4558=4559

[QSQCaito]

Hi guys. Yesterday i went to look for an order, while i was taking a look to it, in my list the seller wrote beside where i asked for the 4558, he wrote "4558cont" and gave me a 4559. I know it's also a dual op-amp, but do you know any difference in sound/performance?

Also i asked for 1N914and he put beside that"=1N4148" are they the same, similar?


Thanks a lot from advanced.

Bye bye

DAC

[petemoore]

  The 4559 is supposedly a bit higher gain, they probably sounded about the same, or +/-  'here or 'there, depending on peripherals to the OA.
  Use whatya got, 'socket and discern' ...or not...

[QSQCaito]

Thanks a lot pete. One more question to add, what's the diff between TL072 and RC4558/9?


Any clue about the diodes?

Thanks a lot lot!!


bye bye

DAC

[calpolyengineer]

Those diodes are very similar and in most cases completely interchangeable.

-Joe

[QSQCaito]

Thanks a lot guys, you're faster everyday

Bonus question: do you know the differences in sound between 4558/9 and TL072??

thanks a lot

Bye bye

[analogmike]

TL072 is much higher specs: will be clearer, sharper sounds, more hifi in most circuits. Less noise too I think.

1N4148 is almost always a little better than 1N914 which are really wimpy in any power applications (for clipping you won't hurt a 1N914)

[QSQCaito]

Thanks a lot guys, i think this post is done, time record:P

Bye bye

DAC

[Peter Snowberg]

The 1N914 and 1N4148 both signal diodes but one is rated for twice the current of the other. I think 50mA & 100mA? For 99.99% of stompbox uses they will work exactly the same. For any power applications, use a 1N4001 instead.

The TL072 uses JFET input transistors for super high impedance. That's the main difference. Download the datasheets of each to your computer to have them near by.

The RC4558 is a general purpose opamp, while the RC4559 is a high performance amp. The 4559 is specified to have a noise level four times lower than the 4558.

http://focus.ti.com/lit/ds/symlink/tl072.pdf
http://focus.ti.com/lit/ds/symlink/rc4558.pdf
http://focus.ti.com/lit/ds/symlink/rc4559.pdf

[jota]

I am confused, because the noise is in different units:
4558: 8 nV/√Hz Typ at 1 kHz
4559: 2 μV/Hz Max (20 Hz to 20 kHz)

[PRR]

> I am confused, because the noise is in different units:
4558: 8 nV/√Hz Typ at 1 kHz
4559: 2 μV/Hz Max (20 Hz to 20 kHz)

20Hz to 20KHz is a 19,980Hz bandwidth.

Square-root of 19,980 is 141.3.

141.3 times 8 nV/√Hz is 1,130nV or 1.1uV.

On paper, the '8 appears to be slightly lower hiss.

BUT in real life you must also compute input current hiss against source impedance. The '8 does not specify input hiss current, but the '9 has significantly lower input bias current which may be a clue.

If you are truly hurting for hiss, there are other choices.

[ElectricDruid]

> I am confused, because the noise is in different units:
4558: 8 nV/√Hz Typ at 1 kHz
4559: 2 μV/Hz Max (20 Hz to 20 kHz)

20Hz to 20KHz is a 19,980Hz bandwidth.

Square-root of 19,980 is 141.3.

141.3 times 8 nV/√Hz is 1,130nV or 1.1uV.

On paper, the '8 appears to be slightly lower hiss.

Given the likelihood of "specsmanship", the statement of " 8 nV/√Hz Typ at 1 kHz" might mean *exactly* that, that the nice low noise figure is achieved at 1KHz, and elsewhere it's worse. I'm imagining a sort of U-shaped graph with the low point around 1KHz. To be fair, since that's roughly where human hearing is most sensitive, that's a pretty good way to distribute the noise. The whole story might be somewhat more nuanced, however....

T.

[Rob Strand]

I think the main problem here is comparing typical and max figures.

The Fairchild datasheets are more consistent:

  RC4558  10nV/rtHz  (graphs => typical)

  RC4559  1.4uV rms *typ.* over 20Hz to 20kHz
              which equates to 9.9nV/rtHz
             The Fairchild data is 5uV rms *max*.

The TI data is,

   RC4558 is 8nV/rtHz typ.

   RC4559 data is also 1.4uV rms typ, but only 2uV rms *max*.

So there is a small in disagreement between the RC4558 typ values and a large difference in the RC4559 max values.  (The Fairchild data appears to be at 50ohm source and the TI at 1k ohm, which would help Fairchild if anything.   I didn't labour over looking any fine print here.)

[StephenGiles]

I think the main problem here is that folks forget that at 9v it really makes no difference at all which opamps are used, especially in a band situation. However, in the quiet of your bedroom where you might be trying to impress your girlfriend, who knows!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

[aron]

Every little bit counts though. My memory sounds better to me for sure.