• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Marchaudio P422 Stereo Power Amplifier Renewed Review

JohnYang1997

Master Contributor
Technical Expert
Manufacturer
Joined
Dec 28, 2018
Messages
7,074
Likes
17,177
Location
China
Good work! One thing I would point out is that, here you are using 20khz low pass filter setting which causes slight deviation in measured noise thus SINAD from Amir's 22.4khz low pass setting. On the other hand, Wolf has always used 20khz setting. So I would suggest either switch to 22.4khz low pass or if you are sticking with 20khz, it would be better to mention the difference of the filter setting and what difference it may lead to.
 
OP
thin bLue

thin bLue

Active Member
Reviewer
Joined
Jul 29, 2019
Messages
203
Likes
605
Yes, but AFAIK, RF filtering is part of the EVAL1 and most other buffers based on it.

So, good but not a major differentiator.
20220628_194124.jpg


This is input RF filter of my P422.

As I know most of EVAL-1 based amp has no RF filter board on the back of the XLR. Am I wrong?
 
OP
thin bLue

thin bLue

Active Member
Reviewer
Joined
Jul 29, 2019
Messages
203
Likes
605
Good work! One thing I would point out is that, here you are using 20khz low pass filter setting which causes slight deviation in measured noise thus SINAD from Amir's 22.4khz low pass setting. On the other hand, Wolf has always used 20khz setting. So I would suggest either switch to 22.4khz low pass or if you are sticking with 20khz, it would be better to mention the difference of the filter setting and what difference it may lead to.
Thanks for great feedback!!!!
I got another improving points!

Technical support from competent engineer like you helps a LOT!!!!
I'll update them next time!
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
3,313
Likes
6,904
Location
Prague
As I know most of EVAL-1 based amp has no RF filter board on the back of the XLR. Am I wrong?
The point is to have the RF filter as close to the input connector as possible. Best right on the pins. Otherwise the fast impulse spikes are able to bypass it on the PCB. Audio devices are often not designed according to RF design rules. They should be, because though audio signals are slow, interference signals are very fast.
 
OP
thin bLue

thin bLue

Active Member
Reviewer
Joined
Jul 29, 2019
Messages
203
Likes
605
The point is to have the RF filter as close to the input connector as possible. Best right on the pins. Otherwise the fast impulse spikes are able to bypass it on the PCB. Audio devices are often not designed according to RF design rules. They should be, because though audio signals are slow, interference signals are very fast.
What a kind explanation!
Thank you!
 
  • Like
Reactions: pma

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
The point is to have the RF filter as close to the input connector as possible. Best right on the pins. Otherwise the fast impulse spikes are able to bypass it on the PCB. Audio devices are often not designed according to RF design rules. They should be, because though audio signals are slow, interference signals are very fast.

March posted in another thread that the EVAL1 is “fine” with regard to RF filtering.

Have no other details to add as do not have a board on hand.
 
  • Like
Reactions: pma

Matias

Major Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
3,590
Likes
6,768
Location
São Paulo, Brazil
My point is not about the power capacity but the distortion. The column header says at the shown power value distortion is 1%. But, it isn’t. Amplifier either shuts down or clips. In the former there is no distortion as there’s no signal, in the latter it is a value read from a chart that shows hard clipping. Neither value is in anyway an operating value.
Would adding an asterisk to the column header and a side note "* or limited by the amplifier for protection before it reaches 1% THD+N" satisfy?
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
10,016
Likes
28,638
Location
Gold Coast, Queensland, Australia
No arguing from my side. I was speaking about general design rules to minimize RF impact.

Absolutely. Nothing groundbreaking about placing the RF filtering passive components on the input board, right on the XLR, or in the case of half a century of commercial gear, on the RCAs or right behind them on the PCB.
 

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
View attachment 230673

This is input RF filter of my P422.

As I know most of EVAL-1 based amp has no RF filter board on the back of the XLR. Am I wrong?

An EVAL1 has the XLR connectors mounted directly on the board. Based on March's post, the board has RF filters.
 

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,268
Likes
2,353
Location
Southern California
Good work! One thing I would point out is that, here you are using 20khz low pass filter setting which causes slight deviation in measured noise thus SINAD from Amir's 22.4khz low pass setting. On the other hand, Wolf has always used 20khz setting. So I would suggest either switch to 22.4khz low pass or if you are sticking with 20khz, it would be better to mention the difference of the filter setting and what difference it may lead to.
Are you guys working on a premium amplifier to take on these biggies?
 
