I have an Onkyo HT-RC160 and I'm looking to see if it's possible to improve the sound from a digital source by swapping out a few components with higher quality parts, or if I would be wasting my time.
The basic signal path is as follows:
[Cirrus Logic CS42518 DAC]->[NJM4580 LPF]->[Chemicon 47/50 cap]->[Selector/volume/tone R2A15218FP]->[Chemicon 47/50]->[Driver]->[Output amp]
My idea is to replace the LCR channel 4580s with either a NE5532 or an OPA2604 (or...?), and the 47/50 caps with either Elna or Nichicon KZ. I'll probably also install audio grade caps in the power supply.
Can I expect an audible improvement in sound if I perform these modifications? Is the Cirrus DAC very good to begin with?
I use the receiver 90% of the time to listen to music off my computer.
The basic signal path is as follows:
[Cirrus Logic CS42518 DAC]->[NJM4580 LPF]->[Chemicon 47/50 cap]->[Selector/volume/tone R2A15218FP]->[Chemicon 47/50]->[Driver]->[Output amp]
My idea is to replace the LCR channel 4580s with either a NE5532 or an OPA2604 (or...?), and the 47/50 caps with either Elna or Nichicon KZ. I'll probably also install audio grade caps in the power supply.
Can I expect an audible improvement in sound if I perform these modifications? Is the Cirrus DAC very good to begin with?
I use the receiver 90% of the time to listen to music off my computer.
Last edited:
You will not get an improvement, but you might imagine one. This is very common on the audio tinkering hobby. Lots and lots of imagination. 😉
Nippon Chemi-con makes excellent capacitors, so there is no room for improvement there.
It's a modern DAC, so I don't see why it would be bad. There is no point in making a bad DAC these days, as the knowledge and technology is there.
Nippon Chemi-con makes excellent capacitors, so there is no room for improvement there.
It's a modern DAC, so I don't see why it would be bad. There is no point in making a bad DAC these days, as the knowledge and technology is there.
most likely you may experience some improvement with the opamp swap.
audio grade caps in psu are not going to yield any audible result most likely.
a better in general psu might, but on its own depends how well is the original one is made.
there is a level from where it will yield no result at all.
possibly bass could be enhanced if You fit bigger value caps to alter the highpass filter formed at the inputs, but even that is not so certain.
the question is more like what are Your goals to achieve ?
that can make a difference.
audio grade caps in psu are not going to yield any audible result most likely.
a better in general psu might, but on its own depends how well is the original one is made.
there is a level from where it will yield no result at all.
possibly bass could be enhanced if You fit bigger value caps to alter the highpass filter formed at the inputs, but even that is not so certain.
the question is more like what are Your goals to achieve ?
that can make a difference.
I have a TX NR 905 and spent some quality time with the service manuals, after noticing the sound quality from analog sources was not bad, but from digital sources it is unlistenable.
I just read the service manuals of your receiver. It is as bad as mine. It is digital audio technology from the 1990's.
You can do nothing to improve the sound quality from the internal DAC. Sell it and get a receiver designed by guys who know a thing about digital audio (ie, not ONKYO).
I just read the service manuals of your receiver. It is as bad as mine. It is digital audio technology from the 1990's.
You can do nothing to improve the sound quality from the internal DAC. Sell it and get a receiver designed by guys who know a thing about digital audio (ie, not ONKYO).
Hi,
Forget changing any power supply caps.
Can't comment on the coupling caps.
NE5532's are always worth a try IMO,
as they are miles better than any other
op-amp that costs less, by a large margin.
(At lower cost, same can be said about TL072's.)
rgds, sreten.
The above two op-amps "took off". Popularity
led to lower prices, making them even more
popular, and thus lower prices, etc.
Forget changing any power supply caps.
Can't comment on the coupling caps.
NE5532's are always worth a try IMO,
as they are miles better than any other
op-amp that costs less, by a large margin.
(At lower cost, same can be said about TL072's.)
rgds, sreten.
The above two op-amps "took off". Popularity
led to lower prices, making them even more
popular, and thus lower prices, etc.
Last edited:
Don't just go changing opamps without checking for oscillations with a scope. Especially when changing 'slow' opamps with very fast ones. The improvements you may be hearing icould be actually oscillations.
Also, the more or difficult the labor you performed in "upgrading" something, the more is the perception that you improved the sound,
Also, the more or difficult the labor you performed in "upgrading" something, the more is the perception that you improved the sound,
You'll see what you want to see. If you'd like to see more, learn up about how modern DACs work. You may well discover that the dynamics of chaotic noise-shaping feedback loops are not that well understood. But what the hell - they deliver the required numbers.
