Hi
I have a Delta 66 which I'd like to mod.
The most logical place to start was the output caps (22 uF polarised electros!) which I paralleled with some metallised 47nF caps, resulting in a nice balanced sound.
I figured the AD/DA is AKM 4524, which is capable of balanced output. The Delta uses NE5532 in the output stage as a summing amplifier, then somewhere in the breakout box I'm assuming the signal is convereted to pseudo-balanced and sent to the output plugs.
I would like to retain the A/D stage as it is, but change the analog output stage completely. I run my system in balanced mode, my other soundcard is the E-mu 1212M which sounds much better as it uses no output caps and runs in pure balanced mode (hotter, cleaner signal, better DAC and opamps than the Delta).
My question is - can I run the balanced outputs of the AKM into a dual opamp buffer, one for each pole, and use that as a fully balanced output?
Do I need any sort of high pass filtering on the opamps or can I run x1 non-inverting buffers? There's a small cap 100pF-470 pF in the feedback loop of the output buffers in the datasheet and on other applications (such as the one attached). Is this needed for LPF?
And is there a specific reason why the Delta first sums the differential outputs from the DAC and then splits it up again? Would not like to blow up the system!
I was planning to use the PCI slots (need to fins a PCI protoboard) to derive a +/- 12 volt supply to power the output opamps, with some good filtering down to maybe +-9 volts, and use 1/4" sockets on a breakout cable for signal output. I won't be able to route it through the breakout box as the cable has only 4 poles for the output signals - I needed 8 (4 channels x 2).
Connection would be tapped from the diff. outputs of the DAC. The resistors at the outputs (100 ohm each) will be removed so that the previous analog stage is disconnected. Internal connections will be on a flyout cable with fan connectors to pass the signal between the two boards.
I have a Delta 66 which I'd like to mod.
The most logical place to start was the output caps (22 uF polarised electros!) which I paralleled with some metallised 47nF caps, resulting in a nice balanced sound.
I figured the AD/DA is AKM 4524, which is capable of balanced output. The Delta uses NE5532 in the output stage as a summing amplifier, then somewhere in the breakout box I'm assuming the signal is convereted to pseudo-balanced and sent to the output plugs.
I would like to retain the A/D stage as it is, but change the analog output stage completely. I run my system in balanced mode, my other soundcard is the E-mu 1212M which sounds much better as it uses no output caps and runs in pure balanced mode (hotter, cleaner signal, better DAC and opamps than the Delta).
My question is - can I run the balanced outputs of the AKM into a dual opamp buffer, one for each pole, and use that as a fully balanced output?
Do I need any sort of high pass filtering on the opamps or can I run x1 non-inverting buffers? There's a small cap 100pF-470 pF in the feedback loop of the output buffers in the datasheet and on other applications (such as the one attached). Is this needed for LPF?
And is there a specific reason why the Delta first sums the differential outputs from the DAC and then splits it up again? Would not like to blow up the system!
I was planning to use the PCI slots (need to fins a PCI protoboard) to derive a +/- 12 volt supply to power the output opamps, with some good filtering down to maybe +-9 volts, and use 1/4" sockets on a breakout cable for signal output. I won't be able to route it through the breakout box as the cable has only 4 poles for the output signals - I needed 8 (4 channels x 2).
Connection would be tapped from the diff. outputs of the DAC. The resistors at the outputs (100 ohm each) will be removed so that the previous analog stage is disconnected. Internal connections will be on a flyout cable with fan connectors to pass the signal between the two boards.
Since this is a one-man thread, I decided to go for it.
All caps except three have been changed to Pana FC, and the output opamps changed over to LM4562. During the opamp changes I managed to rip a couple of solder pads right off the board.
Results: Absolutely jaw-dropping. Though I'm only using a pair of MX5021 computer speakers as I'm relocated in a much smaller apartment and my main rig is resting, the improvement in mid and high is gripping and immediately noticeable. The older 'ER' caps were sluggish and tubby by comparison, and the NE5532 on the output didn't help.
The opamp supply was a R-C filtered +/-12V from the PC.
I had also previously modded an SB Live! for practice by changing a 4558 to LM 4562 and change the in and output caps on this opamp to FCs. I had used hot air, which completely warped and twisted the board, but the card still works fine. I now stay away from hot air, a pair of cutters and a regular iron work fine for me, plus a micro soldering iron for the chips.
The LM4562 is an amazing little opamp. Highly recommended.
All caps except three have been changed to Pana FC, and the output opamps changed over to LM4562. During the opamp changes I managed to rip a couple of solder pads right off the board.
Results: Absolutely jaw-dropping. Though I'm only using a pair of MX5021 computer speakers as I'm relocated in a much smaller apartment and my main rig is resting, the improvement in mid and high is gripping and immediately noticeable. The older 'ER' caps were sluggish and tubby by comparison, and the NE5532 on the output didn't help.
The opamp supply was a R-C filtered +/-12V from the PC.
I had also previously modded an SB Live! for practice by changing a 4558 to LM 4562 and change the in and output caps on this opamp to FCs. I had used hot air, which completely warped and twisted the board, but the card still works fine. I now stay away from hot air, a pair of cutters and a regular iron work fine for me, plus a micro soldering iron for the chips.
The LM4562 is an amazing little opamp. Highly recommended.
Yeh, I didn't pursue it for a long time. When the internet buzz on the 4562 got to me I thought WTH, and went for it.
The SB live! was much, much tougher to do, the original chip was a narrow SOIC and it ran off the 5Volts line, so I had to cut a couple of traces to get at least a single 12V supply to the opamp, though the chip might've worked at 5V too.
The delta runs on +-12V default, though, so that was a little easier overall if I hadn't goofed on removing the chip.
The SB live! was much, much tougher to do, the original chip was a narrow SOIC and it ran off the 5Volts line, so I had to cut a couple of traces to get at least a single 12V supply to the opamp, though the chip might've worked at 5V too.
The delta runs on +-12V default, though, so that was a little easier overall if I hadn't goofed on removing the chip.
Here ya go :ImageShack Album - 6 images
As you can see, I botched up a few bits here and there. The card didn't seem to mind too much, still works fine
As you can see, I botched up a few bits here and there. The card didn't seem to mind too much, still works fine
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.