@Markw4,
All right. I did some basic digital circuit design and FPGA flashing back in University but in terms of DAC interpolation filter design, you lost me already...
So I guess I'll sit tight and hope you get the chance to work on that someday
In the meantime, I'll focus on getting some decent PSU according to Greg's "audio friend"'s setup.
@Greg,
Would you mind, if you have the info, specifying the exact models of toro transformers he used for:
For the output stage, I'm considering another option:
Twistead Pear's toroidal transfo 15V+15V (50VA) Power Transformer
I would have expected > 15V in order to get the final 15V out of the Placid HD BP 2.1 but if I understand correctly, it is meant to be used with it so I guess it's fine?
Any opinion how its performance would compare to your friend's Antek transfo?
Thanks!
All right. I did some basic digital circuit design and FPGA flashing back in University but in terms of DAC interpolation filter design, you lost me already...
So I guess I'll sit tight and hope you get the chance to work on that someday
In the meantime, I'll focus on getting some decent PSU according to Greg's "audio friend"'s setup.
@Greg,
Would you mind, if you have the info, specifying the exact models of toro transformers he used for:
- Katana/MC
- OPAMP Output Stage
For the output stage, I'm considering another option:
Twistead Pear's toroidal transfo 15V+15V (50VA) Power Transformer
I would have expected > 15V in order to get the final 15V out of the Placid HD BP 2.1 but if I understand correctly, it is meant to be used with it so I guess it's fine?
Any opinion how its performance would compare to your friend's Antek transfo?
Thanks!
Jonathan P:
This one is easy. For the opamp output stage and the TP Placid HDPS, TP specifies and sells a 50VA unit from Avel, model Y236203, see: http://www.avellindberg.com/pdf/avel_y23_range.pdf. The price is $27.
The Antek unit is basically spec'd the same, see:AS-0515 - 50VA 15V Transformer - AnTek Products Corp, but it also has an electrostatic shield. The price is $17.50.
"Audio Friend" called out the electrostatic shield. That was the rational. Digital on one side and the TP HDBP supply on the other supplying the analog output stage. If you need it, great. If not, well - my experience is they sound the same.
As to the cost difference, there's shipping to be considered too. You can combine shipping with TP - the Avel unit and the Placid HDBP supply. So, pretty close.
As to the Kantana DAC board and the micro controller, each using half of the DIYINHK LT3042 regulator board - the board can be configured for either 3.3 or 5 or 7 volts. Obviously, the application here is 5 volts. There's some writing on the board that states a range for the AC input as I recall. Basically, if you're too high, you generate some more heat in the regulator. If you're too low, you drop out of regulation. But, there's not much current here - something like 20mA for the micro controller when the +/-15 is supplied externally, and 165mA at startup and 80-100mA for the DAC when music is playing (those numbers from the Allo spec sheet doc). So, even if you have an extra volt - that's 0.02 and 0.165 watts respectively under worse case conditions. Power dissipation is more steady state though...
It is an LDO reg too. And, this is the low noise application... And, with a 50VA transformer the secondary voltage is going to float a little high due to the light load.
6+6 will work or 7+7 allowing for some other application for the supply in the future. Antek is you're friend again, or even Avel. There's nothing really magic here. All of these toroids are machine wound. Again, they sound the same IME.
Here's a treatment of the tradeoffs from an engineering perspective:
Dimdim's Blog | DIY Audio, Arduino, Computers, Music
There's some margin of error and some guess work involved. From the perspective you're in, at least. So, that's what to pick and a source or two.
It appears was though I may have gone to talking down to you. Please don't take it that way - just trying to help and answer your questions throughly.
This one is easy. For the opamp output stage and the TP Placid HDPS, TP specifies and sells a 50VA unit from Avel, model Y236203, see: http://www.avellindberg.com/pdf/avel_y23_range.pdf. The price is $27.
The Antek unit is basically spec'd the same, see:AS-0515 - 50VA 15V Transformer - AnTek Products Corp, but it also has an electrostatic shield. The price is $17.50.
