Nobody questions that Katana is the best HAT.
It's been Allo who announced the "blow-away-$1000-class" target though. And you didn't mean HATs only.
The latest Beekhuizen Katana 1.2 review states, that his Katana setup
"comes close" to his Chord Mojo setup - a $480 DAC.
That doesn't look good to me on the first glance, it'd be well below
the "blow-away-$1000-class" target.
Ok. I don't really trust his reviews. But that's a different story.
But maybe - he's right or at least has a point?
The problem. It's still pretty much the only review out there.
(As a matter of fact - Allo even refers to his reviews.
) Anyhow.
I do hope that Allo gets the quality issues under control soon.
And I do hope that the new PS puts the Katana at least slightly above that Mojo.
I also hope with all your promising products in pipe we can start over at square 1 in a couple of weeks.
We shouldn't forget though. Just with the new PS it's then gonna be a close to $500 DIY streaming solution (without isolator).
And this amount will make live not that much easier.
Enjoy.
PS: Advise: Allo -- remove that "blow-away-$1000-dacs" statement. This just "blows" right back into your faces.
Finally we're getting there.
A 2-in-one solution. Nice.
Will it be SMPS or Linear?
However!
3+1 A ???
What's the "1" for?
Too low for a RPI 3B+ (according to spec) and too low for Katana !??!
It would then require two of these for a Katana stack.
Or you power the DAC/isolator and the controller/output stage each with one of these two outputs
and use another SMPS for the PI e.g. your $11 SMPS.
Is that the way you're thinking?
Looking forward to it.
We are talking worst cases scenario but we will test it with Katana. In fact power available and design was made for 3.9A and 1.6A but we under quote at this point.
This is an LPS. Problem on SMPS is that every time we lower differential noise , common mode noise increases . Same for leakage current (grounding increases diff noise). In any cases new PCB might solve those problems. We will start testing soon..
Hello,
I am assembling the power supply for my Katana, but I am having an issue with the +-15V.
At first, I used only two supply: 5V for RPI + 5V for MC
Katana works as expected, I hear the music.
Then I tried to add the external +-15V with the following steps:
- turned off the Sw3 switch
- moved the jumpers J28 and J29 to close 1&2 pins
- connected power wires, +15V to pin1(close to sparkos label) GND to pin2 and -15V to pin3
When I try to play music, sometimes there is no sound at all, sometimes I get like a whistle, a medium/high frequency sound. I do not know exactly how to describe it, also because I turn the power off quickly in order to avoid damaging the boards.
Then, when I go back to the standard 2 power supply, everything works normally.
Did anyone have the same issue? Or am I doing something wrong?
Any advice/help?
PS the stack is composed of MC+THDopamp+Katana+Isolator+Raspberry Pi 3B. The OS is Volumio 2.513
I am assembling the power supply for my Katana, but I am having an issue with the +-15V.
At first, I used only two supply: 5V for RPI + 5V for MC
Katana works as expected, I hear the music.
Then I tried to add the external +-15V with the following steps:
- turned off the Sw3 switch
- moved the jumpers J28 and J29 to close 1&2 pins
- connected power wires, +15V to pin1(close to sparkos label) GND to pin2 and -15V to pin3
When I try to play music, sometimes there is no sound at all, sometimes I get like a whistle, a medium/high frequency sound. I do not know exactly how to describe it, also because I turn the power off quickly in order to avoid damaging the boards.
Then, when I go back to the standard 2 power supply, everything works normally.
Did anyone have the same issue? Or am I doing something wrong?
Any advice/help?
PS the stack is composed of MC+THDopamp+Katana+Isolator+Raspberry Pi 3B. The OS is Volumio 2.513
@fabio1068
Are the external +/- 15V supplies two separate individual supplies or is it a single unit with a +/- 15 V rail(s) and a common ground?
If using separate 15V supplies then the 0V of the positive supply and the positive of the negative supply need to be tied together on the GND pin of J27 Pin2.
The power up sequence for this mode( using an external +/- 15V supply) is the external +/-15V goes first, the MicroController next and once the green lights stabilize the Rpi is powered on.
In the absence of pictures of your arrangement this is what comes to mind based on my experience with Katana. Good luck.
