4558's hiss a bit. 5532 and TL072 soldier on, but 5532 use is muddied by the fact that it originally came in a 14 pin package. That is what my suppliers were selling when I tried to buy some. 8 pin DIP package is pretty standard, as is 8 pin SIP. TL072 has lockup if the input nears the power rail, and my first experiment did that immediately, so I don't mess with them. TL082 is not supposed to lock up, I have some but haven't installed them yet. I've used MC33078 in an upgraded disco mixer (RA88a), which has quite a bit less hiss than the OEM 4558. JRC4560 has more drive current than 4558, Peavey uses a lot of them. Quiet enough, I can't hear the difference versus 33078. A lot of people swear by the OPA2134, they are a bit expensive. The latest and greatest is the ti/national LM4562 which is quiet and has a lot of drive current. These are different than the Japan radio JRC4562. More drive current allows you to use bigger capacitors in the feedback loop, which cuts the value of the resistors for a given time constant, which should give less noise.
Anything faster (slew rate) than 4558 TL072 or NE5532 should have bypass disk caps near the socket on power supplies and a 22 pf cap or so around the feedback resistor, to prevent oscillation. My disco mixer oscillated about 1 mhz as soon as I replaced the 4558's with 33078s. Didn't sound much different but made the fan run hard on the power amp.
Anything faster (slew rate) than 4558 TL072 or NE5532 should have bypass disk caps near the socket on power supplies and a 22 pf cap or so around the feedback resistor, to prevent oscillation. My disco mixer oscillated about 1 mhz as soon as I replaced the 4558's with 33078s. Didn't sound much different but made the fan run hard on the power amp.
Thanks
Yea I hear a lot of good things about the OPA2132 as well, thats what I was going to use but the price is a bit steep. I will look into the 4562 see what that looks like, thanks again
Yea I hear a lot of good things about the OPA2132 as well, thats what I was going to use but the price is a bit steep. I will look into the 4562 see what that looks like, thanks again
you really need to deal with surface mount parts for today's best chips
some interesting "audio" parts of the last decade or so:
AD8610/20, ADA4627, OPA827, OPA164x
~1 nV/rtHz fast bjt inputs are available - but the noise performance is only useful at very low source impedance - SiGe helps a little
LME series (now at TI since they bought out National)
ADA4897 is interesting for the poorly advertized "highly linear" input stage - should be great for DAC I/V
quite a few high speed, high output current op amps were developed for xDSL line driving - and could make good audio line driving buffers or headphone amps
the TPA6120 is a renumbered THS6012 with extensive audio specs in the datasheet for just that purpose
chip cost really isn't often a problem - power supplies, cases, connectors add up to several times even pricey op amps for most projects - but using cheaper parts for initial debugging, then replacing with the "better" chips could be a good idea if you often release the magic smoke
some interesting "audio" parts of the last decade or so:
AD8610/20, ADA4627, OPA827, OPA164x
~1 nV/rtHz fast bjt inputs are available - but the noise performance is only useful at very low source impedance - SiGe helps a little
LME series (now at TI since they bought out National)
ADA4897 is interesting for the poorly advertized "highly linear" input stage - should be great for DAC I/V
quite a few high speed, high output current op amps were developed for xDSL line driving - and could make good audio line driving buffers or headphone amps
the TPA6120 is a renumbered THS6012 with extensive audio specs in the datasheet for just that purpose
chip cost really isn't often a problem - power supplies, cases, connectors add up to several times even pricey op amps for most projects - but using cheaper parts for initial debugging, then replacing with the "better" chips could be a good idea if you often release the magic smoke
Last edited:
Hi,
Worth a read for more modern alternatives for DIY :
NwAvGuy: Op Amp Measurements
The TL072 gets a bashing nowadays, the 4558 I've never taken seriously.
The NE3352/4 is still the audio op-amp benchmark for everything else.
rgds, sreten.
Worth a read for more modern alternatives for DIY :
NwAvGuy: Op Amp Measurements
The TL072 gets a bashing nowadays, the 4558 I've never taken seriously.
The NE3352/4 is still the audio op-amp benchmark for everything else.
rgds, sreten.
