PS Audio AirLens

Hi Paul,

if useful in some way to you as soon as you can test the ZENith with AirLens prototype (and really thank you again for this!) please find here below the email I received from Innuos support team:
…when using a separate streamer, i would not connect these to the Streamer port on the ZENith, and instead connect those to the PhoenixNET, so that the PhoenixNET is connected by LAN directly to both the ZENith and the Airlens.

With our new UPnP Bridge feature, it is now possible to use the Sense app to play music on the ZENith through to an external streamer, so long as that streamer uses fairly standard UPnP renderer abilities. We cannot gaurantee this feature works with every streamer as UPnP is very irregular, but so far seems to be okay with the likes of Naim and Linn.

The setup process can be found here;
Connecting Network Streamers – Innuos – High-Fidelity Digital Music Servers and Streamers

Regarding i2s, this interface needs to be standardised first as we get a lot of requests to use it over either USB, Ethernet or HDMI. The industry needs to settle on one of these before we can properly commit to is2 implementation.

Do note that a lot of people promote i2s over normal USB PCM/DSD because they have largely made their evaluations using poor USB sources like Macbooks and so on. When using a very well optimised USB source like a ZENith, it becomes quite a different situation….

A few trusted and respected members here suggested to me to take into account Roon because it sounds as good as Sense if not better (and it provides many other well known qualities compared to Sense, of course - that are not considered in my actual comparisons, because my attention is just focusing on SQ). I will certainly try Roon again after its new release (today) and when the DS MK II will be here in my system. I would only like to know at the moment using the AirLens if Sense (on the ZENith) is an option or not and how it sounds.

Secondly it would be interesting to understand if the new DS MK II, with new USB galvanic isolation improvements, can change the importance of I2S connection. Until today I2S is considered a preferable way (over USB) to connect a streamer to the DS DAC. The AirLens itself which role will play in future if the ability to provide a I2S connection to the DS MK II is going to be less significant compared to new USB upcoming input performances?

Thank you Paul for your work and your effort in clarifying these aspects, elevating my knowledge on this arguments.

1 Like

This is good knowledge and yes, the AirLens has UPnP and so that should work just fine as they describe it.

I2S will always be better than USB if for no other reason than the nature of its separate clocks.

USB on the new DS MKII will be a game changer because it is 100% galvanically isolated.


LMS and squeezelite integration with uPnP continues to impress me, as do the Airplay plugins for it - works in both directions, playing to e.g. appleTVs from LMS but also playing from Apple devices to uPnP renderers.

This convinces me it would also work fine with the AL, and would allow the AL to play Airplay,
so it should be possible and practical to build a system with a Raspberry Pi or two, and an AL, and have it do all this good stuff.

1 Like

Would love to hear more about this and how it is implemented. Was this discussed and I just missed it?

If this is true, and if the MKII’s USB can do DSD256, and the performance gap is narrowed between USB and I2S, maybe I can finally get rid of the Matrix 2. Wishful thinking.

1 Like

Indeed, that has been part of my thinking re the MKII DAC. But regarding the AirLens, I was waiting to hear if the AirLens would support Tidal Connect… which is sounding more like the case from @jamesh.

@ipeverywhere Yes, there has been discussion of galvanic isolation for USB on the DAC MKII… but on the “Wishes Upcoming PSA DirectStream MKII” thread…


A few years ago, I used the Iso Regen from Uptone which provides the galvanic isolation in the USB output and it definitely had a positive effect. The question is whether the galvanic isolation in USB in MKll is equal to I2S or not.

Indeed, the AirLens supports Tidal connect (and Spotify as well).

1 Like

One of the big improvements between DS1 and DS2 is that every input (including USB) is galvanically isolated (as well, the output continues to be galvanically isolated as in the DS1). This is a pretty big deal, sonically speaking, and one of the reasons why I cannot wait for people to wrap their heads around just how big of an improvement the MK2 DS is over it’s parent. It’s been 7 years but now it’s time to take the next big leap in digital audio. We should be delivering in October barring anymore parts pandemic nightmares.


You know, now that you wrote this I remember Ted posting the board layout and all the additional bits of kit on the inputs to make this function without some exotic USB receiver. OK, all caught up. I’m not sure October I’ll be in position for new DAC (I’m evaluating subs right now) but eventually I would really like to get FPGA DAC on the rack and this is the frontrunner for sure.

I’m positive it will be a hit!

If galvanic isolation is a critical component of “good digital” would an off-grid battery powered USB to SPIDF converter like the Audiophilleo1 with its companion PurePower pack (Audiophilleo1 - advanced low jitter USB S/PDIF processor - audiophilleo) be a beneficial companion to those of us with DS MKI’s? This device will deliver up-to DSD64 or PCM384kHz (which is the bandwidth limit of SPDIF) vs the DSD128 of the current DS USB input, but offers full galvanic isolation in return.

I think the problem with the USB in MKI is more than just electrical noise from what I recall. Galvanic isolation should help but I am staying with I2S for MKl. I don’t have the desire to upgrade to MKii immediately but I am eagerly waiting for the release of AirLens.

October for Beta testing or are they all ready out there and October for sale on the website Paul?

1 Like

Beta testing in October if the creek don’t rise.


Will there be a maximum and minimum length of HDMI cable to be used with the AirLens?

Good question and one I do not know the answer to. I would suspect you could get away with 6 to 10 feet but beyond that there might be some loss. Just a guess.

1 Like

Other than length are there any design specifications that make an HDMI cable into a superior I2s cable?

Maybe go over to the “recommended HDMI cable” thread. I think PS Audio came out kinda saying the 48 series from Audioquest is nice. That leaves A LOT of room though to spend a lot more money. Plenty of the folks on this thread have already spent probably 50% of the price of an AirLens on the fancier versions with various mythical beast names… just while waiting on the AirLens. My perspective… without any scientific data whatsoever… but with actual usage with a Silent Angel feeding a PS Audio DAC over the cable… is that the bottom AQ Pearl 48 series is good enough for me. It is six times the price of the usual Amazon HDMI cable. But still under 50 bucks / quid / euro…

1 Like

I2S has redundant clock pins
AirLens is galvanically isolated from the network (input)
DSII is galvanically isolated from the input

If the I2S cable makes a difference then one of those 3 things isn’t implemented correctly. Isolation and redundant clocks should make the cable immune from interacting with what you hear coming out of the DAC…

(this post is to spark discussion, I have no idea of I’m right / wrong in my statements)

1 Like

Is there any benefit from replacing the Bridge II board in a modded DS Sr. with the new AirLens? If it is replaced, I would feed the DS Sr. using an I2S cable from the AirLens.

In my system, the ethernet cable is plugged into the Bridge II. Some Bridge II components are covered in 3M EMI absorber paper AB6005S. The Bridge II is also walled off from other boards using the 3M paper as a ‘screen’. The analogue board in the DS is feed by an external LPS (Farad Supera3).