Yes, PSU rejection detorientates at higher frequencies in OPamps and shunts can have a very low impedance to higher frequencies. The LME series of Opamps on the other hand has very good PSU rejection up to high frequencies. Look at the new LME49990.
The problem with very low impedance shunts can be oscilation on the other hand.
I did not say that i did not hear a difference. I just have a problem when listeners claim gigantic and miraculous gains. At the end of the day it´s a bit like cooking. Everything counts and what we hear is a combination of first order and second order effects so FdW´s solution may sound excellent despite some choices he made that whould be done in a differnt way by others.
When you look into integrated intrumentation OPs also consider the INA163.
The problem with very low impedance shunts can be oscilation on the other hand.
I did not say that i did not hear a difference. I just have a problem when listeners claim gigantic and miraculous gains. At the end of the day it´s a bit like cooking. Everything counts and what we hear is a combination of first order and second order effects so FdW´s solution may sound excellent despite some choices he made that whould be done in a differnt way by others.
When you look into integrated intrumentation OPs also consider the INA163.
I compared the Placid ( discrete Shunt) with the LC Audio Low Noise PSU ( discrete series) i anhanced with 27mH coils plus lokal decoupling. I have some experience with LM317 type of regulators that where not totally satisfying. I can not remember that I used fixed regulators since the LM317 came out and that is a long time ago.
I also have some good experience with he TL line splitter but with a lot of decoupling caps. I own a Horwege DAC and Horwege Phono that both use this and sound very dynamic. I think the webpage is Willkommen bei der HiFiWERKSTATT hoer-wege
I also have some good experience with he TL line splitter but with a lot of decoupling caps. I own a Horwege DAC and Horwege Phono that both use this and sound very dynamic. I think the webpage is Willkommen bei der HiFiWERKSTATT hoer-wege
@Joachim
I compared Placid vs Salas V1.0 with remote sensing +5.5V 500mA in the DAC Buffalo II of Twisted Pear, please be so kind to let me know your opinion.
The Salas reg. was do it with 2.200uF/63V cap PSU filter, the rest of components: Elna Silmic II 100uF/25V, Xicon & Koa Speer both carbon film resistors & 1 trimmer Bourns, I detail the components used because as you said before all count.
For both reg. I used a R-Core Tx 9V 230V 30VA.
I compared Placid vs Salas V1.0 with remote sensing +5.5V 500mA in the DAC Buffalo II of Twisted Pear, please be so kind to let me know your opinion.
The Salas reg. was do it with 2.200uF/63V cap PSU filter, the rest of components: Elna Silmic II 100uF/25V, Xicon & Koa Speer both carbon film resistors & 1 trimmer Bourns, I detail the components used because as you said before all count.
For both reg. I used a R-Core Tx 9V 230V 30VA.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Now I have to try it
I will replace the series regulators with shunt regulators. The current boards do not allow for a discrete build, and thus I am getting a few drop-in replacements. To fit them in their current spot TO220 models are needed (with fixing hole) and I found NewClassD (NewClassD UWB Regulator) with the types that I need (but only up to 15V). After exchanging some mail with the very helpful people there they offered to supply me with 24V versions (custom build). And that is what I am going to do keep the fingers crossed.The quick simulation that I did last night convinced me the shunt regulators could be an improvement. I have the schema of that simulation attached. The schema is not optimized in any way, but it was the most simple implementation of a series and a shunt regulator, both with about the same type and amount of components.
Regards,
Frans de Wit.
You are really dedicated to make somthing good. It will be very interesting to know if the application of a Shunt brings you any benefit. I have seen the design you will use and i am curios how it will work.
I aso did a little circuit that uses your MC input scheme but has lower noise. You could also try the LME49990 that has even less distortion and better PSU rejection.
In the case of the LME49990 i whould use 10 Ohm and 300 Ohm resistors because it can only drive 20mA in comparison to 40mA of the AD chip.
I aso did a little circuit that uses your MC input scheme but has lower noise. You could also try the LME49990 that has even less distortion and better PSU rejection.
In the case of the LME49990 i whould use 10 Ohm and 300 Ohm resistors because it can only drive 20mA in comparison to 40mA of the AD chip.
I will install the LME49990 in the diff input stage.
I am not convinced about the lower values resistors. Using lower values will reduce noise, but, it will also make for a mismatched impedance to the cartridge. The MC preamp is used with a Koetsu Rosewood Signature it is specified as needing a 100ohms load impedance and will then deliver 600uV.
The question remaining, and that I would like to see answered, is this; is using lower value resistors a tradeoff between noise and impedance matching? and to what effect?
Regards,
Frans.
I think the only way is to listen. I found that also the type of resistor is important.
The best sound i got so far is with DaleRN65E. I the case of a conventional input i landed between 150 and 300 Ohm in parallel to the cartridge with Lyra Titan i.
If i understand your circuit right you have not the same impedance in both legs. Taking my variety of your circuit the positive leg whould see 5 ohm and 150 Ohm in series when i ignore the ( high ) input impedance of the positive input of your opamp. The negative leg goes to a 5 Ohm resistor and then to virtual ground so your Koetu whould be loaded with 160 Ohm because both inputs are in seies. Do i see that correctly ?
The best sound i got so far is with DaleRN65E. I the case of a conventional input i landed between 150 and 300 Ohm in parallel to the cartridge with Lyra Titan i.
If i understand your circuit right you have not the same impedance in both legs. Taking my variety of your circuit the positive leg whould see 5 ohm and 150 Ohm in series when i ignore the ( high ) input impedance of the positive input of your opamp. The negative leg goes to a 5 Ohm resistor and then to virtual ground so your Koetu whould be loaded with 160 Ohm because both inputs are in seies. Do i see that correctly ?
The 5ohm resistor is the internal series resistance of the cartridge, and not part of the circuit (sorry I forgot to take it out before publishing).
In my opinion the input impedance is 102.2ohms (51.1+51.1). As the voltage difference between +in and -in is virtually zero, it may be seen as a shortcut, and thus the feedback resistor and ground resistor are no part of the input impedance equation. Mmm… (I think).
Regards,
Frans.
And I checked (due to the Mmm...
) and found:Operational amplifier applications - Wikipedia, the free encyclopedia
Near the header: Differential amplifier
Regards,
Frans.
Or am I way out of line here, and did I read your post wrongly?
Is the best digital sound ever heard till today, now I'm upgrading with film cap in VRef by Salas advice, Salas v1.0 reg with Rs feeding the Buffalo II DAC + passive resistor + coupling cap as suggested me Erno Borbely for I/V (only penalty low gain)while I wait for the modification in his I/V that I own, so the final setup will be Buffalo II DAC + Erno Borbely I/V.
what IV model of Borbley you use here ?
can you show the complete functional connections of Salas shunts ( how many - how do you connect to Buffalo ) etc
Is there a photo?
thanks as I am interested to build one
kannan
Any ideas on this one:
The question remaining, and that I would like to see answered, is this; is using lower value resistors a tradeoff between noise and impedance matching? And to what effect? The bottom line is; what am I paying for the lowered noise floor?
Regards,
Frans.
Actually I'm not using the Borbely I/V till Erno modifie the filter for Buffalo II.
The photo with passive I/V (resistor+coupling cap) few post ago.
The Buffalo II only needs a single +5.5V rail supply so only you need to do a Salas reg. p2p like mine photo or buy a Hipnotyze PCB in the GB.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.