Last edited:
That was a very interesting article!!!
I have been weighing out which opamps would be the best and most cost effective for my latest project of a DIY 31 band eq using the Rane schematic.
Since it needs not be the Rolls Royce of Eq's it does need to have good performance while being cost effective, Else I would be using the LME49860 at $3.10 each(Mouser) or such equivalent.
According to the article I gather that these three opamps in the order of performance would meet my criteria,
1 NJM2068 at aprox. $.56 and lowest noise of the Three.
2 NJM4562 at aprox, $1.19 low noise and slightly better performance than the NE5532.
3 NE5532 at aprox. $1.10 all time standard.
Or should I stick with the MC33078 at $.966 that the schematic calls for?
It seems that the NJM2068 is the one to go with.
I am familiar with the NE5532 as it is in my Phillips PSC506 soundcard and I have no issues with it and sounds great even though it is a 16Bit card and it is the best sounding 16bit card I have ever heard to date.
As well as have used the NE5532 before when I replaced all of the opamp's in my old Fostex 454 mixer.
And the same goes for the NJM2068 as it is my Creative X-FI Elite Pro sound card system and the sound is excellent as well.
I am not worried so much about the character of the sound due to the opamp's except that is will be processing square waves and harmonically rich waveform's, As it is to be used between a bunch of guitar effects pedal circuits and the amplifier stack.
But the lowest noise is of concern, As I am also considering using the same opamp's for the high gain effects stages as well.
This is aside from my all time favorite general purpose favorite the TL07x's of yesteryear that I have used in such designs with good results in the past.
Even though I have found the NE5532's for about half as much, I used Mouser's prices for comparison because I may be getting the slide pots (about 50 or so) from them as well.
Some more opinions on this matter would be good as my project experience in the past has only been with the NE5532's and the TL07x opamps.
jer
I have been weighing out which opamps would be the best and most cost effective for my latest project of a DIY 31 band eq using the Rane schematic.
Since it needs not be the Rolls Royce of Eq's it does need to have good performance while being cost effective, Else I would be using the LME49860 at $3.10 each(Mouser) or such equivalent.
According to the article I gather that these three opamps in the order of performance would meet my criteria,
1 NJM2068 at aprox. $.56 and lowest noise of the Three.
2 NJM4562 at aprox, $1.19 low noise and slightly better performance than the NE5532.
3 NE5532 at aprox. $1.10 all time standard.
Or should I stick with the MC33078 at $.966 that the schematic calls for?
It seems that the NJM2068 is the one to go with.
I am familiar with the NE5532 as it is in my Phillips PSC506 soundcard and I have no issues with it and sounds great even though it is a 16Bit card and it is the best sounding 16bit card I have ever heard to date.
As well as have used the NE5532 before when I replaced all of the opamp's in my old Fostex 454 mixer.
And the same goes for the NJM2068 as it is my Creative X-FI Elite Pro sound card system and the sound is excellent as well.
I am not worried so much about the character of the sound due to the opamp's except that is will be processing square waves and harmonically rich waveform's, As it is to be used between a bunch of guitar effects pedal circuits and the amplifier stack.
But the lowest noise is of concern, As I am also considering using the same opamp's for the high gain effects stages as well.
This is aside from my all time favorite general purpose favorite the TL07x's of yesteryear that I have used in such designs with good results in the past.
Even though I have found the NE5532's for about half as much, I used Mouser's prices for comparison because I may be getting the slide pots (about 50 or so) from them as well.
Some more opinions on this matter would be good as my project experience in the past has only been with the NE5532's and the TL07x opamps.
jer
Last edited:
Totally agree. Very neutral and linear. A miracle. The one below I use to replace all the old Opamps.
An externally hosted image should be here but it was not working when we last tested it.
...
For "ultra-high permance", some people use the LME49860, pretty expensive, though.
Yes ,I have some LME49860's and I really like them a lot but they would be quite costly for this project and a bit over the top performance for this particular project as well as the end product is not mean't to be super clean or clear for the most part.
The noise of the amps alone will destroy what ever benefit I would gain from using the LME49860's.
