• 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!

Purifi 1ET400A with linear PSU tested to comply with FTC rule

Status
Not open for further replies.

pma

Master Contributor
Joined
Feb 23, 2019
Messages
5,377
Likes
13,218
Location
Prague
Purifi 1ET400A test with linear PSU

I was curious to test the 1ET400A module with linear power supply and to test it to comply with the FTC (05/2024) rule.
My linear power supply gives 2 x 61Vdc to Purifi module at idle and 2 x 56Vdc at THD1% module power. The test load was 8ohm.

1. THD check in the linear area
1ET400A_linearPSU_1k_8R_103W.png

We can see that the distortion is very low and we have no issues with mains spuriae, despite the fact that conventional design linear power supply is used.


2. Power for THD1%
1ET400A_linearPSU_1k_8R_THD1%.png

THD1% is for 160W/8ohm. Interestingly, the rise of noise floor for THD1% is lower than in case when SMPS was used (another thread).


3. THD vs. frequency and THD1% vs. frequency

1ET400A_linearPSU_thdfreq_8R_103-160W.png

At 160W, the amp keeps THD near 1% up to 4kHz. Then, the distortion goes down to much lower numbers. I have already described this behaviour of the Purifi module, and it is independent of the power supply used.
Plots for 130W and 103W are shown as well. Please note rise of LF distortion in the 130W plot. (maybe familiar to @Sokel )

4. THD vs. level
1ET400A_linearPSU_THDlevel_8R_100Hz-10k.png

Please note that the highest power achieved is at highest frequency (10kHz here).

1ET400A_linearPSU_THDpower_8R_1k.png



5. 5 minute test at 160W/8ohm/1kHz
1ET400A_linearPSU_THD1%_5min_8R_160W.png


The amplifier has passed the 5 minute test easily, staying cold, and with a negligible rise of distortion. The power supply voltage at 160W/8ohm was 2 x 56Vdc.
 
Last edited:
How was the module and power supply mounted during the test? Are they in free air or enclosed?
 
How was the module and power supply mounted during the test? Are they in free air or enclosed?
Purifi module is mounted in this case:

The power supply is inside this case, with another ampllifier:

3 individual wires are used to inter-connect the PSU in one box with the module in the other box. Both boxes are covered and the Purifi stands on the PSU case. Purifi stays cold under any conditions. The case used for Purifi is 2U x 19" (440 x 295 x 95 mm), with 3mm Al bottom panel that serves as a heatsink as well. The SMPS that is inside the box with Purifi module was disconnected for this test.
Again, I can see no problem with the 5 minute THD1% tests.
 
Purifi module is mounted in this case:

The power supply is inside this case, with another ampllifier:

3 individual wires are used to inter-connect the PSU in one box with the module in the other box. Both boxes are covered and the Purifi stands on the PSU case. Purifi stays cold under any conditions. The case used for Purifi is 2U x 19" (440 x 295 x 95 mm), with 3mm Al bottom panel that serves as a heatsink as well. The SMPS that is inside the box with Purifi module was disconnected for this test.
Again, I can see no problem with the 5 minute THD1% tests.
Cool. Literally. :)
 
Purifi 1ET400A test with linear PSU

I was curious to test the 1ET400A module with linear power supply and to test it to comply with the FTC (05/2024) rule.
My linear power supply gives 2 x 61Vdc to Purifi module at idle and 2 x 56Vdc at THD1% module power. The test load was 8ohm.

1. THD check in the linear area
View attachment 417418
We can see that the distortion is very low and we have no issues with mains spuriae, despite the fact that conventional design linear power supply is used.


2. Power for THD1%
View attachment 417419
THD1% is for 160W/8ohm. Interestingly, the rise of noise floor for THD1% is lower than in case when SMPS was used (another thread).


3. THD vs. frequency and THD1% vs. frequency

View attachment 417420
At 160W, the amp keeps THD near 1% up to 4kHz. Then, the distortion goes down to much lower numbers. I have already described this behaviour of the Purifi module, and it is independent of the power supply used.
Plots for 130W and 103W are shown as well. Please note rise of LF distortion in the 130W plot. (maybe familiar to @Sokel )

4. THD vs. level
View attachment 417421
Please note that the highest power achieved is at highest frequency (10kHz here).

