I tried the new Lm4562 Opamp in a Tjoeb '99 cd-player as a replacement for the standard BB OPA2604. It works but not as good as it should, i guess. The sound is rather thin and hard and there is some kind of distortion in the higher frequencies even after more than 70 hours of burning in.
I think its not suited as a direct swap in this configuration, although it sounds very promissing in some ways. There is already much more detail and separation in the soundstage than before.
Does anyone understand what goes wrong and what can be done to make it work properly?
I think its not suited as a direct swap in this configuration, although it sounds very promissing in some ways. There is already much more detail and separation in the soundstage than before.
Does anyone understand what goes wrong and what can be done to make it work properly?
Thanks for your respond!
Yes, i broke it in for about 70 hours or so.
I haven't put any decoupling on the opamp. If this is necessary, could you please do some suggestions how to do this? Do you think this could solve the problem?
I don't know what the operatingvoltage is but it is on the print of a Tjoeb '99, based on a Marantz CD38.
Yes, i broke it in for about 70 hours or so.
I haven't put any decoupling on the opamp. If this is necessary, could you please do some suggestions how to do this? Do you think this could solve the problem?
I don't know what the operatingvoltage is but it is on the print of a Tjoeb '99, based on a Marantz CD38.
Hi,
What is the output of the opamp connected to, is it connected directly to a capacitor ? If so, try connecting a small resistor in series with the output of the opamp - it won't like driving more than a few 10's of pF directly.
For decoupling start with 100nF soldered directly between the + and - supply pins on the opamp. There should also be 10-47uF between the supply pins and gnd within a few cms of the opamp.
Hope that helps.
Cheers,
Jon
What is the output of the opamp connected to, is it connected directly to a capacitor ? If so, try connecting a small resistor in series with the output of the opamp - it won't like driving more than a few 10's of pF directly.
For decoupling start with 100nF soldered directly between the + and - supply pins on the opamp. There should also be 10-47uF between the supply pins and gnd within a few cms of the opamp.
Hope that helps.
Cheers,
Jon
Yes, 100nF between the + and - supply pins directly on the opamp. Then one 10-47uF between + and gnd and another between - and gnd (mind the polarity!) close by the opamp.
I've looked at other Marantz players produced around the same time and you may find the output of the opamp connected directly to a coupling cap to block DC at the output, probably bipolar (unless there is a deliberate DC bias) and between say 10 and 100uF in value.
With inadequate supply bypassing you may find that the opamp is oscillating at high frequency and this is the cause of your problems. Adding the 100nF may just do the trick.
I've looked at other Marantz players produced around the same time and you may find the output of the opamp connected directly to a coupling cap to block DC at the output, probably bipolar (unless there is a deliberate DC bias) and between say 10 and 100uF in value.
With inadequate supply bypassing you may find that the opamp is oscillating at high frequency and this is the cause of your problems. Adding the 100nF may just do the trick.
Sorry for this delayed reaction. Finally i got my hands on a working soldertool.
I haven't soldered a ceramic capacitor (0,001uF) right across the power supply leads but soldered it underneath an 8 pin dip socket in wich i put the opamp itself. Could this have a bad influence on the functioning of the the capacitor?
Now it works and sounds(!) better than before because the sound is more natural and relaxed but there still is a bit of hardness to the sound, especially in high frequencies. Because of my limited soldering skills i don't like to solder on the print itself so the suggestions of putting caps from the chip to ground are not an option immediatly.
What else can be done? If i use a 0,01uF cap instead of the 0,001uF, could it make a difference?
I haven't soldered a ceramic capacitor (0,001uF) right across the power supply leads but soldered it underneath an 8 pin dip socket in wich i put the opamp itself. Could this have a bad influence on the functioning of the the capacitor?
Now it works and sounds(!) better than before because the sound is more natural and relaxed but there still is a bit of hardness to the sound, especially in high frequencies. Because of my limited soldering skills i don't like to solder on the print itself so the suggestions of putting caps from the chip to ground are not an option immediatly.
What else can be done? If i use a 0,01uF cap instead of the 0,001uF, could it make a difference?
Bassivus said:
Ceramic caps on the power supply pins are the preferred caps to use. They will snuff out any RF oscillation as was previous stated. Also use some nice Panasonic FC caps. low ESR, as bypasses for PS pins.
Ceramic caps as SIGNAL caps are to be avoided at all costs. This is probably what you might have heard about ceramics.
Regards, Marcus
To decouple a supply you need caps with'*some* esr. The idea of 'damping' is to take energy from the system to prevent it from ringing or oscillating. You can't do that with pure C (or L, for that matter).
Using audiophile (read: almost zero esr) caps for supply decoupling may even make the problem worse. Those ceramics with their losses are excellent for the purpose. If you read data sheets for low-drop-out regs (where you need the same caps to prevent oscillation) they even specify a MINIMUM esr, certainly not the lowest possible.
Jan Didden
Using audiophile (read: almost zero esr) caps for supply decoupling may even make the problem worse. Those ceramics with their losses are excellent for the purpose. If you read data sheets for low-drop-out regs (where you need the same caps to prevent oscillation) they even specify a MINIMUM esr, certainly not the lowest possible.
Jan Didden
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.