Are the external +/- 15V supplies two separate individual supplies or is it a single unit with a +/- 15 V rail(s) and a common ground?
If using separate 15V supplies then the 0V of the positive supply and the positive of the negative supply need to be tied together on the GND pin of J27 Pin2.
The power up sequence for this mode( using an external +/- 15V supply) is the external +/-15V goes first, the MicroController next and once the green lights stabilize the Rpi is powered on.
In the absence of pictures of your arrangement this is what comes to mind based on my experience with Katana. Good luck.
I am going finally use the Katana in my main system. I have approximately 143 hours of burn in time using the Katana connected to powered speakers. I have it shut down, that is fifth time i have down that to allow the caps to charge up and down.
I stil only have two power supplies connected at this time. I am hoping that next week, I can connect a +/-15 volt LPS to the SQ board.
I do not like the ideas of using the usb connections so hopefully I can connect a LPS to the RPI below that board and on the MC board, as well.
I stil only have two power supplies connected at this time. I am hoping that next week, I can connect a +/-15 volt LPS to the SQ board.
I do not like the ideas of using the usb connections so hopefully I can connect a LPS to the RPI below that board and on the MC board, as well.
What power supply are you using? Can you attach a picture?
Can you measure the +-15v power supply outputs right after you turn on power? You could try measuring DC volts first, then if you have time try measuring AC volts to see if any ripple or other AC there.
Do you have an oscilloscope to look at the power rails?
@ktham and @Markw4
I think I found out what is happening, even though I do not know why.
I measured with a multimeter (sorry no oscilloscope) DC and AC voltage relative to GND at J27 pins and at the wires of the big black capacitor on MC.
Measure with only the katana stack powered on, raspberry not powered.
RCA cables connected to speaker ampli, to hear any sound/whistle.
For the first 8 minutes from power on, I get:
J27: DC -15.00 V
DC +15.06 V
AC 40mV on both rails
Cap: DC -14.06 V
DC -0.8 V
The MC board is not powering the positive rail of OPAMP board. That is why I cannot hear anything.
Then, after more or less 8 minute (much less if the stack is already warm), I get a sudden whistle from the speakers that last about 10 seconds.
When the whistle stops, I measure:
J27: DC -14.98 V
DC +15.00 V
AC 40 mV
Cap: DC -14.06
DC +14.14
AC 40 mV
the OPAMP board is now powered correctly, and if I turn the raspberry on, I can play music normally.
The voltage regulators used for both the 5V on MC and +-15V are:
Module d'alimentation lineaire regulee OP TL072 MJE15034 28V - Audiophonics
All regulators are powered by a single transformer with multiple secondaries. For that reason the 5V and +-15V are powered at the same time.
Is just the power sequence creating this malfunctioning? It looks weird to me.
Maybe somebody else is able to replicate this issue? Or is my MC board that is faulty?
Here some pictures of my setup
I think I found out what is happening, even though I do not know why.
I measured with a multimeter (sorry no oscilloscope
) DC and AC voltage relative to GND at J27 pins and at the wires of the big black capacitor on MC.Measure with only the katana stack powered on, raspberry not powered.
RCA cables connected to speaker ampli, to hear any sound/whistle.
For the first 8 minutes from power on, I get:
J27: DC -15.00 V
DC +15.06 V
AC 40mV on both rails
Cap: DC -14.06 V
DC -0.8 V
The MC board is not powering the positive rail of OPAMP board. That is why I cannot hear anything.
Then, after more or less 8 minute (much less if the stack is already warm), I get a sudden whistle from the speakers that last about 10 seconds.
When the whistle stops, I measure:
J27: DC -14.98 V
DC +15.00 V
AC 40 mV
Cap: DC -14.06
DC +14.14
AC 40 mV
the OPAMP board is now powered correctly, and if I turn the raspberry on, I can play music normally.
The voltage regulators used for both the 5V on MC and +-15V are:
Module d'alimentation lineaire regulee OP TL072 MJE15034 28V - Audiophonics
All regulators are powered by a single transformer with multiple secondaries. For that reason the 5V and +-15V are powered at the same time.
Is just the power sequence creating this malfunctioning? It looks weird to me.
Maybe somebody else is able to replicate this issue? Or is my MC board that is faulty?