I will definitely use them when I build a set of Eq's for my studio though.
Even though in the article it was stated that the NJM2068 has a 3db lower noise figure the NE5532 has a 50% faster slew rate at 9v/us compared to 6v/uc for the NJM2068.
I can get the NE5532's for about the same price if not a bit cheaper as I have found them for as low as $.35 before.
The faster slew rate makes me lean more towards the NE5532's.
Even though the lowest noise is what I am after I don't thank that -106db is going to make any noticeable difference compared to -109db when you consider the noise that will be generated by the previous high gain stages.
I will be benefiting right off of the bat of at least a 10db lower noise floor by using either one over the use of the TL07x's.
jer
The noise of the amps alone will destroy what ever benefit I would gain from using the LME49860's.
I will definitely use them when I build a set of Eq's for my studio though.
Even though in the article it was stated that the NJM2068 has a 3db lower noise figure the NE5532 has a 50% faster slew rate at 9v/us compared to 6v/uc for the NJM2068.
I can get the NE5532's for about the same price if not a bit cheaper as I have found them for as low as $.35 before.
The faster slew rate makes me lean more towards the NE5532's.
Even though the lowest noise is what I am after I don't thank that -106db is going to make any noticeable difference compared to -109db when you consider the noise that will be generated by the previous high gain stages.
I will be benefiting right off of the bat of at least a 10db lower noise floor by using either one over the use of the TL07x's.
jer
Last edited:
Yes, I have had this issue with a Current Boosted power amp application that I recently designed.
I first used a TL072 and when I swapped it out with the LME49860 I found some Oscillations on the positive peaks of the test sine waveform on the output into my 4 ohm load.
A simple disc capacitor solved that issue, I forget how I used it as it was a one night experiment and since then I have cleared off my protoboard for other circuit designs.
Other than that it worked flawlessly!!
Since I am doing a ground up build I am ready for such issues to correct as I go along.
I will be implementing locations in the PCB design should I need to add them as they are already shown in the schematic.
The Rane schematic isn't any much different than any other design except that it uses the MC33078's in the filter stages and it is the first time I had ran across that particular device and I am not familiar with it.
So, I googled it vs NE5532 and that is how I ran across this thread and a few other excellent ones on this subject.
The Rane EQ is very expensive if it is bought outright and the only thing I found different about it from other designs is the use of the MC33078's.
I used to have a Peavey 231 EQ and it was rated as one of the best and lowest noise Eq's at half the cost of the Rane as I paid about $450 for it back when it first came out in 1994.
But, I don't have the schematic for that one, nor have I tried to look it up yet.
I know it maybe more cost effective to just get one already made but this is a custom design for a friend and I am still weighing out the cost differences as I may not need all of the frequency bands offered as it is for a guitar rig.
In the long run my payment would be to have enough parts leftover to build a couple more to use in my recording setup.
I hope that this isn't being too off topic as I am just giving a little insight as to my requirements as others may find helpful as well to the subject of this thread.
If I were building a small preamp for my stereo system or just swapping out CD output stages this wouldn't be much of an issue.
I am talking about employing as much as 50 opamp stages in this design from Eq's to Effects and simple mixing circuits.
jer
I first used a TL072 and when I swapped it out with the LME49860 I found some Oscillations on the positive peaks of the test sine waveform on the output into my 4 ohm load.
A simple disc capacitor solved that issue, I forget how I used it as it was a one night experiment and since then I have cleared off my protoboard for other circuit designs.
Other than that it worked flawlessly!!
Since I am doing a ground up build I am ready for such issues to correct as I go along.
I will be implementing locations in the PCB design should I need to add them as they are already shown in the schematic.
The Rane schematic isn't any much different than any other design except that it uses the MC33078's in the filter stages and it is the first time I had ran across that particular device and I am not familiar with it.
So, I googled it vs NE5532 and that is how I ran across this thread and a few other excellent ones on this subject.
The Rane EQ is very expensive if it is bought outright and the only thing I found different about it from other designs is the use of the MC33078's.
I used to have a Peavey 231 EQ and it was rated as one of the best and lowest noise Eq's at half the cost of the Rane as I paid about $450 for it back when it first came out in 1994.