But there's no will to use all that knowledge and tech. Hence poor DACs which measure extremely well with the current measurement set.
Thank you all for your responses. Basically I'm trying to achieve better clarity with "large" signals. The sound seems to get muddy or harsh with a 'loud' source signal, but like peufeu mentioned, when I listen to vinyl on the receiver through an external preamp it sounds better. I suspect the DAC is probably the weak link.
This is not the reason why these receivers suck so much. Granted, there are better DAC chips than CS42518, but I'm sure that a decent implementation of it would sound nice. Of course you can't use SPDIF with it and get usable results, since it has a Cirrus PLL in it, which is kind of annoying if you want SPDIF inputs...
The reason is that ONKYO's implementation piles up layer upon layer of FAIL. Everything that they could fail, they did, and then they went to great lengths to invent new ways to fail even more.
Here's how it works in the TX-NR905 :
Ethernet media player : Inside the receiver, the ethernet media player is on a stand-alone board which has some fat RISC CPU, off-chip flash, SDRAM, PCI ethernet chip... almost a complete PC... with ... a PCI Envy24 sound chip on it. This chip has 2 xtals, one for 44.1k based sample rates, and one for 48k based rates. And the thing outputs some SPDIF which is routed to the SPDIF input mux.
Let's follow the trace to the DACs !
First some traces and cheap connectors, through various boards : microcontroller board, Speaker terminal board, DAC/DIR Board... it goes everywhere...
A little MUXing with the other SPDIF inputs (74VHC157)
A SPDIF DIR (CS42516, a DIR especially notable for its jitter attenuation performance which is precisely zero)
Then, let's follow the MCK coming out of this DIR :
- crap connector to DSP Board
- a little MUXing with other clocks coming from HDMI board (the mux is of course shared with audio data)
- some traces (and long stubs to a fat DSP)
- a bit more MUXing with other clocks (the mux is again shared with other crap)
- crap connector to DAC/DIR Board
- and finally, alleluiah, the MCK reaches the DAC pins.
Of course, there is no ground plane on the DAC board, whose layout is horrendous at a vomit-inducing leveI. The boards are sandwiched so the fat DSP with its noisy memory busses sits just on top of the high-end NE5532 opamps. All the "crap connectors" noted above have something like 1 ground pin for 20 fast I2S signals (inc. MCLK) ... Repeat : the DAC board layout is worse than the worst DAC board any of you guys have ever seen in this forum. Let that sink in, lol.
I tried to probe this board. Since it has no real ground, no matter where the probe GND and tip are placed, there is always a huge amount of HF noise on every signal, on every trace, everywhere.
When using HDMI as an audio input, well, let's just say the HDMI receiver chip specifies "0.1UI jitter" on its recovered audio clock, and it is fed as-is to the DACs (I'll spare you the signal path, it's quite convoluted too). When using SPDIF over HDMI, same thing.
They could at least have implemented a decent local clock on ethernet playback, but no.
I was unprepared to this level of failure.
Check out some measurements :
http://www.diyaudio.com/forums/digi...nesses-common-ht-receivers-4.html#post3836253
The only way to fix these receivers is to use an external DAC.
Trying to "fix" the internal digital audio subsystem is like trying to "fix" a Trabant.
Last edited:
Wow, I didn't realize onkyo designed the dac section so poorly. It doesn't look like there's much I can do to help the sound.
Who makes receivers with good digital sections?
Who makes receivers with good digital sections?
Last edited:
I don't know. In the other topic good things were said about Pioneer.
It's a shame about the Onkyo because the analog parts sound good...
It's a shame about the Onkyo because the analog parts sound good...
Yikes. I didn't know it could be that bad. That's a real shame for sure, the power amps in the NR905/SR875 are pretty fancy (OK, I don't like the 1970s Elektor style bias much, but they've got two pairs of output transistors and cascoded LTP input and all that jazz), and it's even got a decent headphone amp.
Current Onkyo models may be entirely different inside, as they stem from the TEAC collaboration. They're not super powerful or super fast, but our NR616 runs only lukewarm in 4 ohm mode (electronic rail switching?).
Pioneer isn't a bad idea, at least they tend to be very powerful for the price. They used to have awful user interfaces and remotes though, no wonder they didn't make a dime. (A '922 went straight back because of such issues.) Looks like that's been addressed in current models though.
Current Onkyo models may be entirely different inside, as they stem from the TEAC collaboration. They're not super powerful or super fast, but our NR616 runs only lukewarm in 4 ohm mode (electronic rail switching?).
Pioneer isn't a bad idea, at least they tend to be very powerful for the price. They used to have awful user interfaces and remotes though, no wonder they didn't make a dime. (A '922 went straight back because of such issues.) Looks like that's been addressed in current models though.
