Cheap component in the Rotel 1098 !!!!

[microfast]

All do known the good DAC CS43122 and the op-amp OPA2604, used for the output stage for the front channells and ( in a more simplified circuit ) for the center and the subwoofer channells, in the RSP1098.

This is good, but anyone have never said that are been used also two NE5532 ( cheap, but decent op-amps ) for the output stage for the rear and back channells, and ten TL072C ( very cheap and noisy op-amps ) for the multichannels input, the others analog inputs, and directly after the DACs as buffer ( on the little dac board ) !!!!!

Opinions on this ?

Regards
Marco

[Ossi]

So how about replacing them? Just need to keep the original ones in case of a service problem And while we are on it, let's replace some caps by Black Gate or the Panasonic FC.
Has someone got the shematics of the rsp 1098 to see where replacing would be of a sonical advantage?

[AlvaroD]

For what is worth,

speaking about electronic components, specially op-amps, it is my experience that cheap does not necessarily means bad as expensive doen't necessarily means good, sound wise.

Rotel is well know to choose whatever component sounds good for the minimum price. I shall trust them on the component choice. The final result is the only one that matters to my ears

[microfast]

So how about replacing them? Just need to keep the original ones in case of a service problem And while we are on it, let's replace some caps by Black Gate or the Panasonic FC.
Has someone got the shematics of the rsp 1098 to see where replacing would be of a sonical advantage?

The use of the TL072Cs in a pre/dec of this class is simply a shame.

The bad thing is that there is not a path where the signal do not cross these 0,3 $ chips.

I have already buyed several smd OPA2604 and i am almost ready to heat the solder !!!!!!!!!

I have not the schematic but will be very useful ..........

The more accurate test have said that the integral resolution is not better that 15.6 dB. , the S/N ( A ) on the multichannels input 95 dB. and the noise floor no better than -130dB. ( this value tested a 24 bit 48Khz. is modest; for example in a Denon AVCA1SE is -140dB or better ).

Regards
Marco

[eelco74]

Well,

This is what it is all about, the Denon might have better measurement figures. But all of the Denon amps and Reveivers sound flat and muddy.

I have tried several Denon reveicers, including the AVC-1SR and the AVC-11SR. I own a 3801, wich I use as a processor only. None of these sound as good as the 1066/1075/1080 combo.

Rotel is very carefull in choosing the right comonents. The excellent tests and reviews confirm this.

[microfast]

So how about replacing them? Just need to keep the original ones in case of a service problem And while we are on it, let's replace some caps by Black Gate or the Panasonic FC.
Has someone got the shematics of the rsp 1098 to see where replacing would be of a sonical advantage?

The use of the TL072Cs in a pre/dec of this class is simply a shame.

The bad thing is that there is not a path where the signal do not cross these 0,3 $ chips.

I have already buyed several smd OPA2604 and i am almost ready to heat the solder !!!!!!!!!

I have not the schematic but will be very useful ..........

The more accurate test have said that the integral resolution is not better that 15.6 dB. , the S/N ( A ) on the multichannels input 95 dB. and the noise floor no better than -130dB. ( this value tested a 24 bit 48Khz. is modest; for example in a Denon AVCA1SE is -140dB or better ).

Regards
Marco

[Scarp]

Can you actually hear thats its bad? If you can't hear it, who cares?

I wouldn't care less if there where only components in it that are no more advanced as my alarm clock, if it still sounds as good as it does.

Component specs say very little to none about the total quality and sound.

[Tha Freak]

does your ears tell you that??




- - - - - - - - - -

"Are you gonna bark all day little doggy?...or are you gonna bite?
-Mr. Blond in Reservoir Dogs

[microfast]

Can you actually hear thats its bad? If you can't hear it, who cares?

I wouldn't care less if there where only components in it that are no more advanced as my alarm clock, if it still sounds as good as it does.

Component specs say very little to none about the total quality and sound.

Excuse me for my bad english ........

I have not said that the 1098 sound bad, but the story is not so simple ....

Rotel had said ( boasts of this ) that use first choice DAC ( the CS43122, and this is true ) and a very "musical" op-amp ( BB/Texas OPA-2604, and also this is true ) and then without any rational reason put between theme this universally acclamed as a very cheap and "second-rate" op-amp, good for a walkman.

Good(DAC)+second-rate(buffer)+musical(Output stage): i do think that even if do not sound bad was possible do it better : Good(DAC)+good(buffer)+good output stage, this is very simple.

Also put these op amps in the multichannels input ( the only way to enter from a SACD or DVD-AUDIO player, because there is not any I-LINK input ) is not a good choice.

From years there is no more a quality pre/decoder that use these TL072Cs in so critical sections, even use the modest NE5532 was a better choice.

Is not possible that for 15-20$ of components, if you want a perfect product, without any compromise, you must buy a Lexicon, a Parasound etc etc .............

My previous pre was a Denon AVCA1-SE ( 5800 ), using already external power amplifiers, and i have found some improvment, but very little.

If you are interested i will do known the result of my test.

