Considerations about opamp based preamp

[Tolu]

Hi community

I am thinking about a new preamp based on two stages opamp amplification with LM4562.

Why not dicrete?

Cause I've made very good experiences with a (commercial) preamp that has a similar concept.

But why building a new one?

Cause there are some issues for improvement.

1. It has input coupling caps (1µF MKT)

2. It has a motorized blue alps between the two stages

3. Power supply has simple 7815/7915 regs.

The circuit is similar to the ESP P88 but with some important differences.

1. Behind every input there is an opamp and than there is a reed relais in front of the alps pot

2. It has fewer parts.

Thoughts about improvements:

1. No caps in the signal path

How can I manage this? Opamp based DC servo. Uggh. Never build one, never calculated or simulated one. Is it recommended to use the LM4562 for DC servo? I am a simple mind in electronics but this could be the solution.

2. Replacing the blue alps against series resistance/ shunt optocoupler controlled by a DAC and µC.

OK, this will work but will it be better. I hope so!

3. PS with LM317/LM337 or Lm2937ET-15/LM2990T-15


Propable worsening: only one opamp behind the input selection relais not in front of each reed!

So, open minds are invited to discuss (and help )!

[myhrrhleine]

Originally posted by Tolu
Hi community

How can I manage this? Opamp based DC servo. Uggh. Never build one, never calculated or simulated one. Is it recommended to use the LM4562 for DC servo? I am a simple mind in electronics but this could be the solution.

i'd use OPA827

[fredex]

Hi, for a servo RIAA preamp using LME4970 (half LM4562) go to National website and look for "AN1651" (7/2007).
There are two high performance RIAA circuits there by Joe Curcio, one passive and one with active eq, both use servos, no input or output caps.

[nelsonvandal]

Yes, use a good FET input opamp, and you don't have to worry about DC offset and bias currents. Even the newer better bipolar input opamps like LM4562 has very low DC offset and input bias current. Since you intend to use them in a preamp with no or little gain, I don't think you'll need a servo. Or are your plans a RIAA amp?

I can't see the point using LM4562 for the servo. OPA2134 and the likes of it has less DC offset and are more stable.

The more opamps you use in sereies, the more degraded the sound will be. No monolithic opamp is without coloration. If you for instance use two LM4562, you'll get a very distant sound with hollow mid/center information. You can get a very neutral timbre combining different opamps. My favourite combination is AD8599 (too brittle with distant mid) and AD825 (too forward with accentuated mid). But still, every stage will be a step from close encounters with the musicians - IMO.

You'll find some very interestning preamps here at DIYaudio.com. Some ridiculously simple with a single JFET some more (or even more) complex.

[lineup]

Originally posted by fredex

Hi, for a servo RIAA preamp using LME4970 (half LM4562) go to National website and look for "AN1651" (7/2007).

There are two high performance RIAA circuits there by Joe Curcio, one passive and one with active eq, both use servos, no input or output caps.

fredex means LME49710.

my self would also recommend a look at the new LME49870 (single), LME49860 (dual)
for +-22 Volt operation = 44 Volt supply

Now, only for very high voltage gain amplification
I would consider these expensive modern Ultra Hi-fi op-amps.
In this league we have also:
LM4562, LME49710, LME49870, OPA627, OPA637, LT1028, AD797 + some.
These I could possibly think of buy,
For applications like x1000 (+60dB) RIAA or x100(+40dB) MicroPhone pre-amplifiers.

For normal lower voltage gain, filters and for buffering ( gain= 1 )
it is well enough to use OPA2134 (Audio JFET) or NE5534 (budget, low-noise, bipolar).
This is my thinking. And I think it is not far fromreality.

Lineup
Lineup Op Amps for Audio, Discrete & IC variants

================
INFO, LINKS, DATASHEETS
The LME49710 is a high speed op amp with excellent phase margin and stability.
Capacitive loads up to 100pF will cause little change in the phase characteristics of the amplifiers
and are therefore allowable.
http://www.national.com/ds/LM/LME49710.pdf
AN-1651: Keeping Up with the Expanding Demands of High-Performance Audio
http://www.national.com/an/AN/AN-1651.pdf

LME49870 - 44V High Performance, High Fidelity Audio Operational Amplifier

....

[PMA]

Hello Thomas,

how are the things? A new product?

Regards,
Pavel

[Tolu]

Hi Pavel,

no, just for fun!

Perhaps it would be enough to use just one opamp as buffer or with little gain (x2 or x4).

Yesterday I compared passive blue alps (10k), KR Audio Pre with 2 LM4562 and the Lightspeed passive pre. And the winner was: the Lightspeed!
Though I have never built something from Nelson Pass, I think I should try his Lightspeed buffer with 2SK170 and 2SJ74 and compare it with the LM4562 (or another HQ opamp). (Or should I
try your Dispre II? )

Is it really necessary to use caps with this opamp (e.g. LM4562)?

[Mooly]

Have you considered FET switching rather than relays ?
Oh yes OPA604AP for me

[nelsonvandal]

One vote for a discrete solution .

Why not the new B1 from Pass http://www.passdiy.com/articles.htm

[Bonsai]

Tolu,
I think if you are using low value volume control pots (like PMA says, lower is better - I prefer <=25k, but 10k is also ok - just don't expect a tube source to drive it too well).

If the pre-amp gain is low, you should be able to get away without any caps. Many power-amps use servo's, so most of them would easily null out a few 10's of mV offset from the pre-amp.