Op amp swapping

[fastvideo]

Originally Posted by liquias

Resurrecting, since I promised to post feedback from my tests. (And since I have some new questions...)

OK, so I bought LM4562 op amps and as Mooly recommended I also bought the OPA2604 op amps.

I finally had the time to conduct one of the tests.

I have only tested the Sherwood CD1000C player so far. It's easier since it has only one (dual) op amp at it's output.

So I switched between the jrc4560 op amp and the opa2604 amp. sound took a couple of steps for the better. everything is more emphasized and colorful. all the instruments got presence, and musicality. Voice became a couple of meters closer. I'm very happy !

Then I tried the LM4562. This is a strange one. At first what I could hear was a little less emphasis of everything (like the sound is a little less colorful) but a very large difference in detail. everything got "crisp" and clear. I liked this sound for the few first seconds.

But then I noticed I pickup a distinct distortion. It sounds like the twitter in my speaker has gone bad. (this gave me quite a scare BTW...) I tried decoupling the op amp with 0.1uF cap (between pins 4 and 8) and this doesn't change anything. sound is still distorted, I have no idea why, or how to solve this. Changing back to the opa2604 solves this problem of course.

Any ideas why the 4562 gives me this kind of trouble ? Am I doing something wrong ?

Distortion on one or both channel?
Bad socket connection or Bad 4562.

[liquias]

thank you guys!

@fastvideo , I think not, since reseseating opa2604 on the same socket solves this problem. and I bought 10 4562 op amps, so unless the whole batch is bad i cannot see this as an explanation. thank though.

OH, and BTW:
distortion is on both channels.

[SoNic_real_one]

The decoupling of an OpAmp that is fed differential (and the load connected to ground) HAS to be done between each power pin and ground. NOT between the pins (like the urban legends are going arround).

Your LM4562 is probably oscillating - it's a faster OpAmp so it needs good supply.

[fastvideo]

Originally Posted by liquias

thank you guys!

@fastvideo , I think not, since reseseating opa2604 on the same socket solves this problem. and I bought 10 4562 op amps, so unless the whole batch is bad i cannot see this as an explanation. thank though.

OH, and BTW:
distortion is on both channels.

Can you post the Sherwood CD1000C player
schematic in here.

[liquias]

@SoNic_real_one, I'm not sure what you mean, my op amp has one positive pin (8) and one negative pin (4). pin 8 (+) is already coupled to ground using a 47uF/10V cap. if I understood correctly, you suggest that I connect pin 4 (-) to ground also using a capacitor in series ?

@fastvideo, yes of course. It's attached now.

@Xoc1, I will try replacing the cap to a slightly larger value.

@Mooly, I can easily try that, and I will.
However I'm not sure what to look for with a scope, I can throw in a wave pattern CD and play square or sine wave and see what happens on the outputs, but I'm not sure what to look for. Also, since now I attached the schematic for the player, maybe you can have a look, maybe c114L/R or c110L/R load the output with capacitance ? Should I remove them ? And one last question, won't adding resistance in series (47 ohm) to the output effect the output level considerably?


Thanks guys

[Mooly]

First thing is to connect scope across the audio output sockets. If it's unstable you will know... the waveform will have HF noise etc all over it. Try with a music CD and sine square test disc too. A 1khz squarewave from CD always looks like in the picture here.

You are using sockets... have you tried carefully soldering a small cap directly across pins 4 and 8 on the IC itself ?

The scope check is the first thing though...
C114 is totally isolated by the resistor feeding it so problem. C110 is part of the LPF network and again in itself shouldn't be an issue.
If you have loads of 4562's try bending the output pins so they don't go in the socket and add a 47 ohm so it's in series with the output as shown.

