Bruno Strikes Again. Purifi 1ET7040SA VTV Amplifier.

[mrmojo2022]

The Australian HiFi test is on 230V nominal, likely 240V at the time of test. Our standard was 240V (and still sits at 240V). I'm currently sitting on 239V here. Steve Holding and "newport test labs" have no credibility here and never have had. Read the review- he thinks the 'noise floor' is 140dB down...
View attachment 182227

Clearly he hasn't heard of FFT gain when looking at his spectrum analyzer.

140dB is 10 million times. With respect to full power, that means a residual of 4uV. The Purifi is quiet, but it aint that quiet.

True about the fft gain. is Steve part of Newport labs or the magazine?

I thought they were separate (not sure) and Newport an independent test house that just produce the data. what's your reason for saying Newport's data has no credibility?

 

[restorer-john]

I thought they were separate (not sure) and Newport an independent test house that just produce the data. what's your reason for saying Newport's data has no credibility?

Of course you thought that. That's only natural to believe what you read. But it aint remotely true.

Newport Test Labs is a fabrication, in my opinion. It's registered (gotta love publicly searchable ASIC company information, huh?) to, wait for it, the long term editor of Australian HiFi magazine, Greg Borrowman. Why Newport you ask? Because he clearly likes the suburb in Sydney as they own the house/s at this address:

And then we have this, from public property ownership records:

So we know the official registered address for the Australian company "Newport Test Labs" is this lovely looking place, part-owned by the editor of the Magazine...

So, will the real "Steve Holding" (if he even exists) please stand up?

 

[mrmojo2022]

Of course you thought that. That's only natural to believe what you read. But it aint remotely true.

Newport Test Labs is a fabrication, IMO. It's registered (gotta love publicly searchable ASIC company information, huh?) to, wait for it, the long term editor of Australian HiFi, Greg Borrowman. Why Newport you ask? Because he clearly likes the suburb in Sydney as they own the house/s at this address:

View attachment 182272

And then we have this, from public property ownership records:
View attachment 182274

So we know the registered address for the Australian company "Newport Test Labs" is this lovely looking place part owned by the editor of the Magazine...

View attachment 182277

Will the real "Steve Holding" please stand up?

Interesting findings. So not independent, although that was my inference not their claim

However can you explain why you think the data from the testing has no credibility?

 

[restorer-john]

However can you explain why you think the data from the testing has no credibility?

Sure, when you can tell me how to put old heads on young shoulders...

Do your own research. Read 40 years of those magazines. Look at the frequent errors in their reviews. Misprints, f#ck ups etc. Go to the trade shows. Meet the players. Work in the retail HiFi business. Take up engineering at UNI. Repair, study and restore HiFi gear. I did all that. It was hard work, but I love it.

Why do you think I am on ASR? To make money? Nope, I'm here to help people cut through baked-on the bullsh#t of this industry and give some perspective and balance. Just like Amir, it may upset a few snowflakes now and then, but that's good. (not saying you're a snowflake).

 

[mrmojo2022]

Sure, when you can tell me how to put old heads on young shoulders...

Do your own research. Read 40 years of those magazines. Look at the frequent errors in their reviews. Misprints, f#ck ups etc. Go to the trade shows. Meet the players. Work in the retail HiFi business. Take up engineering at UNI. Repair, study and restore HiFi gear. I did all that. It was hard work, but I love it.

Why do you think I am on ASR? To make money? Nope, I'm here to help people cut through baked-on the bullsh#t of this industry and give some perspective and balance. Just like Amir, it may upset a few snowflakes now and then, but that's good. (not saying you're a snowflake).

So no actual reason then.

I think I will leave it there. not interested in your speculation or self aggrandisation.

 

[boXem]

Erm you might want to revise your website claims in that case.

I was initially wondering what you were talking about. Found it. Thanks!

 

[boXem]

However can you explain why you think the data from the testing has no credibility?

227 W 1% distortion in 8 Ohm means 42 V RMS.
42 V RMS is 60.2 V peak, i.e 4.8 V away from the 65 V rails in the Purifi datasheet.

The test found 236 W in 8 Ohm at an unknown distortion. This means 61.5 V peak with a power supply delivering 63 V unloaded.

Kind of defy physics no?

 

[Yasuo]

I like that shirt

 

[mrmojo2022]

227 W 1% distortion in 8 Ohm means 42 V RMS.
42 V RMS is 60.2 V peak, i.e 4.8 V away from the 65 V rails in the Purifi datasheet.

The test found 236 W in 8 Ohm at an unknown distortion. This means 61.5 V peak with a power supply delivering 63 V unloaded.

Kind of defy physics no?

I'm not clear on the point you are making. these supplies aren't regulated. they will also have tolerances. they aren't all smack on 63 volts. your own data earlier showed this to be the case with what you described as weak units.. their output will also vary with mains input voltage to a degree. I don't see where the laws of physics have been broken. In both cases the amp output voltage is lower than even the nominal psu voltage.

 

[boXem]

I'm not clear on the point you are making. these supplies aren't regulated. they will also have tolerances. they aren't all smack on 63 volts. your own data earlier showed this to be the case with what you described as weak units.. their output will also vary with mains input voltage to a degree. I don't see where the laws of physics have been broken. In both cases the amp output voltage is lower than even the nominal psu voltage.

Purifi states that to get less than 1% distortion the peaks need to be at least 4.8V away from the rails. To get 246 W 1% you would need rails at 67.5 V. The ratio Mains/rails of the SMPS1200A400 is 3.7. To get rails at 67.5 V unloaded you need mains at 250 V. That is indeed what the report states since mains voltage varied between 234 and 253 V. So you are right, physics are ok. Luckily for March his amp's power was measured with mains at 253 V...
In EU mains are at 230V, not 253.

