The Sentinel DAC - Design Update #1

Here is a sneak peek at an old prototype assembly for the Sentinel DAC. You may notice the 2021 date stamp on the PCB. We have often explained how a lot of technology has already trickled down from this project to the Digital Director and eventually the Cascade DAC. At this level of performance, it takes many iterations to create a product we are truly 100% satisfied with.

This is one of two DAC channels that include16 Hybrid DAC MkII Modules each, 32 modules per DAC. We already started a complete re-design of this section of the DAC, but this gives a glimpse of where we started. I don’t recommend reading too far into what you see, this will be changing dramatically.

Stay tuned here on the forums for updates on our progress as we work towards bringing our most ambitious project to life.



1 Like

Did I understand that the DAC modules will be redesigned from the Cascade to the Sentinel or just the supporting circuitry for the modules?

Thanks
Eric

1 Like

Just the supporting circuitry! The Sentinel will use the Mark II version of the hybrid DAC modules just like the Cascade DAC. Unlike the Cascade DAC’s 8 modules, the Sentinel will feature 32 Hybrid DACs.

2 Likes

Are the extra’s separated on each channel or are they split in a fully balanced signal chain. No need to answer if that answer divulges any proprietary data. I’m curious because I am awaiting delivery of a Cascade DAC (that replaced Wadax) and anticipate an upgrade to Sentinel if justified by the performance and improvements. Thx Eric

Eric, congratulations on the upcoming Cascade DAC! Please let us know your thoughts after you receive it.

I can discuss a little. There will be two of the modules pictured, one for each channel. Fully isolated right and left channels. That will result in 16 Balanced Hybrid DAC’s per channel. Each channel is planned to have two dedicated outputs for bi-amping etc, that is why you see two outputs on this prototype. Subject to change of course.

Thanks Daniel. I was a long time Select II owner. I went to Wadax frankly because I didn’t see a lot of new products of this level coming from MSB. You have fixed that problem so I am back!

Eric

Got it. I think your DAC is shipping today actually. Will be curious to hear your thoughts when it arrives.

As a small company, we do take some time to develop new tech, and we only release it if it really moves the needle forward. The Cascade is one of those products, and then the Sentinel is going to be over the top! Very excited to see it come to life. I do have my work cut out for me though.

Will the bi-amping capabilities include a cross-over? I’d guess that the Sentinel will have more than enough DSP computing horsepower to implement a cross-over.

That is not part of the current plans for the dual output. Still, nothing is set in stone. It will certainly have the processing power for it.

1 Like

If this is too long a question, please just say so, I don’t want to waste your time. I know you don’t just measure to a spec but that you a lot of critical listening, and that listening is your primary determinate for how something works.

I think it would be interesting if, when you do your listening test, you do one set without pass filters on the amplifier inputs and another with pass filters on the inputs.

Here is why I think that…

I asked about the cross-over because of the way I think bi-amping works, which, of course, could be completely out to lunch :slight_smile: Here goes my understanding. Hopefully others here will clear up my bi-amp understanding…

First of all, as far as I know, speakers that allow bi-amping still use their internal low and high pass filters. When you are not bi-amping the input is just sent to both filters… the woofers see the low frequency stuff and the mid/tweeters see the high frequency stuff.

When you bi-amp the connection between the high and low pass filter inputs is broken. One amplifier sends a signal to the low-pass filter in the speaker and the other amp sends a signal to the high-pass filter.

Now comes the part that confuses me about bi-amping. If you don’t have a cross-over before the amps, then each of the amps is sending a full-range signal to each of the pass filters in the speaker. The low-pass filter “throws away” all the high-frequency energy and the high-pass filter “throws away” all of the low-frequency energy.

I don’t understand the interaction between a speaker pass filter and an amp. I seems like, if you don’t use a cross-over, that a 100 Watt amp will act like it has less power than spec’d because some of it is “thrown away”. Also, there is a lot of power going across a speaker cable, rather than a signal cable where almost no current flows.

To me, a cross-over in a speaker just redirects energy among some speakers, it never “thows away” any. But that can’t happen if the link between the pass filters is broken.

On the other hand, if there are pass filters are on the input to the amplifiers (like they are on my McIntosh MC901s), the pass filters on the speakers have almost nothing to “throw away”.

Dan

Dan, great question. You are poking at a very big dilemma in this industry. As we “approach” the best designs, we lose the ability to interact with other products. This line of thought would eventually bring us to active speaker designs. Which, technically, have the best possibility for the best performance.

But from the perspective of a manufacturer, the market is not there to support this product category. People want the ability to choose components and build their ideal system. The more specific and dedicated the products become, the more we isolate consumer choice and become more of a walled garden.

You have the right ideas, but it would have to be customized per amplifier/speaker solution which isn’t sustainable for most companies.

2 Likes

Daniel, I see what you are saying, it makes sense.
Dan

1 Like

We try not to leave anything on the table with our designs, but user input is always appreciated! Please feel free to reach out with any ideas large or small. Thanks Dan.

Daniel,

Thanks for taking the time to go through my comments. I will post design more questions/ideas as they come to me.

Dan

1 Like

Nice pictures on the dac modules. As a new and happy Cascade owner, I am very pleased to know that the modules in my dac is of your latest development. Could I ask you why it is called “hybrid”?

2 Likes

Wonderful! You are one of the few to receive their Cascade so far. The Cascade DAC is actually a trickle down design from the Sentinel… it is just the Sentinel was too big of project to get out first. The Hybrid DAC’s in your Cascade are the same as the Sentinel and will be compatible for the upgrade path.

They are called the Hybrid DAC’s because dynamically configurable between a discrete native DSD, or discrete native PCM DAC. The DAC has two modes: a multibit PCM mode and a massive parallel single-bit discrete DSD mode. They’re not limited by audio formats, sample rates, or even speed. We can operate our current DAC modules at up to 6 MHz for PCM and up to 50 MHz for DSD.

Thanks for the reply. Technically speaking, would there be a difference between feeding music upsampled to dsd to the digital director (by hqplayer, for example), compared to feeding the native pcm?

I don’t foresee any benefit to Upsampling before the DD. If you are going for the most original/natural/accurate version, send the native information over. There are no file changes you can do that take more advantage of the Cascade etc than the native files.

If you want to modify or change the experience, this would be a great way to tailor it to your tastes. We do have a lot of users who like to play around with with source processing. If you like it, you are doing it right :wink: This is more about modifying the recording than optimizing the playback process.

Thanks again for the kind reply. Since we are talking about a hobby, experimenting is almost irresistible.

1 Like

Absolutely!!

1 Like