Great to read your posts Greg. I concurr completely with your findings:
I've been trying the USBridge Sig with and without an Isolator and with various other Allo DACs before I put the Katana atop of it. The improvement over the Pi3 is evident with only a Piano or Boss DAC. I can only conclude that it will easily outperform the Pi4 with its even greater PSU noise. The isolator is to my ears a backward step. The Kali together with a Piano DAC is already better than both of my well developed Katana stacks when it is on the USBridge Sig, very much as described by Greg's friend. A separate power supply for my USB connected music drive brings about a small improvement as well.
I have to get used to just how good this sounds before I replaced the Kali/Piano with one of my Katana stacks.
Power supplies are currently non-choke regulated AMB Sigma 11's fed by 9-0-9v toroids. Load current for the USBridge Sig and the Kali/Piano settles to appro 500mA. USB drive supply oscillates between 45mA and 220mA as it gathers data. OS is Picoreplayer and local LMS with Edimax WiFi adaptor. A very neat and extremely satisfying combo. Shame we cant get a second crystal integrated into the Pi Compute module.
A quick question about the USBridge Sig's Clean Power input if anyone knows please? What is the difference between it and the micro USB socket power input and what benefits will there be from raising the applied voltage to the 9volts max?
Many thanks
John
I've been trying the USBridge Sig with and without an Isolator and with various other Allo DACs before I put the Katana atop of it. The improvement over the Pi3 is evident with only a Piano or Boss DAC. I can only conclude that it will easily outperform the Pi4 with its even greater PSU noise. The isolator is to my ears a backward step. The Kali together with a Piano DAC is already better than both of my well developed Katana stacks when it is on the USBridge Sig, very much as described by Greg's friend. A separate power supply for my USB connected music drive brings about a small improvement as well.
I have to get used to just how good this sounds before I replaced the Kali/Piano with one of my Katana stacks.
Power supplies are currently non-choke regulated AMB Sigma 11's fed by 9-0-9v toroids. Load current for the USBridge Sig and the Kali/Piano settles to appro 500mA. USB drive supply oscillates between 45mA and 220mA as it gathers data. OS is Picoreplayer and local LMS with Edimax WiFi adaptor. A very neat and extremely satisfying combo. Shame we cant get a second crystal integrated into the Pi Compute module.
A quick question about the USBridge Sig's Clean Power input if anyone knows please? What is the difference between it and the micro USB socket power input and what benefits will there be from raising the applied voltage to the 9volts max?
Many thanks
John
@matthiasw, I've asked him. I'll relay when I get an answer.
@John Luckins, great to read one of your posts as always. AND glad you like the USBBridge Sig as much as I do.
Interesting to try it with the Kali/Piano stack first. I got a little jaded when I tried my first RPi DAC HAT with that used separate output stage devices. That add IMHO lifted any competent build over anything I could do with any of the ES9023/PCM51xx-based DAC HATs. BUT the Piano on a Kali, just like the original Boss, are great for that genre.
Curious on your results with that combo with the Isolator. Before a reclocker is one of the situations where they could shine. BUT of course they are going to provide the best benefit when the source device is noisy, which the USBBridge Sig is not in my experience. Good to hear that personal experience and confirmation of how that combo works. Datapoints are ALWAYS good!
Bummer about the non-choke supplies... you are no longer using one of the most impressive RPi/DAC power setup in existance! Though Ian Canada's setup where he has 2 3000F Ultracaps as a final output buffer C for the RPi power has got to be pretty close!
On the power to the USBBridge Sig, that 9V is for the Digione Signature, not the USBBridge Sig. The power for the USBBridge Sig is specified at 5V and I remember @cdsgames said max is 5.4V or 5.5V. DON'T go over that recommendation. I poked around on the board and quickly found an LP5907 regulator, which is rated for an absolute max input of 6V.
On what voltage, I've gone as high as 5.2V and as low as 4.9V without hearing any clear correlation in the sonics. BUT I was not looking for that at that time.
AND based on the results my 'audio friend' reported bypassing the USB-C on the Katana and what I heard when I did that, I would definitely use flying leads at the 5V clean power inputs on the USBBridge Sig.
I'll have some updates on the Katana without Isolator on the USBBridge Sig AND some further thoughts on Isolator use in the near future.
Later!
Greg in Mississippi
@John Luckins, great to read one of your posts as always. AND glad you like the USBBridge Sig as much as I do.
Interesting to try it with the Kali/Piano stack first. I got a little jaded when I tried my first RPi DAC HAT with that used separate output stage devices. That add IMHO lifted any competent build over anything I could do with any of the ES9023/PCM51xx-based DAC HATs. BUT the Piano on a Kali, just like the original Boss, are great for that genre.
Curious on your results with that combo with the Isolator. Before a reclocker is one of the situations where they could shine. BUT of course they are going to provide the best benefit when the source device is noisy, which the USBBridge Sig is not in my experience. Good to hear that personal experience and confirmation of how that combo works. Datapoints are ALWAYS good!