"Audio Friend" called out the electrostatic shield. That was the rational. Digital on one side and the TP HDBP supply on the other supplying the analog output stage. If you need it, great. If not, well - my experience is they sound the same.
As to the cost difference, there's shipping to be considered too. You can combine shipping with TP - the Avel unit and the Placid HDBP supply. So, pretty close.
As to the Kantana DAC board and the micro controller, each using half of the DIYINHK LT3042 regulator board - the board can be configured for either 3.3 or 5 or 7 volts. Obviously, the application here is 5 volts. There's some writing on the board that states a range for the AC input as I recall. Basically, if you're too high, you generate some more heat in the regulator. If you're too low, you drop out of regulation. But, there's not much current here - something like 20mA for the micro controller when the +/-15 is supplied externally, and 165mA at startup and 80-100mA for the DAC when music is playing (those numbers from the Allo spec sheet doc). So, even if you have an extra volt - that's 0.02 and 0.165 watts respectively under worse case conditions. Power dissipation is more steady state though...
It is an LDO reg too. And, this is the low noise application... And, with a 50VA transformer the secondary voltage is going to float a little high due to the light load.
6+6 will work or 7+7 allowing for some other application for the supply in the future. Antek is you're friend again, or even Avel. There's nothing really magic here. All of these toroids are machine wound. Again, they sound the same IME.
Here's a treatment of the tradeoffs from an engineering perspective:
Dimdim's Blog | DIY Audio, Arduino, Computers, Music
There's some margin of error and some guess work involved. From the perspective you're in, at least. So, that's what to pick and a source or two.
It appears was though I may have gone to talking down to you. Please don't take it that way - just trying to help and answer your questions throughly.
Last edited:
@jrocker, Thanks for your answers!
Greg already gave some hints about soldering quality and all but I'm wondering what you guys use in terms of:
Greg already gave some hints about soldering quality and all but I'm wondering what you guys use in terms of:
- Solder diameter for assembling the Placid HD BP 2.1
- Solder + flux paste or flux core solder
- Wire gauge for connecting power directly to katana, MC, RPI
Hi guys,
Wondering if any of you have tested the XLR output on the Katana.
I figured it would be smart to take advantage of the balanced input on my amp if I understand correctly, the Katana XLR output is not truly balanced?
@Markw4 suggested a simple circuit to balance the output but I wonder if adding this circuit would make it worth to use over RCA?
Getting the best out of Allo.com's new Katana DAC...
New FIFO buffer for RPI/SBCs
Wondering if any of you have tested the XLR output on the Katana.
I figured it would be smart to take advantage of the balanced input on my amp if I understand correctly, the Katana XLR output is not truly balanced?
@Markw4 suggested a simple circuit to balance the output but I wonder if adding this circuit would make it worth to use over RCA?
Getting the best out of Allo.com's new Katana DAC...
New FIFO buffer for RPI/SBCs
Thanks Indiglo!
And regarding the iron temperature, does that depend more on the solder type or on the sensibility of the part to solder?
Does something change with the sound? If so, how would you describe what you hear that sounds different?
Deeper sound stage? Not sure what that means. I wrote something about listening to dacs for sound quality the other day. It is post #48 in the following thread: https://www.diyaudio.com/forums/dig...proving-ak4490-thd-figures-5.html#post5666915 You might read that and try listening to the changes once more. Depth of sound stage is a perceptual experience that can be changed by a little as .1dB to .01dB volume level change (a fact often used as a part of listening tests because people normally don't perceive such a small level change for what it actually is).
@jonners, thanks for the 2nd report on more-direct power connections to the Katana. When I first did a +-15V supply, because of my previous experiences that using the most direct connection will produce better results, I input the supply into the center pins of J28 & J29 instead of using J27 and then havine the rails go through jumper contacts. Just made sense. I'll be adding direct connect pigtails to the Katana I've been running in the next day or so... will post some pix.
