Done some searching but still confused.
I'm just doing some experimenting with some samples and wondering if it's ok to just drop in Bi or Di FETs for J-fets.
Should they work or will I damage the circuits.
Basically for a phono stage running on +and - 15v. OPA 2604 now.
New one is OPA 2107.
Maybe pre amp experiments also with 2604 but that may be hard to beat.
I'm just doing some experimenting with some samples and wondering if it's ok to just drop in Bi or Di FETs for J-fets.
Should they work or will I damage the circuits.
Basically for a phono stage running on +and - 15v. OPA 2604 now.
New one is OPA 2107.
Maybe pre amp experiments also with 2604 but that may be hard to beat.
any FET (or CMOS) input op amp should have low bias current and work with the same bias arrangement
phono preamp can have DC gain that makes input offset V annoying so check circuit and op amp spec
do check supply V limits - some newer chips have lower supply Vmax; for example the AD8610/20 is +/-13 V max
phono preamp can have DC gain that makes input offset V annoying so check circuit and op amp spec
do check supply V limits - some newer chips have lower supply Vmax; for example the AD8610/20 is +/-13 V max
Hi,
besides the afore mentioned points and of course a compatible pin-layout one of the main differences will be the noise of the circuit.
Above a source impedance of a couple of kOhms (which is the case for MM-pickups) FET-input OPamps are comparable or lower in noise than those with bipolar inputs. Still though the voltage noise figure should be around 5nV/sqr(Hz) or lower and the 1/f-frequency should be as low as possible. OPA2107 and OPA2604 feature 1/f frequnecies around 100-200Hz, while a bipolar OP27 has lower voltage noise and a lower 1/f at 2.7Hz. Other bipolar alternatives are OP27/37, NE5532/5534. So don´t expect a very good Signal-to-Noise ratio from FET-OPamps as input device in a phono stage.
The noise figures of FET-OPamps is rather at the upper limit for MM-pickups and too noisy for MC-pickups (with much lower impedance values than MM-pickups, around a couple of ten ohms).
For an MC-pickup You should rather use a very lownoise bipolar OPamp like LT1018, LT1115, AD797, INA103/163 et al.
jauu
Calvin
besides the afore mentioned points and of course a compatible pin-layout one of the main differences will be the noise of the circuit.
Above a source impedance of a couple of kOhms (which is the case for MM-pickups) FET-input OPamps are comparable or lower in noise than those with bipolar inputs. Still though the voltage noise figure should be around 5nV/sqr(Hz) or lower and the 1/f-frequency should be as low as possible. OPA2107 and OPA2604 feature 1/f frequnecies around 100-200Hz, while a bipolar OP27 has lower voltage noise and a lower 1/f at 2.7Hz. Other bipolar alternatives are OP27/37, NE5532/5534. So don´t expect a very good Signal-to-Noise ratio from FET-OPamps as input device in a phono stage.
The noise figures of FET-OPamps is rather at the upper limit for MM-pickups and too noisy for MC-pickups (with much lower impedance values than MM-pickups, around a couple of ten ohms).
For an MC-pickup You should rather use a very lownoise bipolar OPamp like LT1018, LT1115, AD797, INA103/163 et al.
jauu
Calvin
Hi,
You might consider a phono stage using two gain stages. This way You can choose a superior OPamp for each stage and split up the RIAA-equing, which easens the design of the equalizing parts, offers more flexibility and leads to better measurement results.
Don´t be shy to use Grandpas like the old NE553x or OP27/37 as first gain stages. They perform on par noisewise with e.g a more modern and much faster AD797 (which is optimized noisewise for sources around 200 Ohms. No advantage for sources with impedance values >2kOhm).
In any case I´d opt for a bipolar input OP as first stage. The second stage could be both bipolar as well as FET, though I personally prefer FETs in this position, since the equing filter that sits between both stages somewhere becomes high ohmic in value. FETs are preferrable in high ohmic value applications -> lower noise and lower offset.
Such a topology is described in the thread "Schematic for Pro-ject phono box" starting at #62, the circuit Mzperx built after Bobolix and my proposals. It offers one of the most flexible and precise phonostages and easily competes with expensive commercial products (of which it outperforms most at least measurement wise).
jauu
Calvin
You might consider a phono stage using two gain stages. This way You can choose a superior OPamp for each stage and split up the RIAA-equing, which easens the design of the equalizing parts, offers more flexibility and leads to better measurement results.
Don´t be shy to use Grandpas like the old NE553x or OP27/37 as first gain stages. They perform on par noisewise with e.g a more modern and much faster AD797 (which is optimized noisewise for sources around 200 Ohms. No advantage for sources with impedance values >2kOhm).
In any case I´d opt for a bipolar input OP as first stage. The second stage could be both bipolar as well as FET, though I personally prefer FETs in this position, since the equing filter that sits between both stages somewhere becomes high ohmic in value. FETs are preferrable in high ohmic value applications -> lower noise and lower offset.
Such a topology is described in the thread "Schematic for Pro-ject phono box" starting at #62, the circuit Mzperx built after Bobolix and my proposals. It offers one of the most flexible and precise phonostages and easily competes with expensive commercial products (of which it outperforms most at least measurement wise).
jauu
Calvin
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.