Padel writes:
I have been having the same thoughts myself.
Padel writes:
Just looked up the specs on the AKM5383. It has balanced inputs.
Davey writes:
Hey Davey,
Wanna open another can of worms?
Any chance you could post the analog input circuitry leading up to the ADC's? Then all of us arm chair designers could speculate on ways to get the goesinnas to work as good as the goesoutas...
All the best and a happy new years!
Gary
I have been having the same thoughts myself.
Padel writes:
Just looked up the specs on the AKM5383. It has balanced inputs.
Davey writes:
Hey Davey,
Wanna open another can of worms?
Any chance you could post the analog input circuitry leading up to the ADC's? Then all of us arm chair designers could speculate on ways to get the goesinnas to work as good as the goesoutas... All the best and a happy new years!
Gary
Hi
As far as I know most of the users will enter their DCX through the AES digital input where there's very litlle to mod.
Oelrich has done a nice mod though :
Oelrich's mod
Now, if you are talking about analog inputs here are the schematics :
You'll notice that the crappy JRC4580 opamps can be very easily replaced with better ones.
It's what I did with OPA2604 for a very nice improvement. And I replaced capsC26, C41, C27,&C42 with straps for another improvement. Some people will prefer to enter the ADCs with input trannies which can be done very easily as the ADC have a balanced design.
Thierry
As far as I know most of the users will enter their DCX through the AES digital input where there's very litlle to mod.
Oelrich has done a nice mod though :
Oelrich's mod
Now, if you are talking about analog inputs here are the schematics :
An externally hosted image should be here but it was not working when we last tested it.
You'll notice that the crappy JRC4580 opamps can be very easily replaced with better ones.
It's what I did with OPA2604 for a very nice improvement. And I replaced capsC26, C41, C27,&C42 with straps for another improvement. Some people will prefer to enter the ADCs with input trannies which can be done very easily as the ADC have a balanced design.
Thierry
Gary P said:
Davey writes:
Hey Davey,
Wanna open another can of worms?
All the best and a happy new years!
Gary
Gary,
The complete schematic has been posted over on the DCX forum for a long time now. Look in the files section. There are three PDF's. The one labeled "REAR" is the one of primary interest for most folks.
http://groups.yahoo.com/group/DCX2496/files/
There are a number of people who have suggested/implemented excellent upgrades which show good insight. And some others that haven't.
Cheers,
Davey.
Improved Sonics
Keep in mind that improving the sonics is why we mod. I have been running my DCX with the direct out mod (36 ohms/.1uf) for almost a year with no problems and great sound. Sounds much better than stock and better than any of the 4-5 of the output schemes that I have tried. Swapping the dac chip to AK4395 and running direct out will elevate the sonics of the DCX to near state of the art. If it SOUNDS better, it is better!
Keep in mind that improving the sonics is why we mod. I have been running my DCX with the direct out mod (36 ohms/.1uf) for almost a year with no problems and great sound. Sounds much better than stock and better than any of the 4-5 of the output schemes that I have tried. Swapping the dac chip to AK4395 and running direct out will elevate the sonics of the DCX to near state of the art. If it SOUNDS better, it is better!
sendler,
Is the AK4395 a full drop in replacement or did you have to change something besides the chips also?
Did you do the simple RC out before the change of DAC IC's or after? If after then what were the most obvious changes in sound that made it better?
I have ran the DCX with simple RC also for quite long but as I need a volume after DCX and I have different amps (one 4 channel and one 2 channel) I had problems with the arrangement. Somehow the sound is not as good if a passive volume is inside the DCX and there is no active stage after it. And it is not just in highs. The mid and bass also lack freedom and livelyness.
So this is why I'm working on a version where I will use RC followed by 6 channel 10k ohm pot and followed by opamp to get some gain and to buffer the output.
It also makes it possible to have different gains in analog stage for different channels and I do not have to sacrifice so much DAC resolution in tweeter channels.
Ergo
Is the AK4395 a full drop in replacement or did you have to change something besides the chips also?
Did you do the simple RC out before the change of DAC IC's or after? If after then what were the most obvious changes in sound that made it better?
I have ran the DCX with simple RC also for quite long but as I need a volume after DCX and I have different amps (one 4 channel and one 2 channel) I had problems with the arrangement. Somehow the sound is not as good if a passive volume is inside the DCX and there is no active stage after it. And it is not just in highs. The mid and bass also lack freedom and livelyness.
So this is why I'm working on a version where I will use RC followed by 6 channel 10k ohm pot and followed by opamp to get some gain and to buffer the output.
It also makes it possible to have different gains in analog stage for different channels and I do not have to sacrifice so much DAC resolution in tweeter channels.
Ergo
I had it 400 ohm in both + and - side and 2200p cap between the two and then 6.8u metal foil cap to output from + side.
I agree that there is no comparison with the sound that the original analog board made.
Still I believe that if I can get the volume pot working between the DAC out and DCX final output and make it sound good I will like my setup even more.
Ergo
I agree that there is no comparison with the sound that the original analog board made.
Still I believe that if I can get the volume pot working between the DAC out and DCX final output and make it sound good I will like my setup even more.
Ergo
Depends on how you are going to use it.
If it's digital in / digital out, changing any opamp is pointless.
Now if you are using it analog in / analog out or digital in / analog out, there are simple improvements that you can do. Burr Brown OPA2604AU is a direct replacement for JRC4580 with better audio performance. It an SMD job though...
If it's digital in / digital out, changing any opamp is pointless.