[Xoc1]

Looking at the sherwood schematic it looks like a very cheaply constructed player.
I would not really recommend you to spend to many resources on trying to improve it.
That is not to say that it can not be improved, and it may be a good way to practice your skills before tackling the NAD.
The NAD player looks a lot more sensible and would probably give a lot better results especially if upgraded.
I found a schematic for the NAD here
Service manual, eprom result list | ElektroTanya | Service manuals and repair tips for electronics experts

The Sherwood player has one single pair of + / - 5 volt regulators that are supplying the entire circuit. The resevior capacitance could be increased and the output of the regulators could have a capacitor added.
The output from the op amps has filtering muting and DC blocking. I would check that there is no DC on the output of the op amp, and and replace all the output circuitry from the input of C107 L & R with a single 100 ohm resistor to the output sockets. There are 2 x 47uF capacitors for the op-amp C106 L & R Given the low voltage You could probably go up to 220uF for these.
There are more caps in the signal path, and you might struggle to find replacements that will fit. I have had good results with Panasonic FC caps
The 82 uF 35V cap is useful as it has a 2.5mm pitch and a diameter of 6.3mm so it fits where other upgrade caps will not. If your serious about upgrading this player you could use thes caps to replace C116 C118 C119 C120 on the D/A chip and C108 L & R ,C111 L & R in the signal path.

[fastvideo]

Possible to check the power supply to opamp.
It only use resistor to zener diode. May be 4562 doesn't like it.

[SoNic_real_one]

Originally Posted by liquias

@SoNic_real_one, I'm not sure what you mean, my op amp has one positive pin (8) and one negative pin (4). pin 8 (+) is already coupled to ground using a 47uF/10V cap. if I understood correctly, you suggest that I connect pin 4 (-) to ground also using a capacitor in series ?

Put a 0.1 uF parallel with the 47uF ones - one for pin 4 and other for pin 8. You are solving nothing by connecting 4 and 8 with a capacitor...

[liquias]

Sorry for the delayed reply guys, I've had no time for fun these lasy couple of days.

here goes:

@Mooly:
I started with the 47 ohm in series (this was easiest for me). no change whatsoever. so I left the resistor in, and I connected the 'scope to the outputs of the op amps (using the 4562 of course), played music and seen nothing special. Then I played the sine waves. distortion is very visible (pic1 attached). This picture is at 10kHz, it gets worse as freq rises. Again, if I replace the op amp to jrc4560 (original) or opa2604 everything is fine(pic2). I started thinking maybe I have a whole batch of 10 defective lm4562. So I decided to rig up a small test circuit (pic3) on a matrix to see what happens when I feed the op amp using my oscillator. well results are quite bad (pic4). I tried giving the circuit larger values of +/- Vdc. It helps a little and the distortion moves a little towards the lower end of the sine wave. at this time R2=2k R1=1k, total Av of 3, Vin~1V @20kHz lowering the freq. improves the distortion but doesn't solve it, lowering Vin to about 150mV, solves it.
Today I talked to a friend of mine that suggested to raise R2 to 10k and R1 to 5k. I have to say, this really seems to solve my problems. I can inject Vin, as high as 4V and still get no distortion!
I can't really say what is going on, or why, and I am still not sure if the batch is defective or not. Should I find higher values for the resistors in my CD player, or Is it a defective batch? any suggestions?

@Xoc1:
I know the Sherwood is far from brilliance. I guess in a way I am trying to mod it first, before I tackle the NAD (which I really love!), but as I said in some other thread in this forum, I seem to have some fetish for Sherwood devices.
I am learning allot by modding the Sherwood, and I am not spending too many resources on it doing so. just as you suggested .also there is the technical issue mentioned above, which is this CD has only one op amp and this makes things very easy for me at the moment.

I'll be sure to look into the modifications you suggest for the Sherwood, as soon as I finish dealing with this strange lm4562 op amp situation.

About the service manual for the NAD, Thanks, I have it already. I am using it to plan my future modifications to the machine.

@fastvideo,
I connected my bench power supply in parallel to the players power supply and I didn't get any better results. As I stated above, you might have been right the first time, this might be a complete defective batch of IC's. we'll see...

@SoNic_real_one,
I haven't had the time to try your suggestion, I'll probably try it tomorrow, but I am also very skeptic, since as I mentioned above I connected my bench power supply in parallel with the players P.S. and got no improvement. I will try though.

I know this is getting very long.
Thank you guys, for all the time and patience, and for helping me out.
yours,
liquias.