@Jonathan P (& @Indiglo), On solders, my recommendation of the Wonder Solder is not due to its 'audiophile' properties, but due to it being eutectic and having a very good rosin-core flux. When it first came out in the 1980's, I had a number of pretty extensive DIY projects under my belt... a couple of pairs of own-design speakers, a couple of Hafler DH-200 kits and a DH-101 plus rebuilds of both ala the Audio Amateur Pooge articles, a Shallco-based stepped attenuator, tonearm rewirings, etc. I already knew how to make good solder joints. Still, with hobby-grade soldering irons and typical 60%-40% tin-lead solder that took care and skill. I bought some Wonder Solder to try based on the claims of better sonics, but stayed with it (and recommend it) due to its much better workability over typical hobby-grade solders. I just built a Bottlehead Eros phono stage kit over the last week or so... not a bad solder joint in the build AND it worked first power-up. I am sure there are other solders that are as good to use, but of those I've tried I haven't found one yet. For circuit-board builds I generally use the 22awg size. PartsConnexion recently redid their website, but in the previous incarnation they had some blurbs on their products. I remember some strong comments about Johnson rosin core solder as another with good workability... AND it is almost half again more expensive than the Wonder Solder.
On soldering irons, up until 5 years ago I mostly used Weller 25W soldering irons for small work and 40W ones for larger work such as speaker cables and crossovers. Then I was on a project at work where we used nice Weller digital temperature-controlled soldering stations for both mid-sized component assembly and SMD board fabrication / modification. I fell in love! While I didn't want to spend nearly $300 for that particular model, the Weller NES51 at about $100 and the NESD51 at about $115 (both available at Amazon) are good alternatives. I love my NES51. BUT for a beginner doing small circuit-board and low-gauge wiring (>12g), a Weller 25W will do well. I used them for over 3 decades with good results!
Jonathan, also, with all due respect, based on some of your questions and comments, I can tell you don't have much experience or knowledge in electronics builds. For example, your comment "I would have expected > 15V in order to get the final 15V out of the Placid HD BP 2.1" tells me you aren't familiar with how AC-DC rectification works. There are some good overview articles if you search the internet, but basically the most common AC-DC conversion circuit, the full-wave-bridge, produces a DCV output that is roughly 1.414x the ACV. So the 15VAC Twisted Pear transformer produces about 21VDC once rectified.
Some of your other questions such as types of solder and soldering irons to use also suggest low-experience and knowledge in this area.
There's nothing wrong with that, we all started there. BUT I would strongly suggest that a Twisted Pear Placid HD Bipolar is NOT an appropriate kit for someone with that level of knowledge. Either stick with pre-made modules that only require hookup of the AC or DC inputs and outputs (and maybe adjusting the output voltage). OR build up a couple of 'training' kits from some of the electronics retailers AND for a good first audio kit, look at something fairly simple with mostly capacitors, resistors, and a few monolythic chips like the Twisted Pear Centaur (for 5V supplies) and Low Current Bipolar Supply (for +-15V supplies). AND/OR find someone local who can assist and mentor you, there is NO substitute for having someone demonstrate, look over your shoulder while you try, and then critique.
I hope this all helps!
Greg in Mississippi
@Jonathan P (& @Indiglo), On solders, my recommendation of the Wonder Solder is not due to its 'audiophile' properties, but due to it being eutectic and having a very good rosin-core flux. When it first came out in the 1980's, I had a number of pretty extensive DIY projects under my belt... a couple of pairs of own-design speakers, a couple of Hafler DH-200 kits and a DH-101 plus rebuilds of both ala the Audio Amateur Pooge articles, a Shallco-based stepped attenuator, tonearm rewirings, etc. I already knew how to make good solder joints. Still, with hobby-grade soldering irons and typical 60%-40% tin-lead solder that took care and skill. I bought some Wonder Solder to try based on the claims of better sonics, but stayed with it (and recommend it) due to its much better workability over typical hobby-grade solders. I just built a Bottlehead Eros phono stage kit over the last week or so... not a bad solder joint in the build AND it worked first power-up. I am sure there are other solders that are as good to use, but of those I've tried I haven't found one yet. For circuit-board builds I generally use the 22awg size. PartsConnexion recently redid their website, but in the previous incarnation they had some blurbs on their products. I remember some strong comments about Johnson rosin core solder as another with good workability... AND it is almost half again more expensive than the Wonder Solder.
