DIY phono preamp - cheap and simple!

[peranders]

rjm, it seems that you have forgotten one important part!

The whole noise generating source consists of L+R+C. You have forgotten the C of the cable and the input matching. 100-400 pF must be added to your circuit. This may have some influence of the noise.

[carlosfm]

rmj, you could do a much better circuit layout.
That's no way of using an op-amp.
No PSU caps near the chip?
I wouldn't be surprized if the NE5532 sounds better on this circuit that the OPA2134...

Anyway, for a phono pre with a double op-amp I would go for the OPA2228 instead of the OPA2134.

[peranders]

Except for forgetting(?) decoupling caps of 100 nF (common practice) I think his effort is alright if the goal was to make it simple.

Please check datasheets of the IC's you have used and you'll see how you should power up the IC's.

[carlosfm]

Per-Anders, I'm just trying to be constructive and help rjm make a better circuit layout.
Spare two cheap caps doesn't make sense to me, even on a simple design.
And, I've said it several times here, you can ruin the sound of an OPA (or any modern/fast op-amp) just by using it the wrong way.
Believe me.
So... why bother to use an OPA2134 on this circuit?

Gotta tell you a secret: the OPA2228 behaves much better without bypassing.
If you try it on this circuit as it is, it will SOUND much better than the OPA2134.

[rjm]

P-A, is the cable capacitance in series to the cartridge impedance, R+L+C? I would have thought the capacitance is between the input and ground. In any event even 400pF isn't going to make much of a contribution at these frequencies, and pales in comparison to cartridge inductance.

As for the bypass question:

In my view this is not a high frequency circuit. Small signal bandwidth is about a megahertz depending on the gain setting and op-amp used. So "close" to the opamp may be taken in cm rather than mm.

The low impedance electrolytics, which are only 3 cm of straight wire from the op-amp's power pins, should provide good bypassing over the bandwidth of the circuit.

But, just because I don't always trust my own analysis last week I did put 0.1 uF ceramics in the "proper" place next to the power pins of the op-amp.

[I should add that most people seem happy if the cap itself is next to the power pins and don't worry particularely how far away it is to the load return. Be that as it may ... ]

I could, just about, hear a difference, though it was at the limits of my perception. The treble took on a slightly damped, closed in quality. With the caps gone the amp sounded more open and natural, but perhaps noisier.

For now the caps are gone, I think I prefer it this way but I'll keep an open mind while I continue the evaluation of the circuit in general and have a chance to check out how an OPA2134 sounds in there.

-rjm

[carlosfm]

rjm, it's more important to have capacitance near the chips than just a 0.1uf cap.
I'd put two small 22~100uf caps between the chip's pins and ground.
If there's no space to bypass with a small one, then don't bother.
The OPA2132, that I know very well (similar to the 2134) has a detached bass, uncontrolled, flappy, dominant, if used on a circuit like you have now.
Treat it as it likes and it will give you a tight, dynamic, detailed presentation, very impressive.
High-end, really.
I talk by experience with these chips.
I mean building and LISTENING experience.
Not just electronics by the book.

My passion is Audio.


BTW to all, take or leave my advice, it's up to you.

[rjm]

To change the subject a little, what do you think of this then?

VSPS Ultra

It's a little extreme, in the early stages, and may yet be consigned to the dustbin - but feel free to chew into it anyway. I'm not too familiar with moving coils, so expect bugs.

-rjm

[carlosfm]

The value of the caps on the feedback loop is too high...
And I don't understand why you put only 220uf before the regulators.
And why 220uf after the regulators.
Better would be to bypass the regulators' output with a small cap and put the caps where they are needed: AT THE OP-AMP'S SUPPLY PINS.
Does it make this simple design more expensive?
Don't think so.
Just better.

[rjm]

I re-wired the VSPS filter caps as shown on the pic below.

The output from the regulator is wired directly to the IC power pin, and this 220uF cap is no in effect wired directly to the IC power pins too.

Granted the leads are still a bit longer that one might like, but any bypass cap would have to be connected to the load return point (the red oblong on the ground next to R7) so it isn't that bad a compromise.

So we now have the IC "directly" bypassed by a large value cap, as Carlos wanted. Simple experiment:

If I'm right then this connection change will make absolutely no difference in the sound. The circuit is equivalent but for a few cm of wire.

If I hear any difference then I'll have to start paying attention to his posts.

Guess what: pretty big difference. Similar but more significant than when I tried bypassing with the 0.1uF ceramics: less treble energy, less noise. The balance of the music is shifted quite dramatically downwards from the treble to the midrange. This time though the naturllness of tone is retained.

Go figure. I'm going to look at the layout again to see what can be done to tighten up the distances.

-rjm

[carlosfm]

rjm, next time pay more attention to my words.
I'm still not crazy and I know what I'm talking about.


I mean something like this, and this is still "cheap and simple".