Bummer about the non-choke supplies... you are no longer using one of the most impressive RPi/DAC power setup in existance! Though Ian Canada's setup where he has 2 3000F Ultracaps as a final output buffer C for the RPi power has got to be pretty close!
On the power to the USBBridge Sig, that 9V is for the Digione Signature, not the USBBridge Sig. The power for the USBBridge Sig is specified at 5V and I remember @cdsgames said max is 5.4V or 5.5V. DON'T go over that recommendation. I poked around on the board and quickly found an LP5907 regulator, which is rated for an absolute max input of 6V.
On what voltage, I've gone as high as 5.2V and as low as 4.9V without hearing any clear correlation in the sonics. BUT I was not looking for that at that time.
AND based on the results my 'audio friend' reported bypassing the USB-C on the Katana and what I heard when I did that, I would definitely use flying leads at the 5V clean power inputs on the USBBridge Sig.
I'll have some updates on the Katana without Isolator on the USBBridge Sig AND some further thoughts on Isolator use in the near future.
Later!
Greg in Mississippi
Last edited:
Always interesting that there's still any surprise that bypassing USB-C
connectors is well worth the effort.
It's not that we havn't been discussing it since years (before it was uUSB).
Especially when feeding power hungry analog rails through them.
I'm pretty sure that this tweak alone will lower the distance to a
setup with external 15V. Keep in mind. Using 15V gets rid of the
USB-C as part of the deal. That alone will make a difference.
Unfortunately Allo is not listening (I do not believe that's the case though) or simply not willing to provide us
with easily accessible soldering joints/holes to be able to attach proper wires for a quality power cabling of the tower.
(the same issue we face on Shanti)
IanCanadas solutions e.g. FifoPi are simply great. Two screw sockets
one for the dirty (incl. RPI) one with the clean part. Yep, life can be that convenient.
Anyhow. I'm not sure if I asked that before. What's actually the best spot to attach external quality 5V to Katana?
I really do not want to risk to smoke that unit. And I also want to have the option to properly revert back.
connectors is well worth the effort.
It's not that we havn't been discussing it since years (before it was uUSB).
Especially when feeding power hungry analog rails through them.
I'm pretty sure that this tweak alone will lower the distance to a
setup with external 15V. Keep in mind. Using 15V gets rid of the
USB-C as part of the deal. That alone will make a difference.
Unfortunately Allo is not listening (I do not believe that's the case though) or simply not willing to provide us
with easily accessible soldering joints/holes to be able to attach proper wires for a quality power cabling of the tower.
(the same issue we face on Shanti)
IanCanadas solutions e.g. FifoPi are simply great. Two screw sockets
one for the dirty (incl. RPI) one with the clean part. Yep, life can be that convenient.
Anyhow. I'm not sure if I asked that before. What's actually the best spot to attach external quality 5V to Katana?
I really do not want to risk to smoke that unit. And I also want to have the option to properly revert back.
Last edited:
Always interesting that there's still any surprise that bypassing USB-C
connectors is well worth the effort.
It's not that we havn't been discussing it since years (before it was uUSB).
Especially when feeding power hungry analog rails through them.
I'm pretty sure that this tweak alone will lower the distance to a
setup with external 15V. Keep in mind. Using 15V gets rid of the
USB-C as part of the deal. That alone will make a difference.
Unfortunately Allo is not listening (I do not believe that's the case though) or simply not willing to provide us
with easily accessible soldering joints/holes to be able to attach proper wires for a quality power cabling of the tower.
(the same issue we face on Shanti)
IanCanadas solutions e.g. FifoPi are simply great. Two screw sockets
one for the dirty (incl. RPI) one with the clean part. Yep, life can be that convenient.
Anyhow. I'm not sure if I asked that before. What's actually the best spot to attach external quality 5V to Katana?
I really do not want to risk to smoke that unit. And I also want to have the option to properly revert back.
They did provide easy-to-access 5v input on the latest USBridge Sig/Audiophile Pi.
To quote myself:
Let's backpower our PI4s and attach some OSCON SEPCs to that rail (done since years)...
...and see what we end up with in comparison.
Enjoy.
Katana no longer in production?
For the past few days, I have gone on Allo's website and tried to order the Katana DAC. Every time I try to add it to the cart it states "Sorry, this product is not available at this time!" I emailed their support to ask them when the product would be available, and I got no response for the past few days. Can anyone here give me an update?
For the past few days, I have gone on Allo's website and tried to order the Katana DAC. Every time I try to add it to the cart it states "Sorry, this product is not available at this time!" I emailed their support to ask them when the product would be available, and I got no response for the past few days. Can anyone here give me an update?
Hi Mr Greg Stewart,
I try to connect +/-14.3V Salas Shunt reg direct to the pad pins (+15V to pin 1, -15V pin 2, Ground to pin 3), the pad is removed and I get a problem. There is a noisy sound (eee...)from Opamp board. I try to reassambly all 4 stacks many time but can not fix the issue. Lucky, the all of stack still sound ok, ofcause it bonus a bit noisy. ( At the and I try to remove opamp out, and connect it to +-14.3V psu and I find the noisy sound come from here. Any one get the same ? Pls give me some advice in this case!