On soldering irons, up until 5 years ago I mostly used Weller 25W soldering irons for small work and 40W ones for larger work such as speaker cables and crossovers. Then I was on a project at work where we used nice Weller digital temperature-controlled soldering stations for both mid-sized component assembly and SMD board fabrication / modification. I fell in love! While I didn't want to spend nearly $300 for that particular model, the Weller NES51 at about $100 and the NESD51 at about $115 (both available at Amazon) are good alternatives. I love my NES51. BUT for a beginner doing small circuit-board and low-gauge wiring (>12g), a Weller 25W will do well. I used them for over 3 decades with good results!
Jonathan, also, with all due respect, based on some of your questions and comments, I can tell you don't have much experience or knowledge in electronics builds. For example, your comment "I would have expected > 15V in order to get the final 15V out of the Placid HD BP 2.1" tells me you aren't familiar with how AC-DC rectification works. There are some good overview articles if you search the internet, but basically the most common AC-DC conversion circuit, the full-wave-bridge, produces a DCV output that is roughly 1.414x the ACV. So the 15VAC Twisted Pear transformer produces about 21VDC once rectified.
Some of your other questions such as types of solder and soldering irons to use also suggest low-experience and knowledge in this area.
There's nothing wrong with that, we all started there. BUT I would strongly suggest that a Twisted Pear Placid HD Bipolar is NOT an appropriate kit for someone with that level of knowledge. Either stick with pre-made modules that only require hookup of the AC or DC inputs and outputs (and maybe adjusting the output voltage). OR build up a couple of 'training' kits from some of the electronics retailers AND for a good first audio kit, look at something fairly simple with mostly capacitors, resistors, and a few monolythic chips like the Twisted Pear Centaur (for 5V supplies) and Low Current Bipolar Supply (for +-15V supplies). AND/OR find someone local who can assist and mentor you, there is NO substitute for having someone demonstrate, look over your shoulder while you try, and then critique.
I hope this all helps!
Greg in Mississippi
Last edited:
Good point about using the centre pins, I'm now doing that.
Since that USB3 connector can be inserted either way it must be doing some switching of the supply which can't be ideal?
Anyway, since bypassing it I can now use 1000uF caps on both rails of my 15V PSU (instead of only one) and get it to connect.
Last edited:
First, I updated the 1st post in this thread with additional info on the power supply @jonners & @Fabio1068 are using. BTW, that supply got some good press over in CA here: A novel way to massively improve the SQ of computer audio streaming - Page 512 - Music Servers - Audiophile Style
Those posts had me soften my comments on it in the 1st post, but I still suspect use of it is for the experienced DIY'er only.
I also mentioned where I described how I separate the boards in the Katana to prevent damaging the inter-board connections.
@jrocker, I forgot to acknowledge your very informative response to @Jonathan P on transformer selection. Thanks for that!
AND thanks to both my 'audio friend' and to @jonners for reports on bypassing the Allo connectors on the Microprocessor and DAC boards. I did that today and listened a bit. What I heard was an increased sense of inner details, such as instrument and vocal textures. The buzzy-ness of the vibration of various strings, the vocal intonations that result from how we shape our mouths and hold our tongues and similar inner details were more evident and easier to hear. Also, as @jonners said, I also heard '... a slightly deeper sound stage and more ambient information.'
Not a huge improvement, but much more than a subtle one and well worthwhile!
