tl074 instead of 5532
I think that using tl074 instead of the 5532 could be a good thing, as long as we are using twice the opamps (instead of paralleling 32 ne5532, we parallel 32 tl074, so twice opamps).
It wold be cheaper, and for sure quieter and with a warmer sound because of the Jfet input of the tl074.
What do you think guys?
I think that using tl074 instead of the 5532 could be a good thing, as long as we are using twice the opamps (instead of paralleling 32 ne5532, we parallel 32 tl074, so twice opamps).
It wold be cheaper, and for sure quieter and with a warmer sound because of the Jfet input of the tl074.
What do you think guys?
Last edited:
multiloop with actual high current outputs in packages designed for Watts power dissipation is the cheaper and easily better performing option
you can choose any great for audio input op amp, wrap a global feedback loop around the output amp(s)
can parallel and parallel/bridge output amps too - but starting from 100s of mA to Amps depending on output chip amps
you can choose any great for audio input op amp, wrap a global feedback loop around the output amp(s)
can parallel and parallel/bridge output amps too - but starting from 100s of mA to Amps depending on output chip amps
Last edited:
TL074 does not have anything close to the drive capability in practice of the 5532 so looks to me like a bad idea. Won't be quieter (the JFET front end is much noisier) but could possibly be cheaper yes.
To get 'warmer' sound from the 5532 improve its power supply (i.e. lower the impedance seen by the opamp).
Just tested one tl074 per channel with my marshall major headphones. Sound is not warm at all, but no power issues. In fact tl074cn are pushing more power than expected...faked devices? possible as I bought from a chinese seller for 20 cents piece...
Thank you for your advice Mr Self. However in mixers, equalisers and preamps, I have heard tl074's and they do not sound as bad as we could imagine.
Furthermore when I look in the datasheets,
Input Noise Voltage for the NE5532:
noise 8nV Hz ⁄ (fo = 30Hz)
Slew Rate: 8V/μs
Short Circuit Current ISC = 38 - mA
And for the tl074:
noise Vn = 18 nV/√Hz Typ at f = 1 kHz
Slew Rate: 13 V/μs Typical
Short Circuit Current ISC = 40 - mA
So in theory we have: higher slew rate and higher short circuit current for the tl074. Also slightly higher noise figure, but since I plan to parallel twice the opamps , I can expect to improve this figure.
Still convinced that it is a bad idea?
Thank you for your advice Mr Self. However in mixers, equalisers and preamps, I have heard tl074's and they do not sound as bad as we could imagine.
Furthermore when I look in the datasheets,
Input Noise Voltage for the NE5532:
noise 8nV Hz ⁄ (fo = 30Hz)
Slew Rate: 8V/μs
Short Circuit Current ISC = 38 - mA
And for the tl074:
noise Vn = 18 nV/√Hz Typ at f = 1 kHz
Slew Rate: 13 V/μs Typical
Short Circuit Current ISC = 40 - mA
So in theory we have: higher slew rate and higher short circuit current for the tl074. Also slightly higher noise figure, but since I plan to parallel twice the opamps , I can expect to improve this figure.
Still convinced that it is a bad idea?
For sound quality, I've always championed the TL0 series and I still rate them very highly.
I think you will find that individual 'opamp for opamp' that the 5532 package will deliver more current than a TL0 device. The noise figure doesn't tell the whole story unless looked at under identical conditions between devices, and very importantly, as applied to a specific circuit.
I think you will find that individual 'opamp for opamp' that the 5532 package will deliver more current than a TL0 device. The noise figure doesn't tell the whole story unless looked at under identical conditions between devices, and very importantly, as applied to a specific circuit.
Look at fig9 on the TL074 datasheet which shows the output voltage swing vs load resistance. If you use 32 TL074s a 4ohm speaker will look like 512ohms - the maximum swing you can get into this load is about 9V for a 15V supply - in other words it won't swing closer than 6V from each rail, roughly 10W. An 8ohm speaker does better - about 4V from the rails and gives about 7.5W.