I try to connect +/-14.3V Salas Shunt reg direct to the pad pins (+15V to pin 1, -15V pin 2, Ground to pin 3), the pad is removed and I get a problem. There is a noisy sound (eee...)from Opamp board. I try to reassambly all 4 stacks many time but can not fix the issue. Lucky, the all of stack still sound ok, ofcause it bonus a bit noisy. ( At the and I try to remove opamp out, and connect it to +-14.3V psu and I find the noisy sound come from here. Any one get the same ? Pls give me some advice in this case!
I seem to recall that the middle pin of the 3 is the common 0v. If the pins are numbered in sequence then you have mis-applied the DC.
John
John Luckins is right: Quoting from the Katana Tech Manual:
J27 input for external +/-15V: +15V (pin1), GND (pin2), -15V (pin3)
NOTE THAT MARKINGS ON THE PCB ARE WRONG FOLLOW PIN +/-/GND ABOVE
Also check that you have the switch and jumper settings as shown in Table-2 just above this.
J27 input for external +/-15V: +15V (pin1), GND (pin2), -15V (pin3)
NOTE THAT MARKINGS ON THE PCB ARE WRONG FOLLOW PIN +/-/GND ABOVE
Also check that you have the switch and jumper settings as shown in Table-2 just above this.
Last edited:
Thanks for yours reply. But I solder direct to 4 pins pad on Opamp pcb. Not 3 pins on MC pcb.
This step I follow Mr Greg guide on few previous page.
Thanks !
Hi there.
Anybody out there who has a proper Katana powering scheme in place?
BG:
Allo strongly suggests to comply to a power-up sequence for ultimate sound.
Which is
1. DAC first
2. RPI next
Now they offer Shanti as ultimate PS for that combo. (I wouldn't agree to that, but that's another topic).
Shanti apparently does not offer a power-up-sequence.
Has anybody figured out how such a power-up sequence could be established with Shanti?
Or any other setup?
And would that power-up sequence actually be as important as Allo suggests?
Anyhow. I do have an idea.
I mentioned somewhere else that IMO Shanti requires relays on it's DC outputs to eliminate these awful turn on/off thumps and cap-sucker-noise (Katana).
If we now would have these relays, I'd add to that wishlist a delay/timer feature that'd allow for a specific power-up sequence.
Thx.
Anybody out there who has a proper Katana powering scheme in place?
BG:
Allo strongly suggests to comply to a power-up sequence for ultimate sound.
Which is
1. DAC first
2. RPI next
Now they offer Shanti as ultimate PS for that combo. (I wouldn't agree to that, but that's another topic).
Shanti apparently does not offer a power-up-sequence.
Has anybody figured out how such a power-up sequence could be established with Shanti?
Or any other setup?
And would that power-up sequence actually be as important as Allo suggests?
Anyhow. I do have an idea.
I mentioned somewhere else that IMO Shanti requires relays on it's DC outputs to eliminate these awful turn on/off thumps and cap-sucker-noise (Katana).
If we now would have these relays, I'd add to that wishlist a delay/timer feature that'd allow for a specific power-up sequence.
Thx.
Last edited:
OK. Are you sure then that your Salas PSU is stable? Can you try a different +/-V supply?
Regarding power-up sequence, I think I saw somewhere a comment from Allo that it's now OK to power all up together. That's what I do (4 supplies in total) and it's fine.
NOW OK??
NOW, because Shanti simply lacks that option?
Inside the offical Katana DAC documentation you apparently still find at the very beginning:
And if I recall it correctly, this sequence had an impact on their measurements. That's why they
recommended it.
Last edited:
To all , please disregard the above message. Its technical vodoo. Power up sequence and measurements impact... None at all
Katana only rule is to power the uController before RPI is booted
RPI takes a few dozens seconds (at least) to boot .
So if you power both sides in same time...its fine. If you power Katana before RPI its fine .
The only thing to watch out is...power RPI...wait to boot and then power Katana side (50% of the time if will have wrong clock to start)
Katana only rule is to power the uController before RPI is booted
RPI takes a few dozens seconds (at least) to boot .
So if you power both sides in same time...its fine. If you power Katana before RPI its fine .
The only thing to watch out is...power RPI...wait to boot and then power Katana side (50% of the time if will have wrong clock to start)
cdsgames. I'm a bit worried about you.
You basically confirm a potential clock issue. That's neither Voodoo nor nonsense.
And then I've been quoting your very own product manual! You basically own the Voodoo part.
It's been you who was bringing the whole issue up in the first place.
Most of us would have never thought about a power-up sequence of that HAT tower.
And if that clock issue is as you describe, my power-up sequence/Shanti question is very valid.
You guys need to get your act together.
And next time think before posting.
Last edited:
I will involve moderators if you keep saying stuff like " And if I recall it correctly, this sequence had an impact on their measurements. "
It does not and it makes no sense ..
"You guys need to get your act together.
And next time think before posting."
"I think" that you should learn how to express your view in a civil manner .
It does not and it makes no sense ..
"You guys need to get your act together.
And next time think before posting."
"I think" that you should learn how to express your view in a civil manner .