Suggestions/help on op amp ID and substitution

[whoaru99]

I've been reading some on people changing op amps and getting different sound from their players.

I have a cheap player (Toshiba 3980) but can't seem to find much on specific mods for this player. The earlier models like 3950/3960 are pretty widely documented, but for some reason not too much on this one.

So, the question is, will someone help me identify which op amp(s) to change on this board? I'm using this with a strictly analog setup so the digital output is of no consequence/benefit.

On the attachment, I have ID'ed three primary components; the DAC, and two op amps. No desire to change the DAC, but anyone venture a guess as to which op amp(s) to replace for the analog output?

Thinking of putting a SO-8 to DIP-8 socketed adapter for trying few different op amps.

[ThomasW]

Don't see any attachment....

[whoaru99]

It's there now. The original was 125K so I had to shrink it a bit more.

[Dennis H]

My opinion, if you don't understand the reason for the mod and don't know what to do without asking, don't do the mod. Here's what happens when people try to "improve" SL's ASP with "better" opamps.

http://orion.quicksytes.com/viewtopic.php?t=107

Changing to a 90 MHz gain-bandwidth opamp like the LT1469 fore the ASP takes a bit more than what I see done here.

First of all no spec has jumped out at me as having a benefit over the OPA2134. The wide bandwith makes circuit stability problematic and must be checked with a spectrum analyzer. I would assume that you have local oscillations which definitely can affect the sound.

Now the bypass cap experimentation that I see in the photo is rather naive for high frequency oscillation suppression and inconsequential for low frequencies. There is too much lead inductance and the C-value is too high for this structure to work. Besides, the supply impedance is not necessarily a problem. What is needed here are measurements, not listening where any change is for the better.

The whole thing is a rather misguided use of the LT1469 and probably comes from someone's belief in the benefits of circuit speed and misunderstanding of bandwidth.

SL

[whoaru99]

Dennis, I understand what you are saying. However, it's a $50 player so I'm not particularly concerned if something goes wrong.

I guess I see it as something like tube rolling, except a fair bit more difficult because it's not a plug-in; yet.

[Marzen]

Not to rain on you parade - but I would ask do you have access to a scope and the schematic? The latter isn't completely neccessary, but it sure speeds up the process. Finding the data sheets for the existing components will give you an idea of what constraints you have for the replacement parts. You don't have limit yourself to using a dual channel opamp for the replacement, you can flip a 16 pin ic upside down for a dead bug mod. But, without a scope & a reasonable idea of what you're looking to improve you're pretty much shooting in the dark. Good luck with it tho'.

[whoaru99]

No schematic, but I do personally own a Fluke Scopemeter and if that's not enough I have access to other DSO's and analog scopes at work - in addtion to signal generators, etc.

However, I think much of the potential concerns regarding substitution are mostly figured out. My understanding is that most op amps of the OPA2604, OPA2132, AD8620, etc. type are pretty much drop ins.

Really my biggest question is taking a stab at which one is the analog output driver, for lack of better terms. I know it's not the PCM1742, that's the DAC. Guess I'll start with the 4558...

[Marzen]

Really my biggest question is taking a stab at which one is the analog output driver, for lack of better terms. I know it's not the PCM1742, that's the DAC. Guess I'll start with the 4558...

I'd say that's a pretty safe bet considering it's proximity to both the DAC & the analog rca outs. I'll take a look at work and see what I have there for SO-8 packages, at least then I could recommend what not to choose for hi fidelity. Burr-Brown's seem to be all the rage. You could burn a test cd of tones panned hard left or right if you need to trace backwards from the outs. Also probe around all the IC's & see what noise is on the VCC inputs, you may be able to add an extra decoupling cap there. I'm no audio tech, but I'd think you could start with looking at the basics: clean stable power & clock; low noise; over & undershoot, & ringing. I would also imagine there is a manual adjustment for the laser along with a probe point.
Ward

Added: something I found on the Zoran Vaddis 6:
"This unit uses Zoran Vaddis 6 chip, which is also used by lots of other units accessed the same way. 1. Turn on your DVD player 2. Open disc tray either using the EJECT button on the Player or the OPEN button on the remote 3. Press 2812 4. A menu should pop up, press ENTER until 9 appears 5. Your DVD player is now region-free (you can play any countries DVD's) 6. The menu also gives the option to enable or disable Macrovision and SVCD playing."

[whoaru99]

Thanks. I'll have to give that stuff with the Zoran chip a try.

I have a Pioneer DV-578 that is running some hacked firmware which enables a number of things like region free, NTSC/PAL video, blacker-than-black, audio and video equalizer/adjustment functions, output volume control (only accessible from the player remote control), and DIVX(?) playback, amongst other things. No adverse affects I can find - although I don't have any non-region 1 discs, or PAL discs, or DIVX, to be sure those things are working.

I did inadvertantly enable PAL video and could not navigate the menus to set it back to NTSC. Had to reflash the firmware which had NTSC set as default. The hacked firmware also claims something about using an improved DSP algorithm from the newer PIO model DV-588.

[Marzen]

I'll take a look at work and see what I have there for SO-8 packages, at least then I could recommend what not to choose for hi fidelity. Burr-Brown's seem to be all the rage.

Turns out we use the Burr-Brown PCM1725 coupled to a Nat'l LM833 dual OPA using their published filter with a Fc = 20Khz. It measured pretty darn flat from 20-20k. I should ask to borrow a unit for the weekend and give it a serious listen.
Ward