That sounds excellent.
I was just asking because the previous figures that you quoted were "THD+N" so I just wanted to be sure that we were comparing apples to apples.
Apple to apples its the right way..however a lot of the N in the THD+N comes from actual PSU and some (in testing) from the Gdn of the RPI.
THD is more from the dac design and analog stage.
No question!
However.
I'm not talking about tweaks here!
Ian introduced a seperate HAT for the outputstage.
This way it can easily get swapped. This is what you IMO should look at at.
Many people appreciate output transformer stages. They prefer them over
opamp stages. I'm just saying...
And then there will be people who'd love to see a tube-HAT-stage.
It's about future options.
All that doesn't mean opamp stages - if done right - are a bad thing.
It's a good option to start with.
Does the power/control board act as an isolator between the Pi and the rest of the stack, or would you need to use the isolator board in addition to the 3 board stack to remove noise coming in from the Pi?
No question!
However.
I'm not talking about tweaks here!
Ian introduced a seperate HAT for the outputstage.
This way it can easily get swapped. This is what you IMO should look at at.
Many people appreciate output transformer stages. They prefer them over
opamp stages. I'm just saying...
And then there will be people who'd love to see a tube-HAT-stage.
It's about future options.
All that doesn't mean opamp stages - if done right - are a bad thing.
It's a good option to start with.
@Soundcheck,
I'm a bit confused by what you wrote. Allo has had a separate output stage board in the Katana design from the beginning:
I'd say the difference between Allo & Ian's approaches is that:
- Allo's DAC is intended to give you the best SQ they can manage in a R-Pi DAC HAT-sized set of boards that can be easily by a single 5V supply, but provides some flexibility for tweaking.
- Ian's approach is to give you very good sounding DAC base and one (or maybe eventually more) output board(s) where you can roll your own.
The 1st is great for a product that can be used by a wide swath of consumers, the 2nd is great for many of us DIY'ers here in DIYAudio.
I'm happy that both are out there and intend to use both when they are available.
Just my 2 cents.
Greg in Mississippi
P.S. For a DAC HAT with a separate transformer output stage already available, you can go buy a TauDAC at about 650 Euro.
Last edited:
Hmmmmm....we found out that our AP machine has a noise of 6uV on RCA and 10uV on XLR by default ( using loop back).
Our noise in RCA is of 13uV right now and 25 of XLR (removing base noise thats 7uV on RCA and 14 on XLR)
THD+N is solved. I don't have the exact calculation but its better than -105 THD+N on both RCA and XLR (1Khz).
Final version of PCBs (after measuring and tweaking every power rail , about 45 different ones) is being worked on and will sent to PCB house for mass production.
Also as a side note , we have developed the 5V SMPS that goes along with it. 10 days to new prototype. We expect to set a new record in 5V SMPS ( low noise) in both input (towards mains) and output (towards your DAC/RPI) . Special attention was directed towards "leakage currents" on output and EMI reduction at the source. (diode and mosfet)
Its hard to explain how many hours we slaved in reducing loop area on mosfet (imagine a tetris game) we changed the mosfet 3 times , last time to gain about 4mm sq (less) of loop area . Also ferrite beads were strategically placed in gate driver , drain and r2dc snubber (+ 2 on source resistors)
Along with less EMI , we have as a side effect that we need less Y capacitance to pass EMI test and less Y capacitor = less leakage .
Our noise in RCA is of 13uV right now and 25 of XLR (removing base noise thats 7uV on RCA and 14 on XLR)
THD+N is solved. I don't have the exact calculation but its better than -105 THD+N on both RCA and XLR (1Khz).
Final version of PCBs (after measuring and tweaking every power rail , about 45 different ones) is being worked on and will sent to PCB house for mass production.
Also as a side note , we have developed the 5V SMPS that goes along with it. 10 days to new prototype. We expect to set a new record in 5V SMPS ( low noise) in both input (towards mains) and output (towards your DAC/RPI) . Special attention was directed towards "leakage currents" on output and EMI reduction at the source. (diode and mosfet)
Its hard to explain how many hours we slaved in reducing loop area on mosfet (imagine a tetris game) we changed the mosfet 3 times , last time to gain about 4mm sq (less) of loop area . Also ferrite beads were strategically placed in gate driver , drain and r2dc snubber (+ 2 on source resistors)
Along with less EMI , we have as a side effect that we need less Y capacitance to pass EMI test and less Y capacitor = less leakage .
No question!
All that doesn't mean opamp stages - if done right - are a bad thing.
It's a good option to start with.
Its done right
Handmade. (rather checked manually for offset). Only film caps on the path (or LPF) analog side.No it does not. This week we have scheduled to test the isolator board as well to see whats the reduction in noise. However when you arrive at 6-7uV its hard to see the improvement.
Its been weeks that we are in testing now, we have hundreds of reports.
Unit sounds..like a ess9038q2m with a class A stage..
Noise on oscillators rails is 600uV , analog stage 900uV . Everything is under 1mV.
This week we might release the driver of the DAC , or rather the driver of the MCU.
We have final PCBs , we will check it a few times and send it to PCB house
Unit sounds..like a ess9038q2m with a class A stage..
Noise on oscillators rails is 600uV , analog stage 900uV . Everything is under 1mV.
This week we might release the driver of the DAC , or rather the driver of the MCU.
