AVR modification; in this case, the Anthem MRX series

ever wondered what's inside your audio video receiver?
here's the inside of Anthem's MRX series receiver.

the usual power transformer, poweramp circuit, hdmi receiver board and etc etc.

the hdmi receiver board is in the way of the much more interesting analog section and DAC. let's remove it.

on the far left is the anthem dock board. the large board below the anthem dock board is the DAC and analog section.

here's the hdmi receiver board if anyone is interested:


the analog board and DAC section:


the avr is using a multichannel ADC-DAC from Cirrus Logic (CS42528). the specs are nothing to shout about. definitely not flagship, but like most other makers nowadays, anthem is implementing a multichannel DAC to reduce costs.
but hey, a flagship DAC sitting inside a crappy implementation would still sound more or less, like crap. when it comes to DAC, it all boils down to how good the implementation is.

enough with all the talk. let's move on to more important things. like the work itself. all talk no work no good.

as anthem's mrx series shares the same dsp board, analog section board and hdmi receiver board across all the models in the series, this modification is applicable to the mrx300, mrx500 and mrx700.

the 2 most significant parts that could be upgraded is the signal decoupling capacitors (DC blocking caps) at the post DAC low pass filter section here:

and at the output buffer section here:


both sections are using solid electrolytic caps valued at 10uF/16V. and as Walt Jung suggested, capacitors are not made the same. especially when it comes to signal decoupling. electrolytics are best avoided for signal decoupling purposes and best to stick to a good film cap like polypropylene, polystyrene or polyester variant.
surely, talkers would tend to say there's no difference in using whatever cap even if it's made with pure turd but talkers are always talkers, never having the knowledge nor expertise to desolder and resolder another cap. an electrical engineer on the other hand, are not talkers and swapping caps are like a walk in the park for them.

my advice is to try swapping it out and if you don't like it or don't hear any difference then could always revert it back. that is, IF you're capable enough. this ain't childs play.

swap it all out for a good quality film and you'll eventually come to this:

i ran out of those 1uF Vishay-Roederstein MKT1822 and had to mix it with some polypropylene for the rears.

and before any sifu come in here and shout something about phase distortion or signal degradation or poor impedence of the film or cable inductance from all that wire and all that crap, think i'll duck my head under my table.

well hey what do you know. i have an RM20k scope right under my feet this whole time. must be a present from santa.


nope, can't see any signal degradation at 20kHz. same for left and right.


looks the same at 80Hz.


wait there's something going there at 40Hz.


right on. there's some phase distortion and bass roll off at 20Hz.


even more phase distortion and bass roll off at 10Hz.

note: the previous waveform are taken with only 1 channel modded, meaning only 1 channel changed to film caps whilst another channel is left with stock capacitors. not a valid comparison if all the caps are already swapped out and replaced, no?

the measurement are taken with the output buffer swapped with 1uF polyester film and the low pass filter swapped with 0.47uF polyester film. further explaination below.
and sorry no pics for this stage as i'm not your slave. just have to trust me on this.

the output buffer section have an impedence of 100 kiloohm to and from the opamp buffer. at 100kohm impedence, 1uF would provide a corner frequency of 1.59Hz and a 0.47uF would provide a corner frequency of 3.38Hz

the low pass filter section on the other hand, have an impedence of around 80kiloohm. a 1uF would fit better here, as it provides a lower corner frequency, but for the heck of it, i tested the amp first with 0.47uF at the filter, providing a corner frequency of 4.23Hz

thus you have the previous waveform. there's a slight phase distortion and signal rolloff caused by the 0.47uF. tho the corner frequency is lower than 10Hz, but the phase is stained up till 24Hz. it goes back to normal at 25Hz.
tho this might not be of any issues since the left and right speakers would have their signal cutoff at 80Hz to the sub, it might have some mental effect to clueless audiophiles.
and furthermore, i doubt there's any passive speakers that could produce bass signal below 30Hz flat. the speakers itself would have rolled off all that signal before reaching 25Hz.

could be easily rectified by not using any 0.47uF caps (or any lower than 1uF) for replacement of the solid electrolytic. by using only 1uF, you would have this kind of waveform:

flat with some minor phase distortion at 20Hz compared to the original circuit using the electrolytic.

after all of that, you redo the calibration and sit back and enjoy your hard work. this mod removes the brutal and uncontrolled signature of the anthem and produce a more musical tone to it.

a more musical tone beating the marantz sr6003 in stereo reproduction. as the anthem in stock condition is already good in movies, this makes it an even better amp for both movies and music.

and once you're done with that maybe you would consider going to the extreme and swapping out the original Japan Radio Company's JRC2068 opamp used for the filter and output buffer and replacing it with some BurrBrown goodness.

and while you're at it, why not bias everything into Class A as it takes only a resistor and a low noise JFET to do the job. an audio grade low noise JFET from toshiba is selling for only RM1 each at jalan pasar. the cost is so low that it's stupid not to do it, right? could always remove it if you don't like how it sounds like (like that's even gonna happen).

and to top it all off, why not build a nice PCB to house all the mods done. do so and you'll eventually wind up with this:


a Class A biased anthem MRX with socketed Burr Brown OPA2111 opamp and Vishay-Roederstein MKT1822 1uF caps for the front and center and vanilla 1uF polyester bypassed with Vishay MKP1837 polypropylene caps, coupled with Nippon-Chemicon SXE series low impedence power caps.

and you have to ask, why even take such an effort,
i'd say because the anthem in stock condition is beaten in some area to the avr it's supposed to replace.

as the previous avr is also heavily modified, it only makes sense to try and improve the anthem in areas it loses out, mainly the center channel openness and integration between the front speakers and back.

and even after all this mod, the anthem still loses at the center channel openness. i can live with that since the other amp have far better power regulation (think of super regulator, a few of it) and lots more work (and time) put into it. it requires time to tweak (mod) the anthem to the level of the other amp.

and if you're looking for an easier way to the modification path, a friend pointed out a company that does similar service: http://www.upgradecompany.com/ tho it's in US.
and no i'm not associated to the company. and no i don't provide this service to anyone.

enough with all the talk. let's move on to more important things. like the work itself. all talk no work no good.

Yes, enough of the talk. Show us the work.
Photos/pictures say a thousand words.

does it really worth the mod where u r still using the same power supply?

sph wrote:

enough with all the talk. let's move on to more important things. like the work itself. all talk no work no good.

Yes, enough of the talk. Show us the work.
Photos/pictures say a thousand words.

pictures are already there, ain't it?

alfred wrote:does it really worth the mod where u r still using the same power supply?

the power supply is already linear PS and not switched mode. tho power supply have some effect to the final sound signature, it's not as significant as upgrading the signal decoupling caps.

implementing a dual tracking super regulator for this anthem would take some time as the space is quite limited. surely not a walk in the park then.
and not to mention the anthem is already using tantalums as bypass. it would take a great PS to get a good audible improvement from a PS upgrade.

I don't seem to be able to see the photos. My apologies.

Perhaps my settings are not correct. Can someone help me with this?

sph wrote:I don't seem to be able to see the photos. My apologies.

Perhaps my settings are not correct. Can someone help me with this?

the images are there. all are uploaded to imageshack and linked here. perhaps you should try refreshing the page a few times of using some other browser.