I've decided to look for a <$300 DAC/headphone amp with a line out that can connect to an analog input of the receiver. There are a bunch of options out there but ones that caught my attention are the JDS O2+ODAC combo (said to work well with my HD650's), Maverick TubeMagic D1, the Schiit modi/magni... I'm sure there are others I'm not aware of.
With the O2+ODAC I like the fact you can order all the unassembled parts to build it yourself, and it knocks $100 off the price of a completed unit. It's USB only but I don't have any other digital sources besides my computer.
With the O2+ODAC I like the fact you can order all the unassembled parts to build it yourself, and it knocks $100 off the price of a completed unit. It's USB only but I don't have any other digital sources besides my computer.
Last edited:
Member
Joined 2009
Paid Member
I started the thread referred to above. I've been looking at improving the sound of my HT system for years. I started with dedicated amps driven from the analogue outputs of my Blueray player but this meant I had to supply my own volume control and I had no room correction or many of the other conveniences my family wanted. So I bought a receiver.
I bought the service manual for the receiver. It's clear that the power amplifier is the weakest link. Sure, there are better DACs or better volume control chips and probably a few other items but nothing major.
The power amps are not good. They are crippled from the start by various cost saving measures.
a) the power supply has insufficient capacitance and does not use low impedance wiring to connect to the amplifier board
b) the heatsink is not that big and the amplifier has many channels crowded onto it so it can't be set up to generate much heat - it is under-biassed so distortion is higher than it need be
c) the amplifier topology is very basic, almost poor and easily bettered.
I suspect the best improvement would be to take off the signals before they reach the power amplifier and bring them out to external power amplifiers.
The output mute relays are also inadequate to protect a speaker if the amplifier develops a dc fault - they do not switch the speaker terminal to ground to extinguish an inductive arc, they have no catch diodes and are underrated in terms of current handling. An external amplifier with proper dc protection would help.
This is the direction I'm taking at the present time. I'll simply convert my receiver into a home theatre pre-amp and retain all the conveniences but add external power amplifiers of my own design and construction.
I bought the service manual for the receiver. It's clear that the power amplifier is the weakest link. Sure, there are better DACs or better volume control chips and probably a few other items but nothing major.
The power amps are not good. They are crippled from the start by various cost saving measures.
a) the power supply has insufficient capacitance and does not use low impedance wiring to connect to the amplifier board
b) the heatsink is not that big and the amplifier has many channels crowded onto it so it can't be set up to generate much heat - it is under-biassed so distortion is higher than it need be
c) the amplifier topology is very basic, almost poor and easily bettered.
I suspect the best improvement would be to take off the signals before they reach the power amplifier and bring them out to external power amplifiers.
The output mute relays are also inadequate to protect a speaker if the amplifier develops a dc fault - they do not switch the speaker terminal to ground to extinguish an inductive arc, they have no catch diodes and are underrated in terms of current handling. An external amplifier with proper dc protection would help.
This is the direction I'm taking at the present time. I'll simply convert my receiver into a home theatre pre-amp and retain all the conveniences but add external power amplifiers of my own design and construction.
Last edited:
Bigun, I use an Integra 30.2 for HT , but I use a Nad 214 poweramp on the preouts for the main speakers. Doing it this way releases the power supply of the main speakers and it only has to drive the centre and sorround speakers. Works great !
Member
Joined 2009
Paid Member
Hi Jan, that's a good idea, it's not necessary to go with the best amplifier for the surrounds for starters. Mind you, the centre channel is critical because it carries most of the speech.
Actually, my plans are a little more evolved - I plan to try a couple of additional tweaks to the signals between the receiver pre-out (which I have to create as my receiver doesn't have any pre-outs) and the power amplifier
i) allow me to mix the sub-woofer channel into all of the speaker channels - I use large speakers capable of reproducing low frequencies and I want to utilize this to get a variation on the 'multiple sub' scenario
ii) allow me to mix some of the centre channel into the L & R mains. I have a BIG screen so my mains are widely spaced and I need to move some of the speech content to the L & R to get a better sound field.
Actually, my plans are a little more evolved - I plan to try a couple of additional tweaks to the signals between the receiver pre-out (which I have to create as my receiver doesn't have any pre-outs) and the power amplifier
i) allow me to mix the sub-woofer channel into all of the speaker channels - I use large speakers capable of reproducing low frequencies and I want to utilize this to get a variation on the 'multiple sub' scenario
ii) allow me to mix some of the centre channel into the L & R mains. I have a BIG screen so my mains are widely spaced and I need to move some of the speech content to the L & R to get a better sound field.
- Status
- Not open for further replies.