Looking at the supra 2.0 schematic i had an idea to instead of paralleling the opamps at the input, to parallel them at the output, in a master slave configuration, much like some headphone amps that have been seen on this forum.
Lets say 4 ne5532 per channel, 1 side of the double op used as an input then the remaining 7 paralled as a buffer, then the riaa feedback is global and treats the master+ slave as a single gain stage.
Or we can even split 1 of the 5532 into a 2 stage riaa then the remaining 6 channels into a master + slave buffer for the 2nd gain block of the initial 5532.
I know from experience the master slave config can be made to sound very good as a pre/headphone amp and because like drives like (same slew rate) the oscillation is at low risk.
Id like to know how viable this idea would be tho when its a high gain circuit of phono amp
Lets say 4 ne5532 per channel, 1 side of the double op used as an input then the remaining 7 paralled as a buffer, then the riaa feedback is global and treats the master+ slave as a single gain stage.
Or we can even split 1 of the 5532 into a 2 stage riaa then the remaining 6 channels into a master + slave buffer for the 2nd gain block of the initial 5532.
I know from experience the master slave config can be made to sound very good as a pre/headphone amp and because like drives like (same slew rate) the oscillation is at low risk.
Id like to know how viable this idea would be tho when its a high gain circuit of phono amp
Last edited by a moderator:
There is no law of physics forbidding it and it doesn't cause excessive equivalent input noise current like the Elektor scheme (noise-wise, the Supra 2.0 must be the worst moving-magnet amplifier ever designed), but what do you want to achieve with it? A RIAA amplifier usually doesn't drive a very heavy load, so why would it need seven buffers? Regarding stability, you may have to design for a high-frequency gain of two rather than one to accomodate for the extra phase shift of the buffer.
Last edited:
The area of benefit might be miniscule but it is still a superior gain stage than a single opamp of the same type because of greater driving power and the fact that the master opamp only has to drive the high impedance input of the buffer stage thereby nesting itself in a rather ideal condition. I say miniscule but used as a line stage vs a single opamp the difference is night and day. I might use 5534 for the master stage where the bigger cap across the comp pins will make it easier to stabilize, or otho keep the master 5532 in a jung type nested loop to slow it down.
Noise-wise an NE5534A in the first stage is slightly better than an NE5532 (although both will be swamped by record surface noise whenever there is a record playing).
I was gearing up to start designing PCB for Supra 2.0 when I found this thread, seems there is experience with this particular Phono preamp.
So without further ado will put my question here, is it worth it to make this preamp or not?, parts/PCB ( I already have tons of AD797, OPA627, LME49720HA, AD8610/20 and few others) are not a concern already have made RJM Emerald and Hagerman Bugle 2, almost done with engineering garage high end phono preamp.
What I am really interested is ability of the pre for music highs/mids/low and with low noise.
So without further ado will put my question here, is it worth it to make this preamp or not?, parts/PCB ( I already have tons of AD797, OPA627, LME49720HA, AD8610/20 and few others) are not a concern already have made RJM Emerald and Hagerman Bugle 2, almost done with engineering garage high end phono preamp.
What I am really interested is ability of the pre for music highs/mids/low and with low noise.
Last edited:
The moving-magnet version of the Supra 2.0 is a good example of how not to design a moving-magnet preamplifier. The designers completely overlooked the effect of input noise current and used a fortune's worth of ultra low noise op-amps to create the highest noise moving-magnet amplifier ever made.
The moving coil version is a good design, though. Input noise current doesn't matter much for moving coil, because of the much lower cartridge impedance.
Something similar to the Supra 2.0 but using FET op-amps would also work fine for moving magnet. FET op-amps hardly have any input noise current.
The moving coil version is a good design, though. Input noise current doesn't matter much for moving coil, because of the much lower cartridge impedance.
Something similar to the Supra 2.0 but using FET op-amps would also work fine for moving magnet. FET op-amps hardly have any input noise current.
Thanks for the inputs.
I am looking to build the OPAMP version for MM since I dont have any MC cartridge or interested in buying one. May be replace the LT version with OPA1641/2.
I am looking to build the OPAMP version for MM since I dont have any MC cartridge or interested in buying one. May be replace the LT version with OPA1641/2.
Last edited:
Rod Elliott has some example circuits here: RIAA Phono Preamps
The section "phono equalizers" is where the MM circuits start. He chooses the NE5534A for preference due to its low current noise for a bipolar opamp - a FET opamp will work well for this, and there are some low voltage rail-to-rail FET opamps now which allow a lower supply voltage such as 5V or 9V battery to be practical, and your OPA1641/2 suggestion is a great example.
Doug Self has a whole book on Electronics for Vinyl which I'd recommend.
The section "phono equalizers" is where the MM circuits start. He chooses the NE5534A for preference due to its low current noise for a bipolar opamp - a FET opamp will work well for this, and there are some low voltage rail-to-rail FET opamps now which allow a lower supply voltage such as 5V or 9V battery to be practical, and your OPA1641/2 suggestion is a great example.
Doug Self has a whole book on Electronics for Vinyl which I'd recommend.
Would you buy a ship from a ship designer who does not own his own ship?
(In fact I do know a ship designer, and he is not selling ships, but I don't hold it against him.)
Sure just as id take life tips from a dead guy, such as buddha 😀
Going back to the learning exercise that is my design, im currently looking into and weighting the benefits of active feedback vs passive. Looks like 2 very successful design- bugle and muffsy made use of passive filter between 2 gain stages. Im wondering if this would be possible with one gain stage then a monster of a buffer that is going to be 7 ne5532 channels in parallel.
Going back to the learning exercise that is my design, im currently looking into and weighting the benefits of active feedback vs passive. Looks like 2 very successful design- bugle and muffsy made use of passive filter between 2 gain stages. Im wondering if this would be possible with one gain stage then a monster of a buffer that is going to be 7 ne5532 channels in parallel.