Hi, your design is very interesting since I am looking for a phono stage in the new preamp I am planning.
The preamp D Self published in Elektor 2012 uses the input resistance circuit discussed in this thread, I don´t remember the noise improvement he claimed but it is in the article if someone have the magazines.
About the surface mount circuits, put them on the copper side, SO14 is not that hard to solder and I kind of like surface mount now even for self etched boards, it reduces the amount of boring drilling you have to do. A combination of surface mount and through hole components can sometimes make the layout better since you can use both sides of the board without the trouble of making double sides board. If you design with 1206 resistors it is possible to mount 1206 or mini-melf resistors, in those sizes it´s easier to find real metal film SMD resistors.
I notice the OPA1644 have very low input bias current, would it not be possible to skip the input capacitor?
Also consider using two OPA1642, this will make the layout easier I think, and it also makes it possible to use a different opamp for the last stages, people love to play around with different opamps.
The preamp D Self published in Elektor 2012 uses the input resistance circuit discussed in this thread, I don´t remember the noise improvement he claimed but it is in the article if someone have the magazines.
About the surface mount circuits, put them on the copper side, SO14 is not that hard to solder and I kind of like surface mount now even for self etched boards, it reduces the amount of boring drilling you have to do. A combination of surface mount and through hole components can sometimes make the layout better since you can use both sides of the board without the trouble of making double sides board. If you design with 1206 resistors it is possible to mount 1206 or mini-melf resistors, in those sizes it´s easier to find real metal film SMD resistors.
I notice the OPA1644 have very low input bias current, would it not be possible to skip the input capacitor?
Also consider using two OPA1642, this will make the layout easier I think, and it also makes it possible to use a different opamp for the last stages, people love to play around with different opamps.
Hi Martin,
D Self's circuit is a bit of a travesty. The input noise current of the op-amps he uses creates much more noise than the resistor does. It could actually increase the noise if he uses the parallel buffer that he describes in SSAD.
I'm a little apprehensive about leaving out the input capacitor. You are right that there is virtually no bias current, but there is of course the offset of U1.1, through a high value resistor. It should be a bipolar type, really, mainly to protect C4.1 from harm.
I think you are right regarding the mounting, I'd better crack on with the layout!
I do like the idea of fewer packages, and therefore fewer supply rail decoupling capacitors and SMD solder joints. I did a fair bit of research, and I'm confident that nothing comes close to the 1644 for this circuit.
D Self's circuit is a bit of a travesty. The input noise current of the op-amps he uses creates much more noise than the resistor does. It could actually increase the noise if he uses the parallel buffer that he describes in SSAD.
I'm a little apprehensive about leaving out the input capacitor. You are right that there is virtually no bias current, but there is of course the offset of U1.1, through a high value resistor. It should be a bipolar type, really, mainly to protect C4.1 from harm.
I think you are right regarding the mounting, I'd better crack on with the layout!
I do like the idea of fewer packages, and therefore fewer supply rail decoupling capacitors and SMD solder joints. I did a fair bit of research, and I'm confident that nothing comes close to the 1644 for this circuit.
Hi Martin,
D Self's circuit is a bit of a travesty. The input noise current of the op-amps he uses creates much more noise than the resistor does. It could actually increase the noise if he uses the parallel buffer that he describes in SSAD.
I'm a little apprehensive about leaving out the input capacitor. You are right that there is virtually no bias current, but there is of course the offset of U1.1, through a high value resistor. It should be a bipolar type, really, mainly to protect C4.1 from harm.
I think you are right regarding the mounting, I'd better crack on with the layout!
I do like the idea of fewer packages, and therefore fewer supply rail decoupling capacitors and SMD solder joints. I did a fair bit of research, and I'm confident that nothing comes close to the 1644 for this circuit.
Hi Monty!
your attempt to electronic cooling is absolutely proper vs Doug's
I can find them on RS, they should become fairly commonplace due to their excellent specifications. One thing I particularly like is the low quiescent current, 1.8mA, as opposed to the conventional 8mA or so . That'll mean that a smaller power supply can be used.
Voted leave this morning . I don't want Merkel's millions moving in next door .
. That'll mean that a smaller power supply can be used.Voted leave this morning
. I don't want Merkel's millions moving in next door .
the best would be 1641 with well decouplings to each chip but probably this is no an option?
Hmm... That would mean a lot of SMD solder joints for the constructor, as well as extra board space and component cost. Seeing as the chip will be situated on the bottom of the board, the usual restrictions don't apply as much. I could make this thing really rather compact.
I think I shall go with the 1644. To avoid crosstalk issues, I shall use a single chip per channel. Seems to be the best way to do it .
I think I shall go with the 1644. To avoid crosstalk issues, I shall use a single chip per channel. Seems to be the best way to do it
.
No, only the op-amps will be SMD.
I'll leave provision on the board for good large polystyrene capacitors, to keep distortion low.
Doug Self showed that common polyester types generate about 0.03% THD about 6dB below the cut-off.
I'll try and leave as much ground around the tracks, being a single sided board. Hopefully when I finish prototyping this, the board will have a silk screen .
I'll leave provision on the board for good large polystyrene capacitors, to keep distortion low.
Doug Self showed that common polyester types generate about 0.03% THD about 6dB below the cut-off.
I'll try and leave as much ground around the tracks, being a single sided board. Hopefully when I finish prototyping this, the board will have a silk screen
.
Hi!
do you have something like BOM with the footprints or Mouser NO,
I will start to complete the parts in meantime...
PS did you read Bateman articles about capacitors distortion (exemplary attached)?
Polypropylenes are not bad too...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.