Here some pictures of my setup
Attachments
@fabio1068
I am using a similar regulator board with my Katana, but I haven't enabled the separate 15V supplies yet.
You may already be aware of this, but these boards have a current limiting resistor. Mine (from eBay) had 2.2R fitted, which limited the current to about 300mA. That's fine for the 15V supplies, but for a 5V supply that requires more current a lower value resistor is needed. Are you certain that your 5V supplies are not dropping below 5V?
I got some help with this from this thread: low noise Pre-Amp / DAC power supply MJE15034 TL072 Regulator based on STUDER 900
Edit: From the Audiophonics pics, it looks as if those resistors might be 0.22R. If that's the case, they won't limit until 3A!
I am using a similar regulator board with my Katana, but I haven't enabled the separate 15V supplies yet.
You may already be aware of this, but these boards have a current limiting resistor. Mine (from eBay) had 2.2R fitted, which limited the current to about 300mA. That's fine for the 15V supplies, but for a 5V supply that requires more current a lower value resistor is needed. Are you certain that your 5V supplies are not dropping below 5V?
I got some help with this from this thread: low noise Pre-Amp / DAC power supply MJE15034 TL072 Regulator based on STUDER 900
Edit: From the Audiophonics pics, it looks as if those resistors might be 0.22R. If that's the case, they won't limit until 3A!
Last edited:
@jonners
I read that thread. My board has no problem to power the katana with only one 5V supply, so it can deliver more than 1A easily.
@Markw4
It is not a problem of overheating,It is quite the contrary:
When the system is warm, I can play music for several hours without problem.
The issue is there only at startup expecially cold start
I read that thread. My board has no problem to power the katana with only one 5V supply, so it can deliver more than 1A easily.
@Markw4
It is not a problem of overheating,It is quite the contrary:
When the system is warm, I can play music for several hours without problem.
The issue is there only at startup expecially cold start
What are the secondary voltages of the r-core transformer you are using with these regulator boards?
Okay, let's try this again, if you don't mind. On the underside of the MCU board, there are a couple of LDO regulators. They appear to be fed from J28, J29, then the LDO outputs go to the pin header where the big black cap is soldered on. From the symptoms you describe, it sounds like there is a delay in one of the LDO outputs coming up to voltage. Since you say it takes about 8 minutes for it to start working, that might suggest there is a thermal intermittent inside one of the LDOs, in the MCU board traces, or an intermittent solder joint affecting one of the connections between the late starting LDO and the associated input or output pins where you can measure. Seems like a plausible theory might be that when the 1st LDO starts working, it starts heating the MCU board in the vicinity of the LDOs. After enough time has passed the heat causes the thermal intermittent connection to close and other LDO output comes up as you can measure at the big black cap.
No matter what else, if I have correctly described the situation then it would appear most likely there is some warranty problem with a thermal intermittent of some kind on the MCU board.
Hi guys,
Can someone confirm or precise the power requirements for each of the Katana 1.2 boards?
I am particularly interested in power requirements for the +-15VDC input of the analog output stage:
Thanks!
Can someone confirm or precise the power requirements for each of the Katana 1.2 boards?
I am particularly interested in power requirements for the +-15VDC input of the analog output stage:
- min/recommended current
- min/max acceptable voltage range
Thanks!
Up to now I've been successfully using three 5V supplies to power Katana, Controller and RPi, with all supplies starting up together. I have now connected separate +- 15V opamp supplies but if I power everything together I now get noises from the speakers but no music.
I have reverted to just three 5V supplies and all is well. Should I use a particular power-up sequence when using external 15V supplies?
I have reverted to just three 5V supplies and all is well. Should I use a particular power-up sequence when using external 15V supplies?
Did you measure that +-15 volts or a little less (relative to ground) are on the two leads of the big black filter cap glued to the top of the MCU board? If no voltage on one or the other lead, then that would be a problem.
Double check the inverter on the MCU board is switched off?
If everything looks okay otherwise, I think there are power sequencing instructions in one of the downloadable manuals and or instruction sheets for Katana on the Allo website. There used to be at least.
By the way, may I ask what +-15v supply you are using?
@BruceJ
the secondaries are rated at 9V and 16V, in reality slightly higher.