Regards
Marco

[Ossi]

@ microfast
I am very interested in your results.
I have myself sucessfully overworked a sony CDplayer xa50es and relaced all relevent parts by panasonic FC caps, new diodes for power supply and yes, the OPA 2604 and OPA2107 with a less sharp filter design made in all a tremendious difference. However I will see how much balck gate caps still can bring of an improvement.
It needs to be remenbered that parts that an owner can change will not cost much, but chosed with wisdom that will have a big impact.
The critical thing is to see where changing will actually bring a benefit.
you will in most cases need the shematics and be able to read those which is not at the hand of everyone.
I do not plan to get a 1098 but possibly a 1068 or sherwood p-965.
And I am pretty sure that for a couple of $$ of better parts those will greatly benefit.

[Scarp]

Microfast,

My question remains, does the 1098 sound bad? If you are missing something compared to others in its class, then maybe this will improve it, otherwise I think its too risky.

Does the 1098 sound better than your previous Denon? I would expect that or otherwise you wouldn't have bought it.


A warning to everybody here:
you will loose your warranty if you modify your 1098. Rotel will not even repair your 1098 if you modify your 1098 too much.
Also any modifications posted here are at your own risk and are not the responsibility of Club Rotel.

[DrBoom]

Maybe these so called "bad" components are needed in the design, just like a certain capacitor in speaker crossover could be replaced with a far better one, but giving worse results because the speaker was designed with the losses of that particular capacitor in mind ?

My opinion, if stuff like this bothers you, buy something like a Mark Levinson N°40, Lexicon MC12 or Classe SSP75 which costs a fortune but uses (lets hope so) first grade components all around.

I'm not saying it isn't a compromise to use a cheaper component, but I think it's a well thought-out compromise.
If they would have to use first grade everything, it would've probably cost 2x or 3x as much as it does now.
Besides, all the best or most expensive components in the world mixed together don't make a perfect product.
It's all in the mix

[microfast]

Maybe these so called "bad" components are needed in the design, just like a certain capacitor in speaker crossover could be replaced with a far better one, but giving worse results because the speaker was designed with the losses of that particular capacitor in mind ?

My opinion, if stuff like this bothers you, buy something like a Mark Levinson N°40, Lexicon MC12 or Classe SSP75 which costs a fortune but uses (lets hope so) first grade components all around.

I'm not saying it isn't a compromise to use a cheaper component, but I think it's a well thought-out compromise.
If they would have to use first grade everything, it would've probably cost 2x or 3x as much as it does now.
Besides, all the best or most expensive components in the world mixed together don't make a perfect product.
It's all in the mix

I am very sceptical that to add noise ( the TL072c have 3 times the noise of a cheap NE5532 ) to the signal just converted from the wonderful CS43122 DAC can improve in some mode the sound.

If some Rotel engineer want explain this i am here ready to learn.

I can say that the very sofisticated test measurements of Audio Review ( Italian magazine ) do speak clear :

A well sounding pre/dec but with some limitation :

Tested with digital cd source the effective resolution is 15,1 dB for the left channel and 14,9 dB. for the right ( many medium price cd player and pre/dec have 15,8-16dB ); noise floor is very clean and at -130dB. ( good result ).

But tested with digital 24 bit source at rate of 48 Khz., begin the problems : the eff. resolution do not go up to 17,5-18 dB like in good dvd-player/pre dec, but only to 15,4 bit for the left channell and 15,3 dB. for the right, and the noise floor is pratically the same of the 16 bit ( it normally should be about 10dB. lower ).

Even the A weighted S/N in the better ( multichannel input ) value is 95dB., not a very bad result but many unit have 100dB and more.

If you compare for example the same test on a Marantz DV8400 ( dvd player ), you can see state of the art performance using more cheap DAC ( CS4392 ) but with better filter/buffer/output stage and probably a better shield between the section.

Honestly with two toroidal transformer ( and a toroidal inductor ) in the power supply section, four premium dacs and the OPA2604 in the output stage there are all the presuppositions for even better performance.

Regards
Marco

[AlvaroD]

microfast,

you, for sure, seem to have some knowledge on technical specs and measurements.
But
have you actually made some experimentations and listened to theses components by yourself ?

Having made this myself, I can assure you that specs like noise floor, slew rate, THD and so on, does not give an accurate idea on how the component really sounds in real life. Audio is a bit more complicated than just the usual parameters used for specs evaluation. That's what distinguishes companies like Rotel from the grand majority of marketing specs fans.
The first one (Rotel and alikes) argues that they listen to their design first, then improve them. The second ones (I'll let you name them) just sell numbers and hope that most people will buy these numbers and don't even bother to listen to the final result

[zeppelin]

Microfast, there is a risk in doing some mod and I suggest if the warranty of your Rotel already expire than by all means go ahead with the mod.
There is a difference between high grade and high tolerance spec in electronics component. Just remember this guys, If you think that cheaper electronics component sounds better than why would you all bother to buy expensive speaker cables or interconnect and why even bother to change the higher spec, higher quality power cables, why not just use the power cables that comes along with your Rotel products.
There is a great price difference in Higher End products and Mid End like Rotel, but compare to the electronics component that they use its not that great a difference its just by a few cents or even a dollar.
So Micofast if you really want to risk your rotel by all means do it and send us your report. For me I'm to lazy to do it cause it really must involve in a lot of work and patience.