What is interesting is that I did not perceive a reduction of background noise or blacker blacknesses. The Katana in my setup with my supplies and to my ears very good in that realm already AND MUCH improved over the inital version AND somewhat improved in its current incarnation over my earlier power supply setups. I'm guessing that the reduction of plug-in junctions AND increase in wire gauge provided a slight increase in the ability to pass dynamic current changes in response to the digital circuitry. Sonically, this sounded similar to other situations where I removed a power-supply plug-connection and replaced it with a hard-wired one and/or better wire.
ALSO, wiring this way bypasses a protection diode on the Microprocessor board... dangerous if you don't know what you are doing, but worthwhile. See @jrocker's write-up a bit back on the similar setup at the MicroUSB on many RPis.
Well worthwhile, though I rate this an 'experienced-to-expert' level modification to the Katana. It takes some care to ensure you don't short or dislodge any of the parts used for the connection points AND you have to know how to confirm you did it right before you power up!
I used my standard Deans high-current plugs for the connectors (these are from my model aviation hobby) AND Audience 18g hookup wire for the wire. On the Microprocessor board, I soldered the - to the - lead of the supercap and the + to the + side of the nearby ceramic cap. An alternative for the - lead is to solder to the metal outer case of the USB-C connector... that is what I did on the Boss 1.2.
On the DAC board, I soldered to the lead stubs from the Allo-provided power connector jack on the bottom side of the board, keeping the power connector intact.
I strain-relieved both with some hot-melt glue... you see that in the attached picture of the Microprocessor board connection, but not in the DAC board picture (I did that right after that picture). Ugly, but effective.
ALSO note that this is a board that I modified based on direction from Allo during the Katana-development listening tests I did. SO the added caps are not the same or even in the same places as what everyone else has. Better or worse? Don't know, I have tried going back and forth. BUT this is the Katana I've used for most of my listening.
AND of course, as with anything like this, your ability to hear the effect of this modification AND appreciate it will be system, person, and personal preference-dependent.
Greg in Mississippi
P.S. @jonners, I forgot to respond on the USB-C connector... it is just designed such that inserting it either way makes the right connections. No switching or anything fancy involved. IMHO, this is how the USB connector should have been designed in the 1st place!
P.P.S. I pulled out one of Ian Canada's IsolatorPi's I have here to try with the Katana. I'm curious if there might be a possible SQ lift using it because it can be easily powered separately from the Katana DAC board, which is not EASILY possible with Allo's Isolator 1.2. OTOH, I slightly preferred Allo's original Isolator to Ian's, I suspect because of the more sophisticated regulators.
Those posts had me soften my comments on it in the 1st post, but I still suspect use of it is for the experienced DIY'er only.
I also mentioned where I described how I separate the boards in the Katana to prevent damaging the inter-board connections.
@jrocker, I forgot to acknowledge your very informative response to @Jonathan P on transformer selection. Thanks for that!
AND thanks to both my 'audio friend' and to @jonners for reports on bypassing the Allo connectors on the Microprocessor and DAC boards. I did that today and listened a bit. What I heard was an increased sense of inner details, such as instrument and vocal textures. The buzzy-ness of the vibration of various strings, the vocal intonations that result from how we shape our mouths and hold our tongues and similar inner details were more evident and easier to hear. Also, as @jonners said, I also heard '... a slightly deeper sound stage and more ambient information.'
Not a huge improvement, but much more than a subtle one and well worthwhile!
What is interesting is that I did not perceive a reduction of background noise or blacker blacknesses. The Katana in my setup with my supplies and to my ears very good in that realm already AND MUCH improved over the inital version AND somewhat improved in its current incarnation over my earlier power supply setups. I'm guessing that the reduction of plug-in junctions AND increase in wire gauge provided a slight increase in the ability to pass dynamic current changes in response to the digital circuitry. Sonically, this sounded similar to other situations where I removed a power-supply plug-connection and replaced it with a hard-wired one and/or better wire.
ALSO, wiring this way bypasses a protection diode on the Microprocessor board... dangerous if you don't know what you are doing, but worthwhile. See @jrocker's write-up a bit back on the similar setup at the MicroUSB on many RPis.