OP
thin bLue

thin bLue

Active Member
Reviewer
Joined
Jul 29, 2019
Messages
203
Likes
605
An EVAL1 has the XLR connectors mounted directly on the board. Based on March's post, the board has RF filters.
Recent reply I got from March audio would help.

20220915_161802013.jpg


Hi @thinblue

Thank you for the re-worked P422 review, the results are now very much as we would expect them to be. Great job ! There are many pitfalls that can occur and it takes a bit of time to learn the correct techniques to obtain accurate data and you have done well .

Just a few observations. I know you have mentioned this in your text, but it’s worth re-iterating that the results are fundamentally limited by the capabilities of your AP525 analyser. With its own measurement limit at 109dB SINAD it is really no better than the device under test (DUT), so we are as much measuring the performance of the AP525, as we are the P422. We really need a measurement system that is at least 12dB, and preferably 18dB better performing than the DUT to gain reasonably accurate results.

For information, as your main measurements were in dBV, below is a measurement of the P422 performance in dBr to be more consistent and comparable to the ASR measurements. It was taken from a unit going through production test this morning. It is a typical result. Please note that our production test measurement system also is not as good as the reference AP555 used by ASR. If the unit was measured on an AP555 it would be showing SINAD >107dB.

It achieves 106.9dB THD+N. Note that the result is extremely clean with no hint of mains spuria. Amp gain is 20.5dB suited for 4 volt XLR sources. Whilst it is not mentioned on our product page, the input buffer already has selectable 26.5dB and 20.5dB gain options. This can be operated via a jumper on the board. When we manufacture another batch of enclosures in the next few months this will be operated via an external switch on the rear panel. Note our gain settings will actually achieve full amp output by 2 and 4 volts RMS input – some other amps from elsewhere don’t cough….(hint – think about effect of gain on noise levels and final SINAD ).

다운로드.jpeg


As you noted in your review, our new (non Hypex) power supply is working very well, as demonstrated by the high 2 ohm output power. The PSU is very low noise, doesn’t suffer bus pumping and maintains output voltage even with high load and reduced mains voltage. This means, unlike other Hypex based units, it will achieve full power rating regardless of what the mains is doing.

Regarding Bruno Putzeys comments, yes he did say that our input buffer was more transparent than his Eval1 board. Purifi have had our products at their lab and used both our amps and speakers in their demo suite at the recent Munich High End show. We think this is a great tick of approval .

Claus Neesgaard – Purifi
다운로드.png



Regarding the RF filtering, yes the Purifi Eval1 has bandwidth limiting, but not an actual RF filter at the XLR input pins.

We have implemented an RF filter right at the XLR pins plus appropriate bandwidth limiting in the buffer. This in itself is nothing new, just good design practice, however other manufacturers do not appear to be doing it. Some manufacturers even still use the Hypex eval board as a buffer which has zero input filtering, and its bandwidth is wide open. It’s a radio receiver!

Anyway, thanks again for your good work. I will talk to you about how you can cost effectively improve your measurement system beyond the capabilities of the AP525 in a cost effective way.

Cheers
Alan
 
OP
thin bLue

thin bLue

Active Member
Reviewer
Joined
Jul 29, 2019
Messages
203
Likes
605
I forgot to mention there was technical supports E1DA Discord Channel(@IVX, @staticV3 ). Now added!
 
OP
thin bLue

thin bLue

Active Member
Reviewer
Joined
Jul 29, 2019
Messages
203
Likes
605
Thanks @thin bLue

Is it possible to test the big boy - March Audio P501 1ET7040SA next?
I thought P501 is over kill, so I didn't think about that seriously.
But now, I think I should! I'll contact to Marchaudio.
 
Last edited:
OP
thin bLue

thin bLue

Active Member
Reviewer
Joined
Jul 29, 2019
Messages
203
Likes
605
Thanks. It would be interesting to see how it compares to the boxem 4222/E1, particularly in terms of output power since March Audio is using this Micro-Audio SMPS instead of the Hypex.
Please check this out!
 
Top Bottom