Anyhow that doesn't change anything to my initial statement: the 1ET7040SA will deliver the exact same power as the 1ET400A above 3.3V when powered by the same PSU.

 

[mrmojo2022]

Purifi states that to get less than 1% distortion the peaks need to be at least 4.8V away from the rails. To get 246 W 1% you would need rails at 67.5 V. The ratio Mains/rails of the SMPS1200A400 is 3.7. To get rails at 67.5 V unloaded you need mains at 250 V. That is indeed what the report states since mains voltage varied between 234 and 253 V. So you are right, physics are ok. Luckily for March his amp's power was measured with mains at 253 V...
In EU mains are at 230V, not 253.

Anyhow that doesn't change anything to my initial statement: the 1ET7040SA will deliver the exact same power as the 1ET400A above 3.3V when powered by the same PSU.

Its all in the hypex datasheet. how the output is proportional to mains voltage.

Is it in the Purifi datasheet about this 4.8volts? Obviously there will be loss and it can't go rail to rail but I've not seen a specific figure. I may have missed it.

I make it 246 volts if this 4.8 volts is included.

Well Europe is Europe. the UK never moved from 240volts. Looks like Australia never did either.

 

[Matias]

The Purifi 1ET7040SA uses Hypex SMPS1200A400 or SMPS1200A700?

 

[boXem]

The Purifi 1ET7040SA uses Hypex SMPS1200A400 or SMPS1200A700?

A400. Otherwise I wouldn't make such statements

 

[Derkkk]

A400. Otherwise I wouldn't make such statements

If the SMPS1200A700 runs at 85v why isn't that PSU utilized? Seems odd if Purifi, for what ever reason, whether its other specs or even just what plugs are utilized to connect the power supply to the module that they would utilize or require a PSU to be used that doesn't leverage the full capability of the module. If the SMPS1200A700 is available and runs at a higher voltage (note my below statement) why wouldn't they design around that? I have no idea if the below specs for the A700 below that state 85Vdc are the voltage specs being discussed above. All curiosity, I have no knowledge of amp design. One of the simplest answers could be cost though, as you said above even if you could leverage the full capability of the module, normal 4ohm and up speakers would see little increase in performance so if the A700 PSU is considerable more expensive then I guess that makes sense.

Mains voltage input	230V / 115V
Automatic input voltage selection	No, Selectable with jumper
SMPS Voltage +/-	2x64Vdc (A400) 2x85Vdc (A700)
Max: Output power	1500W
Max: Output power @ 20Hz	1200W
Dimensions mm	165*105*55
Weight	750g

Note* I'm no engineer or have nay knowledge to the specs and what they mean or their function other than the basics. So take my question as nothing more than curiosity.

 

[boXem]

If the SMPS1200A700 runs at 85v why isn't that PSU utilized? Seems odd if Purifi, for what ever reason, whether its other specs or even just what plugs are utilized to connect the power supply to the module that they would utilize or require a PSU to be used that doesn't leverage the full capability of the module. If the SMPS1200A700 is available and runs at a higher voltage (note my below statement) why wouldn't they design around that? I have no idea if the below specs for the A700 below that state 85Vdc are the voltage specs being discussed above. All curiosity, I have no knowledge of amp design. One of the simplest answers could be cost though, as you said above even if you could leverage the full capability of the module, normal 4ohm and up speakers would see little increase in performance so if the A700 PSU is considerable more expensive then I guess that makes sense.

Mains voltage input	230V / 115V
Automatic input voltage selection	No, Selectable with jumper
SMPS Voltage +/-	2x64Vdc (A400) 2x85Vdc (A700)
Max: Output power	1500W
Max: Output power @ 20Hz	1200W
Dimensions mm	165*105*55
Weight	750g

Note* I'm no engineer or have nay knowledge to the specs and what they mean or their function other than the basics. So take my question as nothing more than curiosity.

Just my 2c, I have no information from Purifi on the matter.
Having the supply rails specified with absolute max voltage +/- 75 V DC like the 1ET7040SA allows to be excluded from the EU Low Voltage Directive scope, with all the advantages associated.

 

[restorer-john]

Having the supply rails specified with absolute max voltage +/- 75 V DC like the 1ET7040SA allows to be excluded from the EU Low Voltage Directive scope, with all the advantages associated.

Interesting.

 

[pogo]

I don't really understand the discussion here. Manufacturers with know-how will adapt the purifi amps built under licence to their needs anyway, whether power supply or DF, see also here:
Link

 

[Maverick AF]

The reason for the +-70v is because they wanted high power at lower impedances. Because the intention of this module is to run multiple subwoofers in parallel. Such as their upcoming 8, 10 and 12’ woofers. For car guys think of the voltage as horsepower, and the current as torque. This module needs more like a Diesel engine than a gas engine because it needs the low end torque. So this means low voltage and high current. 70V and 40A. The Hypex NC1200 is also a 40A module. Just it puts out 700w into 4 ohms rather than 500w. That’s because it requires an 84v supply.

For typical 4-8 ohm nominal speaker use, the 1ET400A is the better module to choose. Because you’ll get the same power to the speakers, and it measures better. Not to mention cheaper.

 

[Maverick AF]

And I don’t believe Purifi has intentions on selling this module to the DIY community.

 

[pogo]

I think there is still so much potential in the Purifi amp designs that still needs to be uncovered.
T+A has probably made a good start here by using non-eval boards, which means that they can guarantee the reliability of spec compliance for years. An optimization of the PCB design, including the elimination of connectors, etc., contributes to this.
A PSU like the Dynaudio Arbiter (battery powered) could also lead to surprises here.