We have final PCBs , we will check it a few times and send it to PCB house
Great to see the progress.
My Qs:
1.
I don't see any supercaps as buffer or any other special caps you've been using on e.g. the Boss for the analog part!?!? Budget? No need anymore ?
Especially on class-A I'd expect plenty of local & fast power to be
a rather good idea.
2.
I don't see any external power options (jacks/jumpers) as we've been discussing several times for your other products!?!?
I mean, you applied related changes to e.g. your recent Boss revision.
Change on strategy once more?
3.
Will the new 5V SMPS, that can probably be used with the new HATs, also
be launched in conjunction with the new DAC?
My Qs:
1.
I don't see any supercaps as buffer or any other special caps you've been using on e.g. the Boss for the analog part!?!? Budget? No need anymore ?
Especially on class-A I'd expect plenty of local & fast power to be
a rather good idea.
2.
I don't see any external power options (jacks/jumpers) as we've been discussing several times for your other products!?!?
I mean, you applied related changes to e.g. your recent Boss revision.
Change on strategy once more?
3.
Will the new 5V SMPS, that can probably be used with the new HATs, also
be launched in conjunction with the new DAC?
1. They are there on the ess9038 PCB (better even then Boss)..dual
2. On the analog stage , not present , present on the controller board . You can feed external 15V. For the 5V you have 2 ways of feeding it + you can feed the RPI independently . I will try to make a doc explaining it.
3. Next week I will have a definitive answer on the SMPS (testing the latest PCB)
Ah. There's another board. A controller board!?!
That's sounds more than just a power controller / power board!??
But that one at least generates +/-15V from 5V - doesn't it?
Only one 5V SMPS would than be required for this HAT solution!?!
Things are getting more complex by the post.
You said. You'll write it all up. We'll be patient.
Just to mention it.
The supercaps and these caps that'd been suggested by Greg (forgot the type) on the Boss do power the Boss DACs internal opamp output stage.
At least that's IMO their main benefit.
What sense does it make to have them now on the DAC HAT only!?!?
To me it'd even make more sense to have them on the output stage.
Anyhow. Never mind. I'm not the HW expert here. I do know usually
you have your reasons for doing things the way they are done.
In the end, what matters is the final result!
I think it would be great if you'd launch both products DAC and SMPS at the same time. Or better, the SMPS first.
It's better from a marketing perspective.
That's sounds more than just a power controller / power board!??
But that one at least generates +/-15V from 5V - doesn't it?
Only one 5V SMPS would than be required for this HAT solution!?!
Things are getting more complex by the post.
You said. You'll write it all up. We'll be patient.
Just to mention it.
The supercaps and these caps that'd been suggested by Greg (forgot the type) on the Boss do power the Boss DACs internal opamp output stage.
At least that's IMO their main benefit.
What sense does it make to have them now on the DAC HAT only!?!?
To me it'd even make more sense to have them on the output stage.
Anyhow. Never mind. I'm not the HW expert here. I do know usually
you have your reasons for doing things the way they are done.
In the end, what matters is the final result!
I think it would be great if you'd launch both products DAC and SMPS at the same time. Or better, the SMPS first.
It's better from a marketing perspective.
New Raspberry Pi 3+
Raspberry Pi 3 Model B+ on sale now at $35 - Raspberry Pi
Another round of testing maybe....It is a little faster and networking has been improved.
Raspberry Pi 3 Model B+ on sale now at $35 - Raspberry Pi
Another round of testing maybe....It is a little faster and networking has been improved.
Is Volumio site going to offer the Kali + Piano 2.1 bundle again
The Volumio site shows out of stock.
Or is there a more important reason why it is no longer offered?
And BTW. What's the general consensus if my goal is to BiAmp (tubes on top, SS on the bottoms) using the Piano 2.1?
My current streaming source is a RPi I2S connected to a HFiBerry DAC running Volumio on the RPi. I know much has matured in this space, and I want BiAmp and a volume control in the app.
The Volumio site shows out of stock.
Or is there a more important reason why it is no longer offered?
And BTW. What's the general consensus if my goal is to BiAmp (tubes on top, SS on the bottoms) using the Piano 2.1?
My current streaming source is a RPi I2S connected to a HFiBerry DAC running Volumio on the RPi. I know much has matured in this space, and I want BiAmp and a volume control in the app.
I see that the Piano 2.1 is out of stock at Allo.com too. No doubt they would be able to tell you when new stock is expected.
Quite a few people (including me) like the Kali/Piano2.1 in dual mono mode best. However , if you are biamping you may prefer to use dual stereo so that each amp has its own input.
you prompted a question...
Have the problems with setting the crossover point in 2.1 or 2.2 mode been solved? Do I still need the TI software to adjust? ... or is the crossover adjustable from the Volumio user interface? And exactly what can I adjust from Volumio?
Are there things I can adjust from the TI software that I cannot from Volumio (assuming I can adjust anything from Volumio)?
Has the process of getting the TI software gotten easier?
Boy! what a bunch of questions...!
Have the problems with setting the crossover point in 2.1 or 2.2 mode been solved? Do I still need the TI software to adjust? ... or is the crossover adjustable from the Volumio user interface? And exactly what can I adjust from Volumio?
Are there things I can adjust from the TI software that I cannot from Volumio (assuming I can adjust anything from Volumio)?
Has the process of getting the TI software gotten easier?
Boy! what a bunch of questions...!