Now if you are using it analog in / analog out or digital in / analog out, there are simple improvements that you can do. Burr Brown OPA2604AU is a direct replacement for JRC4580 with better audio performance. It an SMD job though...
Opamps compared
From my ramblings page.
http://mysite.verizon.net/vze4c5pt/id11.html
I have evaluated Burr Brown OPA134 (over rated)and OPA627 (much better but not quite as good as the LT and way over priced), Analog Devices AD811 (very good but too fussy about power supply design) and my absolute favorite, The LT1360 which sounds very close to the AD811 and has proven to be indestructable. I have compared all the previously mentioned opamps in various applications. Output driver, unity buffer, gain stage at 6-18db, I/V converter, dc servo, inverting stage, differential amp, etc. The sonics of the LT1360 are in general the best. Several steps ahead of the OPA134. Why the LT13xx family isn't more popular I have no idea. I have been raving about them for years. They drop into any application and sound better than whatever you took out and stand up to incredible miss use. The AD amps sound very good as well but are very fussy about implimentation and somewhat pricier so I have given up on them. I don't find much advantage to the even faster 1364 so I just prefer to play it safe and stick to the 1360/1361. The 134 is a good amp and will be a nice step up from the 071 but I find the 1360 sounds more open and see through.
From my ramblings page.
http://mysite.verizon.net/vze4c5pt/id11.html
I have evaluated Burr Brown OPA134 (over rated)and OPA627 (much better but not quite as good as the LT and way over priced), Analog Devices AD811 (very good but too fussy about power supply design) and my absolute favorite, The LT1360 which sounds very close to the AD811 and has proven to be indestructable. I have compared all the previously mentioned opamps in various applications. Output driver, unity buffer, gain stage at 6-18db, I/V converter, dc servo, inverting stage, differential amp, etc. The sonics of the LT1360 are in general the best. Several steps ahead of the OPA134. Why the LT13xx family isn't more popular I have no idea. I have been raving about them for years. They drop into any application and sound better than whatever you took out and stand up to incredible miss use. The AD amps sound very good as well but are very fussy about implimentation and somewhat pricier so I have given up on them. I don't find much advantage to the even faster 1364 so I just prefer to play it safe and stick to the 1360/1361. The 134 is a good amp and will be a nice step up from the 071 but I find the 1360 sounds more open and see through.
Ergo passive multi channel volume control
From my stepped attenuator page.
http://mysite.verizon.net/vze4c5pt/id2.html
There is a big difference in the sonics of passive volume controls when plugged into the source, before the interconnect, (which is the way most people try them?) compared with connecting them directly to the amps. I made up a set of attenuators with the connectors configured to plug into the DCX2496 outputs (The wrong way) in order to compare and the differences are quite obvious even though I’m running 16 gauge of Nordost Flatline for the interconnect. I can see where the drawbacks would be even more obvious when an attenuator is used before a more common audiophile interconnect of 22 gauge or less. The changes to the sound are actually quite wide spread. The bass becomes mushy and shelved down. Dynamics are lessened. The sound stage doesn’t project as tall as before and seems to have less detail. Even with the loss of detail there is an added stridency in the upper mids. Placing the attenuators after the cables, directly on the inputs of the amps improves the sound in every area, particularly in terms of a blacker background with better defined and separated instrument location. When used properly, connected to the amps with the shortest possible cable, a stepped attenuator will sound it's best, better than the best active preamps that I have ever tried.
From my stepped attenuator page.
http://mysite.verizon.net/vze4c5pt/id2.html
There is a big difference in the sonics of passive volume controls when plugged into the source, before the interconnect, (which is the way most people try them?) compared with connecting them directly to the amps. I made up a set of attenuators with the connectors configured to plug into the DCX2496 outputs (The wrong way) in order to compare and the differences are quite obvious even though I’m running 16 gauge of Nordost Flatline for the interconnect. I can see where the drawbacks would be even more obvious when an attenuator is used before a more common audiophile interconnect of 22 gauge or less. The changes to the sound are actually quite wide spread. The bass becomes mushy and shelved down. Dynamics are lessened. The sound stage doesn’t project as tall as before and seems to have less detail. Even with the loss of detail there is an added stridency in the upper mids. Placing the attenuators after the cables, directly on the inputs of the amps improves the sound in every area, particularly in terms of a blacker background with better defined and separated instrument location. When used properly, connected to the amps with the shortest possible cable, a stepped attenuator will sound it's best, better than the best active preamps that I have ever tried.
I finally got around to setting up my DCX. However, I have a problem. I get a ton of noise out of my tweeter and midrange channels, but not the bass.
I am powering all channels with chipamps (lm3886 on mid and tweet and LM4780 on the bass). The bass channel is 2 bridged chips running differentially (hence the - is not connected to a ground, but to an inverted amp.) All channels have a fixed gain of 26db. This noise isn't from the amplifier since when i connect it without the DCX, it is very quiet on all channels.
Can someone offer an expanation or solution to this?
I am powering all channels with chipamps (lm3886 on mid and tweet and LM4780 on the bass). The bass channel is 2 bridged chips running differentially (hence the - is not connected to a ground, but to an inverted amp.) All channels have a fixed gain of 26db. This noise isn't from the amplifier since when i connect it without the DCX, it is very quiet on all channels.
Can someone offer an expanation or solution to this?
akunec said:
That is the biggest problem with digital crossovers - having attenuated signal in front of the xover. You need to get attenuation after the xover - 4 or 6 channel whatever your set up is. Take out preamp and run digital signal straight into the xover. Than attenuate signals before amps. Finding or making volume control for six channels is pain - expensive one. After that is easy.
Good luck