• 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!

Dayton Audio APA1200DSP Teardown (DSP Amplifier)

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,420
Likes
179,177
Location
Seattle Area
I just reviewed the Dayton Audio APA1200DSP amplifier. This is a teardown of the same. A few screws hold the top after which we see all of its bits:

Dayton Audio APA1200DSP teardown Class D Amplifier board.jpg

[click on image for larger size].

Everything seems to be custom design. The large board has a front-end switching power supply which then powers the stereo class D amplifier. Based on size of the parts, it doesn't seem like it can pump out 1,200 watts of amplification.

I did not like that the power supply rectifier was glued to the electrolytic capacitor next to it. In switching supplies, these rectifiers get very warm and you don't want it to cook the capacitor.

Nice to see high voltage wiring clearly kept away from signal wires. The chunk of heatsink in the middle with side fins is much better than typical Hypex amplifier builds. On top of that you have a fan so this thing should run pretty cool.

Speaking of the capacitor, the brand name is "Jamicon." Not exactly a top brand capacitor although it tries hard to associate with them using that "con" ending.

The board on the top left is the main brains handling DSP functionality, codecs and such. Here is a more focused image:
Dayton Audio APA1200DSP teardown DSP board.jpg


Analog devices ADSP-21479 DSP handles computing jobs. See https://www.analog.com/en/products/adsp-21479.html

A pair of Cirrus Logic CS4270 "codecs" (ADC and DAC) handle conversion to and from analog domain: https://www.cirrus.com/products/cs4270/

DAC THD+N spec is 87 dB SINAD. ADC is better at 95 dB.

Conclusions
Interesting to see a full custom solution in this amplifier. This has not come cheap unless they expect huge volumes. Parts quality is average but construction is above average compared to homebrew module based amplifiers we see around.

-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
14,829
Likes
14,550
Location
Canada
Thank you for the teardown. I still say it does a lot of things. It would be a great tool to have in the golf bag/quiver of arrows for different peeps looking for something. It's just those test results holding it back. Peeps said a couple of times that a amp with DSP PEQ and a DAC integrated in it is what is needed.
 

GWolfman

Addicted to Fun and Learning
Forum Donor
Joined
Nov 3, 2020
Messages
520
Likes
754
DAC THD+N spec is 87 dB SINAD.
And you measured 72 dB. Seems like such a loss, but guessing a combination of poor implementation and "ideal" (i.e., maximum) specs from AD (the manufacturer).
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,420
Likes
179,177
Location
Seattle Area
And you measured 72 dB. Seems like such a loss, but guessing a combination of poor implementation and "ideal" (i.e., maximum) specs from AD (the manufacturer).
The 72 dB is for the DAC and amplifier combined. I did not measure the DAC by itself. The job here is to be much better than the amp and it seems to be the case.
 

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
1,856
Likes
1,709
Location
Sweden
And you measured 72 dB. Seems like such a loss, but guessing a combination of poor implementation and "ideal" (i.e., maximum) specs from AD (the manufacturer).
Hmm… which class D chip is it based on ? And maybe its the dsp alone thats holding it back ?
 
Last edited:

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
14,829
Likes
14,550
Location
Canada
Hmm… they could have used a tpa3255 chip with ( maybe ) much better performance , but maybe its the dsp thats holding it back ?
Any idea what the TPA3255 outputs in mono into 8Ω? The datasheet is not very helpful that way.

tpa.png
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
3,296
Likes
6,860
Location
Prague
Build quality is better than we usually see here. Jamicon caps are not the best brand, but they are just good, average product. The obligate SINAD of 72dB would probably still remain inaudible (yes noise is quite high), much worse thing is the frequency response heavily modulated by load impedance at high frequencies and this will be audible.
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
14,829
Likes
14,550
Location
Canada
In practical use, less than 200 Watts in 8 Ohm with 48 V 6 A . Higher output than that and the chip goes in protection mode .
Is that a RMS 200 watts?
I ask because for the amp to put out 200wRMS @8Ω the rail voltage would need to be ~56.6V DC @ ~7.1A peak.
If using a 48V DC @6A supply the amp could put out ~144wRMS into 8Ω.

Where did you get the [email protected]Ω, w/48V 6A power supply figure from?
Maybe I'm missing something but I think I am not. :D
They make all of this so convoluted with the unconventional specs that they provide in the datasheet and elsewhere too.
Regardless of that mumbO jumbO it's impressive that a SMD chip can put out this much power. :D
 

GWolfman

Addicted to Fun and Learning
Forum Donor
Joined
Nov 3, 2020
Messages
520
Likes
754
The 72 dB is for the DAC and amplifier combined. I did not measure the DAC by itself. The job here is to be much better than the amp and it seems to be the case.
Yes, understandable.
 

Vladimir Filevski

Senior Member
Joined
Mar 22, 2020
Messages
392
Likes
485
Any idea what the TPA3255 outputs in mono into 8Ω? The datasheet is not very helpful that way.
From datasheet: "150 W stereo into 8 Ω BTL configuration".
In mono PBTL configuration it will be essentially the same - 150 W into 8 Ω. Parallel configuration is useful only with low impedance load.
 

Azathoth

Member
Joined
Oct 18, 2020
Messages
95
Likes
208
It kinda tickles me that a lot of these devices are just like a bag of chips. Lots of air in them.
 

Harmonie

Major Contributor
Forum Donor
Joined
May 30, 2020
Messages
1,930
Likes
2,019
Location
France
Teardown is the right expression.

Tear it down and keep just the board with the DSP functionality, codecs and the case and connections ...
 
Top Bottom