Well worthwhile, though I rate this an 'experienced-to-expert' level modification to the Katana. It takes some care to ensure you don't short or dislodge any of the parts used for the connection points AND you have to know how to confirm you did it right before you power up!
I used my standard Deans high-current plugs for the connectors (these are from my model aviation hobby) AND Audience 18g hookup wire for the wire. On the Microprocessor board, I soldered the - to the - lead of the supercap and the + to the + side of the nearby ceramic cap. An alternative for the - lead is to solder to the metal outer case of the USB-C connector... that is what I did on the Boss 1.2.
On the DAC board, I soldered to the lead stubs from the Allo-provided power connector jack on the bottom side of the board, keeping the power connector intact.
I strain-relieved both with some hot-melt glue... you see that in the attached picture of the Microprocessor board connection, but not in the DAC board picture (I did that right after that picture). Ugly, but effective.
ALSO note that this is a board that I modified based on direction from Allo during the Katana-development listening tests I did. SO the added caps are not the same or even in the same places as what everyone else has. Better or worse? Don't know, I have tried going back and forth. BUT this is the Katana I've used for most of my listening.
AND of course, as with anything like this, your ability to hear the effect of this modification AND appreciate it will be system, person, and personal preference-dependent.
Greg in Mississippi
P.S. @jonners, I forgot to respond on the USB-C connector... it is just designed such that inserting it either way makes the right connections. No switching or anything fancy involved. IMHO, this is how the USB connector should have been designed in the 1st place!
P.P.S. I pulled out one of Ian Canada's IsolatorPi's I have here to try with the Katana. I'm curious if there might be a possible SQ lift using it because it can be easily powered separately from the Katana DAC board, which is not EASILY possible with Allo's Isolator 1.2. OTOH, I slightly preferred Allo's original Isolator to Ian's, I suspect because of the more sophisticated regulators.
Attachments
Last edited:
@Greg, You're (almost) right about your assumptions, so no offense taken
Actually, I acquired some decent knowledge and hands on experience in some of my electrical engineering classes BUT that was loooong time ago so forgive my memory and my basic questions polluting the thread. I thought I would be better safe than sorry and ask experienced pros like you guys.Regarding solder type, I'm asking because I have no specific experience assembling hi-fi audio components and I have been told that for that purpose, solder with small amounts of silver was recommended. It's not clear though if it affects sound quality, solder resistance/durability and/or ease to work with… I guess I just wanted to try to optimize everything as much as possible but in the end, it’s probably way more important to focus on getting good joins. Nevertheless, the presence of silver seems to be confirmed by looking at the composition of some of the best (allegedly) solders made by renowned hi-fi component manufacturers:
Cardas Solder
S-070-10 | FURUTECH
I appreciate your advice about NOT going with the Placid HD BP. Unfortunately though, it's already on its way, so I'll follow your other advice about refreshing my skills on basic circuits first. Glad to hear the Weller WES51 station is your favorite because I own the exact same model. Hopefully it will help me succeed
Also, just FYI, I received replacement stack from allo and I now have the Katana working with 3 PSU (2 iPower + 1 allo provided). It sounds just slightly better than with only 2 PSU. Maybe I should replace the iPowers with DIYINHK LT3042 regulator and/or allo's PSU with K&K 12w low voltage power supply but my understanding is that the +-15VDC should make the biggest difference, right?
Finally, I apologize for asking again but since I’m due and ready to upgrade my interconnect between DAC and AMP, it would be great to make sure I choose the right/best connector type (RCA vs XLR).
If you have an XLR input on your amp, it make sense to have XLR all way through... There is lot of info on the Internet about the PROs and CONs regarding RCA and XLR including 'shoot-outs'... It's a good moment for using your search engine of choice. Good luck!
The 'Audiophonics' supply (also available via eBay from China) also has a thread here: low noise Pre-Amp / DAC power supply MJE15034 TL072 Regulator based on STUDER 900
I tried using my IsolatorPi with Katana but I thought I heard a slight loss of resolution, so I have removed it.