I totally agree with that, That is why I am going to replace one NE5532, with two tl074 ( 'two opamp for one opamp! instead of 'opamp for opamp'.
10w are more than enough for me
It would be advisable to check the total power dissipation in the opamps when you've got such a poor utilization of the power rails. You may find the primary constraint is thermal rather than the output current.
Hi, Absolutely, rgds, sreten.
Your trying to improve something you don't really
understand applying gross oversimplifications.
Last edited:
Eeeeeemmm, since i am replacing one NE5532 with two tl074 it is not an oversimplification. It is rather a complication, since the original is twice simpler
.Maybe you are giving an opinion to a comment you did not understood
Eeeeeemmm, since i am replacing one NE5532 with two tl074 it is not an oversimplification. It is rather a complication, since the original is twice simpler
Maybe you are giving an opinion to a comment you did not understood
No, your analysis is very oversimplistic, rgds, sreten.
A quad TL074 will not better a dual 5532, its a poor idea.
I would like to hear more from Mr Self, is it a "terrible idea " to replace 1xne5532 with 1xtl074cn because it will not work? (small power headphone version seems to work well-may be too much power is output-).
Or may be because it will sound bad?
Because too little power will be avaible to the speaker?
Knowing that 5watts RMS are more than enough for most of my needs, I am not lokking for too much power...
Please do not misunderstand me as Sreten did, BY NO WAY I am trying to improve Mr Self design or to simplificate it. In fact as power capability and output characteristics of TL's are less than the NE, my mods COMPLICATE the design, at least by using more parts.
It might be interesting to compare the sound of two different opamps.
If opamp design seems "simple", it doesn't mean that it is, in fact some comments that ARE simple are often misunderstood.
I value all your comments , and say thank you for them.
I totally agree with you. The point is to use the TL's I already have, not to overload them or to improve Mr self design , which is beyond my skills.
So far , paralleling 8 opamps (2x tl074) had given good results with my headphones. But as Abraxalito pointed, power issues may arise if I do the same without considering total Pd, since one NE5532 seems to have power dissipation capability twice the tl074.
Or may be because it will sound bad?
Because too little power will be avaible to the speaker?
Knowing that 5watts RMS are more than enough for most of my needs, I am not lokking for too much power...
Please do not misunderstand me as Sreten did, BY NO WAY I am trying to improve Mr Self design or to simplificate it. In fact as power capability and output characteristics of TL's are less than the NE, my mods COMPLICATE the design, at least by using more parts.
It might be interesting to compare the sound of two different opamps.
If opamp design seems "simple", it doesn't mean that it is, in fact some comments that ARE simple are often misunderstood.
I value all your comments , and say thank you for them.
I totally agree with you. The point is to use the TL's I already have, not to overload them or to improve Mr self design , which is beyond my skills.
So far , paralleling 8 opamps (2x tl074) had given good results with my headphones. But as Abraxalito pointed, power issues may arise if I do the same without considering total Pd, since one NE5532 seems to have power dissipation capability twice the tl074.
Hi,
It may be more complicated, but it will be no better.
In terms of the published design what is the point of
the extra complication and board redesign ? None.
Can you build a amp based on TL074's ? Of course.
Is it a better idea than 5532's ? No. Especially given
the attention to the finer details in D.Self's design.
You started out by trying to imply the change would
be worthwhile, when basically it isn't in any respect.
rgds, sreten.
It may be more complicated, but it will be no better.
In terms of the published design what is the point of
the extra complication and board redesign ? None.
Can you build a amp based on TL074's ? Of course.
Is it a better idea than 5532's ? No. Especially given
the attention to the finer details in D.Self's design.
You started out by trying to imply the change would
be worthwhile, when basically it isn't in any respect.
rgds, sreten.
Last edited: