Oh my ...
FYI, I am currently working on the much needed component overhauling because trust me, it is needed. While I stand behind the schematic or architecture, I certainly do not stand behind the passive component selection, my comment to that is
. I am certainly not done, it will take a few days to map out the replacement components but I have a good grasp of what's what. I am not an engineer by trade, but the past days have teached me a few things which is in direct conflict with the decisions made for the UGS MUSES. While measurements was done - thinking of Ine's version - and it is good, it is certainly not the best performance we can squeeze out of this unit. Either the engineers thought "this is good enough and will conform to a low budget" or they didn't think it true. A third option is they were totally unaware about the different levels and quality figures the different classes represent in terms of signal integrity in the amplifiers operating range - temperature as well as gain area.
I am, to be honest, a bit disappointed. But no matter, things will change and improve with the newer version. Because of this, there will be a noticeable acoustical difference between the two versions. The article I am working on which I will publish in a few days should be viewed as an introduction to what to expect of the final version. I am releasing it is as a teaser and preparation, but it should not under any circumstance be used as your shopping cart. The real shopping cart will be released when the project is fully realized which is months away.
My underlying reason for the component overhaul is not only so that the performance is as close to an ideal as possible, but also because this preamplifier (or Line Stage) has the potential to be a top contender and can easily last a few decades before a replacement becomes a concern. It does not contain a D/A converter or any form of media player and is therefore completely immune against changes in the digital domain or power amplifier domain. That is enough reasons to give it the best start and potential from the get go.
To does who love jumping on the next project as soon as possible and therefore focus on a budget friendly and very average or mediocre performance, turn around now - you might not like the cost of this unit once all parts have been collected. On the other hand, if you are like me and see a huge potential in the UGS MUSES architecture, by all means stick around.
To me at least, the UGS MUSES Scion can certainly be an End Game preamplifier for I do not have an itch to look for the next new thing - that is reserved for D/A converters and power amplifiers.
Oneminde
FYI, I am currently working on the much needed component overhauling because trust me, it is needed. While I stand behind the schematic or architecture, I certainly do not stand behind the passive component selection, my comment to that is
. I am certainly not done, it will take a few days to map out the replacement components but I have a good grasp of what's what. I am not an engineer by trade, but the past days have teached me a few things which is in direct conflict with the decisions made for the UGS MUSES. While measurements was done - thinking of Ine's version - and it is good, it is certainly not the best performance we can squeeze out of this unit. Either the engineers thought "this is good enough and will conform to a low budget" or they didn't think it true. A third option is they were totally unaware about the different levels and quality figures the different classes represent in terms of signal integrity in the amplifiers operating range - temperature as well as gain area.I am, to be honest, a bit disappointed. But no matter, things will change and improve with the newer version. Because of this, there will be a noticeable acoustical difference between the two versions. The article I am working on which I will publish in a few days should be viewed as an introduction to what to expect of the final version. I am releasing it is as a teaser and preparation, but it should not under any circumstance be used as your shopping cart. The real shopping cart will be released when the project is fully realized which is months away.
My underlying reason for the component overhaul is not only so that the performance is as close to an ideal as possible, but also because this preamplifier (or Line Stage) has the potential to be a top contender and can easily last a few decades before a replacement becomes a concern. It does not contain a D/A converter or any form of media player and is therefore completely immune against changes in the digital domain or power amplifier domain. That is enough reasons to give it the best start and potential from the get go.
To does who love jumping on the next project as soon as possible and therefore focus on a budget friendly and very average or mediocre performance, turn around now - you might not like the cost of this unit once all parts have been collected. On the other hand, if you are like me and see a huge potential in the UGS MUSES architecture, by all means stick around.
To me at least, the UGS MUSES Scion can certainly be an End Game preamplifier for I do not have an itch to look for the next new thing - that is reserved for D/A converters and power amplifiers.
Oneminde
Two years later has a collection on the wall with different stereo front, UI layouts and volume wheel sizes, weight etc... but no stereo .. LMAO
Component selection for the UGS MUSES Scion Preamplifier - V1
After what felt like an eternity, the Component selection for the UGS MUSES Scion Preamplifier V1 is finally ready. It took all my muster due to the share amount of components and going over what I wanted. Now, this is V1 meaning the components are subject to change. This document will only be available for a limited period. Once V2 is released, this document will be removed (dead link). V2 will be permanently attached in the first entry of this thread as well as in its own post further down the road.
The pass band (frequency response) of the UGS MUSES did not reflect the pass band of the PASS LABS XP-30 and have therefore been adjusted to better reflect the 2hz-60kHz as in XP-30 - see in document.
The original gain stages and MUSES motherboard uses variable resistors. The plan here is to rather end up with fixed values. Variable resistors are used to match channels, MUSES IC and Impedance matching. There will be a set of variable resistor PCB's and a set of fixed resistor PCB's that are drop-in replacement with one another in the form of compatible pin connectors. Once the proper resistance value have been found, remove the variable resistor card and replace it with a fixed precision resistor card. If the amp drifts past the proper value (over time) or you need to perform a new bias or impedance matching, you simply do the process over.
This document does not contain any MCU, display or connectors information. That is reserved for V2 which should be the final edition and will be released after a period of bench top trials & test run as well as burn in and adjustments.
Before you make any comments or ask any question that is related to the information found inside the document, make sure you have read the material in its entirety, this will make life much easier for everyone.
Document link: Component selection for the UGS MUSES Scion Preamplifier V1.pdf - Google Drive
Oneminde
After what felt like an eternity, the Component selection for the UGS MUSES Scion Preamplifier V1 is finally ready. It took all my muster due to the share amount of components and going over what I wanted. Now, this is V1 meaning the components are subject to change. This document will only be available for a limited period. Once V2 is released, this document will be removed (dead link). V2 will be permanently attached in the first entry of this thread as well as in its own post further down the road.
The pass band (frequency response) of the UGS MUSES did not reflect the pass band of the PASS LABS XP-30 and have therefore been adjusted to better reflect the 2hz-60kHz as in XP-30 - see in document.
The original gain stages and MUSES motherboard uses variable resistors. The plan here is to rather end up with fixed values. Variable resistors are used to match channels, MUSES IC and Impedance matching. There will be a set of variable resistor PCB's and a set of fixed resistor PCB's that are drop-in replacement with one another in the form of compatible pin connectors. Once the proper resistance value have been found, remove the variable resistor card and replace it with a fixed precision resistor card. If the amp drifts past the proper value (over time) or you need to perform a new bias or impedance matching, you simply do the process over.
This document does not contain any MCU, display or connectors information. That is reserved for V2 which should be the final edition and will be released after a period of bench top trials & test run as well as burn in and adjustments.
Before you make any comments or ask any question that is related to the information found inside the document, make sure you have read the material in its entirety, this will make life much easier for everyone.
Document link: Component selection for the UGS MUSES Scion Preamplifier V1.pdf - Google Drive
Oneminde
I referring to the over all noise, THD, CMRR, Cross-talk and S/N ... as far as I know, does are that matter. Some of the measurements are Armands and some come from hifisentralen which I think is all Armands data.
My expression for "squeezing out as much performance as possible" is down to how well the PCB is designed and the overall component selections, or at least that is the goal. Wether or not that becomes a reality in real life tests remain to see. My spirit is certainly there when it comes to making it as good as possible.
If I have been looking at two different UGS units, then my references might be a bit off.
Got itAll measurements done on my pre. are done by Armand, here and at HiFi-sentralen. The ones here are done with the UGS 3. The ones at HiFisentralen are done with my latest UGS.
anyway, there will be a significant difference (read as upgrade) in terms of component drift (thermal stability) and component noise on the Scion edition or so I have learned through diving deep into component noise. That part has to matter. Sure, the schematic and things like feedback has to be good from the start, but once you start building the real thing, component quality becomes an issue. That's all I know. But I bet you, someone with much more experience than what I have might be of a different opinion...
PSU
Right now, I am considering a 30VA R-core transformer, individually shielded and epoxy potted for each channel and the MCU unit. That is 3 individual transformers, hooked up to "custom" versions of the Salas UltraBiB shunt regulator.
As with everything, this is subject to change if I stumble upon something better. But should be good for now
Oneminde
Right now, I am considering a 30VA R-core transformer, individually shielded and epoxy potted for each channel and the MCU unit. That is 3 individual transformers, hooked up to "custom" versions of the Salas UltraBiB shunt regulator.
As with everything, this is subject to change if I stumble upon something better. But should be good for now
Oneminde
Diminishing return is in the back of my mind, but it can't hurt to give it a fighting chance. And sure enough, the gain stage architecture matters. But lets see how the UGS V3 performs with Thin Film 10ppm/C parts and BJT's with less than 1dB in Noise FigureHi,
in my (humble) opinion the "upgrade" in this pre. would be the design of the gain-stage , not so much by the upgrading of the components . Stability reg. temp. drift: Yes, might be better .
But, keep up your good work
Right now, I am considering a 30VA R-core transformer, individually shielded and epoxy potted for each channel and the MCU unit. That is 3 individual transformers, hooked up to "custom" versions of the Salas UltraBiB shunt regulator.
As with everything, this is subject to change if I stumble upon something better. But should be good for now
Oneminde
Yes, go all the way !
Attachments
Last edited:
I know you are using the UPS, but the one in the middle, is that one for the MCU perhaps or is that one for the MM/MC RIIA ?Hi, I updated my last post with another picture
What I am proposing is using both the UPS and Salas architecture on a custom PCB, so in the schematic tie them together into a coherent unit, possibly scale the capacitor values since the UPS has 40kuF to 60kuF/rail compared to 6.8kuF/rail on the Salas. The UGS MUSES psu call for 2,7kuF or 2.2kuF... so. I have nothing against a good amount of reserve spread on several capacitors so they share the demand.