View attachment 417430


5. 5 minute test at 160W/8ohm/1kHz
View attachment 417422

The amplifier has passed the 5 minute test easily, staying cold, and with a negligible rise of distortion. The power supply voltage at 160W/8ohm was 2 x 56Vdc.
Very interesting! As an owner of 2 1ET400A I continue to be impressed. Did you do the 5 minute test at all frequencies?
 
Very interesting! As an owner of 2 1ET400A I continue to be impressed. Did you do the 5 minute test at all frequencies?
Not at "all", but at 20Hz, 1kHz and 20kHz, and it passed the tests. I assume it is mainly a question of the power supply, because 1ET400A stays cold, it has very low power loss at high power. And maybe quite large case bottom plate helps as well. BTW, the THD1% again meets the datasheet specs. And, we are not at the maximum allowed supply voltage, which is 2 x 70V. My findings with the linear PSU are valid for 2 x 56Vdc (loaded).

1ET400A_power_vs_supplyvoltage55.png


1ET400A_linPSU_THDlevel_1k.png
 
Maybe one more notable observation. At 160W/8ohm, tested with 24 seconds frequency sweep, the module protection shut it down at 53kHz. With lower power of 130W, it did not shut down. This is an excellent result, IMO.

1ET400A_linearPSU_FR_8R_103-160W.png
 
Maybe one more notable observation. At 160W/8ohm, tested with 24 seconds frequency sweep, the module protection shut it down at 53kHz. With lower power of 130W, it did not shut down. This is an excellent result, IMO.

View attachment 417582
I gather from other threads that as the power increases, the self-clocking frequency reduces (the Eigentakt model). Perhaps such high power at high frequency puts it close to instability.
 
I gather from other threads that as the power increases, the self-clocking frequency reduces (the Eigentakt model).
Yes, as any PWM modulator based on self-oscillating principle. Both reduces and rises depending on signal polarity.

1ET400A_switch_fluctuations.png



Perhaps such high power at high frequency puts it close to instability.

The PWM spectrum (re switching frequency shifts) of the self-oscillating modulator depends on signal frequency as well, mathematically it is pretty complex (@j_j ?). And there is also the LC (2nd order?) LPF filter in the feedback loop, and I suspect that near 50kHz internal circuit amplitude may rise considerably.
 
So you didn't really follow the FTC regulations of testing all frequencies bewteen 20hz-20khz so title thread is misleading.
I believe its rules are open to interpretation, so it's not strictly an either-or situation.
The key point is 5 minutes at different frequencies.
 
No, the key point is 5 minutes at all frequencies I haven't seen a test yet that meets that requirement.
That's the point. It doesn't explicitly state that. However, testing is required across the specified frequency range.
Happy to see that Amir's tests meet the new regulations.
It doesn't. He primarily tests what could be considered burst signals. However, this isn't the place for that discussion.
 
That's the point. It doesn't explicitly state that. However, testing is required across the specified frequency range.

It doesn't. He primarily tests what could be considered burst signals. However, this isn't the place for that discussion.
Correct but I'll reiterate the tests done here at 3 frequencies in no way can be stretched to be an "interpretation " of meeting the requirement of testing all frequencies between 20hz-20khz anymore than Amir's burst signals. So the title thread is misleading in claiming it complies with the FTC regulations. At best it should say here are my tests at certain frequencies that might comply. Gene at audioholics who tested the Yamaha R-N2000A doesn't claim it meets the requirements even though the test he ran is more thorough. He said the test is vague and here's what I came up with. That's all this is.
 
Correct but I'll reiterate the tests done here at 3 frequencies in no way can be stretched to be an "interpretation " of meeting the requirement of testing all frequencies between 20hz-20khz anymore than Amir's burst signals.

I’d say it definitely does due to the duration.
 
Status
Not open for further replies.
Back
Top Bottom