But, I don't have the schematic for that one, nor have I tried to look it up yet.
I know it maybe more cost effective to just get one already made but this is a custom design for a friend and I am still weighing out the cost differences as I may not need all of the frequency bands offered as it is for a guitar rig.
In the long run my payment would be to have enough parts leftover to build a couple more to use in my recording setup.
I hope that this isn't being too off topic as I am just giving a little insight as to my requirements as others may find helpful as well to the subject of this thread.
If I were building a small preamp for my stereo system or just swapping out CD output stages this wouldn't be much of an issue.
I am talking about employing as much as 50 opamp stages in this design from Eq's to Effects and simple mixing circuits.
jer
Yes, I have had this issue with a Current Boosted power amp application that I recently designed.
I first used a TL072 and when I swapped it out with the LME49860 I found some Oscillations on the positive peaks of the test sine waveform on the output into my 4 ohm load.
A simple disc capacitor solved that issue, I forget how I used it as it was a one night experiment and since then I have cleared off my protoboard for other circuit designs.
Other than that it worked flawlessly!!
Since I am doing a ground up build I am ready for such issues to correct as I go along.
I will be implementing locations in the PCB design should I need to add them as they are already shown in the schematic.
The Rane schematic isn't any much different than any other design except that it uses the MC33078's in the filter stages and it is the first time I had ran across that particular device and I am not familiar with it.
So, I googled it vs NE5532 and that is how I ran across this thread and a few other excellent ones on this subject.
The Rane EQ is very expensive if it is bought outright and the only thing I found different about it from other designs is the use of the MC33078's.
I used to have a Peavey 231 EQ and it was rated as one of the best and lowest noise Eq's at half the cost of the Rane as I paid about $450 for it back when it first came out in 1994.
But, I don't have the schematic for that one, nor have I tried to look it up yet.
I know it maybe more cost effective to just get one already made but this is a custom design for a friend and I am still weighing out the cost differences as I may not need all of the frequency bands offered as it is for a guitar rig.
In the long run my payment would be to have enough parts leftover to build a couple more to use in my recording setup.
I hope that this isn't being too off topic as I am just giving a little insight as to my requirements as others may find helpful as well to the subject of this thread.
If I were building a small preamp for my stereo system or just swapping out CD output stages this wouldn't be much of an issue.
I am talking about employing as much as 50 opamp stages in this design from Eq's to Effects and simple mixing circuits.
jer
I first used a TL072 and when I swapped it out with the LME49860 I found some Oscillations on the positive peaks of the test sine waveform on the output into my 4 ohm load.
A simple disc capacitor solved that issue, I forget how I used it as it was a one night experiment and since then I have cleared off my protoboard for other circuit designs.
Other than that it worked flawlessly!!
Since I am doing a ground up build I am ready for such issues to correct as I go along.
I will be implementing locations in the PCB design should I need to add them as they are already shown in the schematic.
The Rane schematic isn't any much different than any other design except that it uses the MC33078's in the filter stages and it is the first time I had ran across that particular device and I am not familiar with it.
So, I googled it vs NE5532 and that is how I ran across this thread and a few other excellent ones on this subject.
The Rane EQ is very expensive if it is bought outright and the only thing I found different about it from other designs is the use of the MC33078's.
I used to have a Peavey 231 EQ and it was rated as one of the best and lowest noise Eq's at half the cost of the Rane as I paid about $450 for it back when it first came out in 1994.
But, I don't have the schematic for that one, nor have I tried to look it up yet.
I know it maybe more cost effective to just get one already made but this is a custom design for a friend and I am still weighing out the cost differences as I may not need all of the frequency bands offered as it is for a guitar rig.
In the long run my payment would be to have enough parts leftover to build a couple more to use in my recording setup.
I hope that this isn't being too off topic as I am just giving a little insight as to my requirements as others may find helpful as well to the subject of this thread.
If I were building a small preamp for my stereo system or just swapping out CD output stages this wouldn't be much of an issue.
I am talking about employing as much as 50 opamp stages in this design from Eq's to Effects and simple mixing circuits.
jer
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.