@Markw4
I checked, and I think the LDOs are fine, since they are also used by the internal DC-DC converter when the stack is powered with only one 5V supply, and in that configuration everything works normally.
I also checked the voltage of the DC-DC at pins J28 and J29 (+14.8V -15.10V) and they are almost the same as my external supply.
I do not see what could be the difference between internal or external +-15V supply.
I only have a doubt. The GND of the power supply is floating, it is not connected to the Earth. If I measure the voltage between GND and Earth I get tens of AC volt. Could it create some kind of issue? do you usually ground the GNDs?
the secondaries are rated at 9V and 16V, in reality slightly higher.
@Markw4
I checked, and I think the LDOs are fine, since they are also used by the internal DC-DC converter when the stack is powered with only one 5V supply, and in that configuration everything works normally.
I also checked the voltage of the DC-DC at pins J28 and J29 (+14.8V -15.10V) and they are almost the same as my external supply.
I do not see what could be the difference between internal or external +-15V supply.
I only have a doubt. The GND of the power supply is floating, it is not connected to the Earth. If I measure the voltage between GND and Earth I get tens of AC volt. Could it create some kind of issue? do you usually ground the GNDs?
@jonners,
The power up sequencing of the +-15V supply does not matter a whit. I have left mine on while rebooting the Katana with no issues... and I have rebooted the Katana and forgotten to plug in the +-15V supply with no issue... in that case you get a very low-volume distorted sound with the feed-through paths of the feedback loops. I mute my amp, plug the +-15V supply in, unmute the amp, and sound is back. In the initial Katana Manual, it said that you could turn the outputs on and off using SW3 on page 4, top of the page (which I've also done)... not sure if that is in the latest version of the manual.
What @Markw4 said are all good points. ALSO confirm you have fed the +, -, and ground into the right pins on J27 (the positive is on the same side as J28, the negative on the same side a J29, with the ground in the center... you can confirm this by checking for continuity between the pins of J27 and the 1 pins on J28 & J29) AND that you have the jumper on pins 1-2 on J28 & J29.
Let us know what you find!
Greg in Mississippi
The power up sequencing of the +-15V supply does not matter a whit. I have left mine on while rebooting the Katana with no issues... and I have rebooted the Katana and forgotten to plug in the +-15V supply with no issue... in that case you get a very low-volume distorted sound with the feed-through paths of the feedback loops. I mute my amp, plug the +-15V supply in, unmute the amp, and sound is back. In the initial Katana Manual, it said that you could turn the outputs on and off using SW3 on page 4, top of the page (which I've also done)... not sure if that is in the latest version of the manual.
What @Markw4 said are all good points. ALSO confirm you have fed the +, -, and ground into the right pins on J27 (the positive is on the same side as J28, the negative on the same side a J29, with the ground in the center... you can confirm this by checking for continuity between the pins of J27 and the 1 pins on J28 & J29) AND that you have the jumper on pins 1-2 on J28 & J29.
Let us know what you find!
Greg in Mississippi
Last edited:
@Jonathan P... EDIT... oops, I didn't see you had quoted that 1st post in the 'Getting the best...' thread. The high-figures in each are minimum requirements to ensure consistent results.
Max voltage range for the 5V supplies will be <5.5V... Allo uses some LDOs that list that as a maximum. You'd be MUCH safer with 4.9V-5.1V.
@cdsgames was going to provide a max voltage for the +-15V in the 'Getting the best...' thread (post 108), but I don't remember that he had. I suspect you'd be good at 16V, I WOULD NOT go over that without specific info from Allo.
Greg in Mississippi
Max voltage range for the 5V supplies will be <5.5V... Allo uses some LDOs that list that as a maximum. You'd be MUCH safer with 4.9V-5.1V.
@cdsgames was going to provide a max voltage for the +-15V in the 'Getting the best...' thread (post 108), but I don't remember that he had. I suspect you'd be good at 16V, I WOULD NOT go over that without specific info from Allo.
Greg in Mississippi
Last edited:
Well then, you might have a similar problem as fabio1068. You would need to check the voltages on the big black cap to to know one way or the other.
Also, you don't need to connect anything to earth ground if you don't want to. You could connect the PCB ground (or probably more correctly, the 'common') at one point to earth as a safety ground if you want.