A mono amps) sounds much better than bypassed even though the Pre 90 technically should have no problem with the impedance based on the numbers found in the ASR review. So, yeah, looks like I followed Bruno's advice as well to try and compare by ear.

 

[pogo]

A good matching of the components gains is important for a good end result, see also here as an example for the M33:
NAD has added a gain stage to increase its headroom so that it works better with the DSP circuits of its room-correction software and tone controls. NAD concedes that this slightly decreases the M33’s signal/noise ratio, but with careful design and premium parts, and because the Eigentakt’s S/N is so low to begin with, the noise is still inaudible.
Source: soundstagehifi.com

Btw, Bruno was probably also involved

 

[Oniiz86]

It appears that Purifi is finally getting ready to list the 1ET6525SA officially on their website with revised power rating specs for 4 Ohms which is now 450W reduced slightly from 480W.

1ET6525SA Datasheet is now available

 

[pogo]

This makes an exchange easier:

 

[Streamc]

I have a Topping Pre90 and buffered (Audiophonics 1ET400A mono amps) sounds much better than bypassed even though the Pre 90 technically should have no problem with the impedance based on the numbers found in the ASR review. So, yeah, looks like I followed Bruno's advice as well to try and compare by ear.

I also compared connecting earlier Topping A90 as preamp. Sound changed minimally but bypassing it resulted in "cleaner" for me. Maybe configuration with Pre90 is another sound for ear.

 

[pma]

I have been waiting to see this data. 115dB SINAD!
View attachment 406049
I think k Boxem posted a quick measurement of their version in another thread, but it wasn't close to this.

SINAD = -(THD+N) so it is 111.9 dB in the plot you have posted.

 

[boXem]

Volume supply has started this month.

 

[PEQuestrian]

When I buy a Purifi amp the dependability, timely order fulfillment and willingness of the builder to go the extra mile is what decides who my next order goes to - NOT a difference of 2-3 dB in SINAD even if real. Frankly, small SINAD differences are TOTALLY unimportant in the listening room. Having purchased amps from both Boxem and March Audio, I would definitely go back to Boxem. Enough said...

Couldn't agree more. My 4216/E2 stereo amp just arrived last week and I find both build- and sound quality, as well as the feature set excellent. Fred was super helpful and worked with me on customizing the low gain input setting to better match the low pre-out voltage (1.4V max) on my Denon X3800H. Good luck finding this level of engineering-first customer centrism from any larger manufacturer. So even though I'm US based, I had zero concerns ordering with Fred.

In general, the continuous power ratings and 2 Ohm stability with these new Purifi modules paired with minimal heat emission (vs older Hypex, or class A or A/B amps), low power consumption and compact form factor in a very nice implementation objectively represent an excellent HiFi value. I appreciate Purifi, Boxem, Buckeye and other smaller vendors here who are pushing Class D forward at reasonable prices for helping democratize sound quality.

 

[pma]

Signal generator noise limits the accuracy.

Yes, that is a common issue, especially it affects the noise. It adds: Vnoise(generator) * Gain(amplifier) = method error noise. I have the same problem.

 

[miero]

A datasheet for 1ET6525SA is now available on the Purifi website including the EVAL6 Stereo 1ET6525SA Evaluation Kit (EUR 706.00 VAT Excluded):
- https://purifi-audio.com/shop/eval6-eval6-stereo-1et6525sa-evaluation-kit-2576

 

[pma]

March basically measured the amp noise with inputs shorted then added the THD component from the measurement.

I know, but it is incorrect. At high power the noise floor rises, thus N in the THD+N rises and the THD+N is worse than THD + (Noise with the shorted input). His method would measure SNR, not THD+N. It would make the numbers to look nicer. Please see below. The green plot, noise only. The plot is in dBW, which is not usual here, but @John_Siau knows well why. If the noise was constant, the green plot would be straight.

 

[pma]

A datasheet for 1ET6525SA is now available on the Purifi website including the EVAL6 Stereo 1ET6525SA Evaluation Kit (EUR 706.00 VAT Excluded):
- https://purifi-audio.com/shop/eval6-eval6-stereo-1et6525sa-evaluation-kit-2576


View attachment 409660

Measuring bandwidth not shown, as usually. So, the THD+N plot without BW mentioned is pointless. I know, somewhere in the datasheet they probably say 20Hz-20kHz. However, they should have mentioned it in the plot notes.

 

[DACs_Lover]

What is its SINAD Value?
It's unbelievable. Better than Benchmark AHB2's.
Do we still hear the difference between these extreme fidelity amps?

 

[boXem]

What is its SINAD Value?

Depends mainly from the input/gain stage it is hooked to. The 1ET6525SA is a part (the main one tbh) used to build an amplifier, not a finished amplifier.
The same way that the DAC chip performance doesn't tell the DAC performance since it is highly dependent from the power supply and output stage used.