Soekris DAM1021 R2R DAC Measurements

[March Audio]

No. No transients there. That is the whole point. Humans recognize sounds as real or not based largely on the attack (it's an evolutionary thing to keep from being eaten by a mountain lion). Looking at the rise time of a square wave will at least tell a little more.

It will tell a lot about the bandwidth of the circuit. What else, and can you explain why its relevant? All digital is by definition bandwidth limited and at defined frequencies in audio, so will only capture a finite number of the square waves harmonics. So unless a dac has a serious frequency response issue........

Impulse response will inform you about the imaging filter but dont forget the point already made about digital being bandwidth limited....

...and of course that music just doesnt contain these sorts of signals.

 

[March Audio]

To clarify the XLR pin 1 issue: @BE718, could you please confirm that both the dam1021 build and the Motu 8a you measured short pin 1 to the chassis (ideally right next to the connector)? These things are definitely not consistently wired in all equipment, and can introduce (audible!) hum when done incorrectly. It’s hard to tell what’s going on from the photo on page 1 of the thread. There may also be PSU and grounding scheme problems unrelated to the DAC itself (which is what I think @soekris was alluding to).

The 1021 build indeed does have pin1 taken to chassis at entry, you can see this in the picture. The psu has been tried in two conigurations. The first is exactly as instructed in the 1021 manual with the transformers AC outputs directly feeding the onboard psu and as seen in the pictures with an intermediate low noise dc psu which is a method also noted in the manual IIRC. Made no difference to the mains noise.

The QA401 measurement device was used with differential input and has no issue with other dacs. Note the mains issues have been seen with other measurement devices and in the measurements independantly made by SBAF.

I will check the motu later.

 

[Blumlein 88]

Music is not steady-state. Under changing signal conditions the states of sigma-delta modulators are often undefined. Can you offer a repeatable, non steady-state test methodology besides music?

Is music so much non-steady state? Transient and changing, but I don't think it fits what people picture when they hear these descriptions.

Take hard struck cymbals. Many think of it as a super fast transient sound. Here are 4 cymbals recorded with a wide bandwidth measurement microphone.

What you see is what you should expect. You strike a metal cymbal. Sound travels thru it at roughly 4 times the speed of sound in air. The impulse from the strike travels over the surface of the metal disc, reflects back rinse and repeat. It takes a few cycles to reach maximum amplitude with a long tail decay. So this is a few high frequency sines that build to a high level and decay from resonances in the cymbal. This is hardly extreme in the direction of non-steady. It isn't close to square waves. It isn't even close to the rise and fall times of a 19+20 khz IMD test tone. Voice, musical instruments all are multiple resonances for the most part. Not steady like a test signal, but not horrid totally non-resonant chaos as is often believed.

I've heard this description of why test signals aren't music. Yet no one seems to nail down what is missed in music vs test signals. We end up with a lot of conjecture and hand waving. I'd like some concrete examples or some explanations beyond hey music isn't steady like a test tone. I've seen some issues with such things in power amps especially power amps connected to speakers, but not really in DACs. Can you give some more insight into how DACs fail to perform or how multi-bit works in a way that delta-sigma designs fail?

 

[Blumlein 88]

No. No transients there. That is the whole point. Humans recognize sounds as real or not based largely on the attack (it's an evolutionary thing to keep from being eaten by a mountain lion). Looking at the rise time of a square wave will at least tell a little more.

Find me some music (not including heavily limited heavily produced pop) that looks anything like a square wave. I read this all the time, and would like to know where this fast attack music is where we are evolutionarily built to recognize sound this way. There are transients in music and real live sound yes, but I don't see things that come close to stressing our line level electronics.

 

[Blumlein 88]

If this is really the case shouldnt the 1021 sound very harsh due to all the non harmonic spuria noted in in the IMD test?

I think the point that has to be remembered by some is that whilst they are welcome to do some "hand waving" about this and that, they need to providecsome evidence to demonstrate the statements. Its no good just stating "delta sigma sounds harsh and has unknown unmeasureable problems". Nor is saying "just listen".

While there are a lot of spuria in the IMD result of the 1021 I don't know it reaches a level to be audibly bothersome. The highest spike was -94 db with the others lower. While there are lots of them I doubt they are noticeable and real music very rarely will put this much stress on the 1021.

 

[March Audio]

While there are plenty of spuria in the IMD test I don't know it reaches a level to be audibly bothersome in this case.

Whilst I may agree with you, the issue here is that in the next breath we will get some people saying that invisible unmeasureable and unkown issues are of audible consequence.

There is an audiophile problem here, one moment the minutia are important, but when it suits apparantly they are not.

 

[March Audio]

While there are a lot of spuria in the IMD result of the 1021 I don't know it reaches a level to be audibly bothersome. The highest spike was -94 db with the others lower. While there are lots of them I doubt they are noticeable and real music very rarely will put this much stress on the 1021.

.....but also make it incapable of replaying 24 bit material, if that is of importance to you or consider it a valid exercise

With a more complex music waveform it will be a sea of IM products.

 

[March Audio]

Find me some music (not including heavily limited heavily produced pop) that looks anything like a square wave. I read this all the time, and would like to know where this fast attack music is where we are evolutionarily built to recognize sound this way. There are transients in music and real live sound yes, but I don't see things that come close to stressing our line level electronics.

People have not made the connection between bandwidth and signal rise time.

 

[Blumlein 88]

.....but also make it incapable of replaying 24 bit material, if that is of importance to you or consider it a valid exercise

With a more complex music waveform it will be a sea of IM products.

Well yes and no. A complex waveform with many different sines making up the composite signal will mean each of those frequencies are lower in level. So the IM products will be lower in level and well mostly they'll drop below the noise floor.

As for the 24 bit noise floors well, I've yet to make any recordings that are even 20 bit. Actually I don't think I have made one at 16 bit. More common is 10-12 bits worth.

 

[RayDunzl]

You strike a metal cymbal. Sound travels thru it at roughly 4 times the speed of sound in air. The impulse from the strike travels over the surface of the metal disc, reflects back rinse and repeat. It takes a few cycles to reach maximum amplitude with a long tail decay. So this is a few high frequency sines that build to a high level and decay from resonances in the cymbal.

I had the opportunity to play with a support cable on a TV tower (1974 or so)

Near the anchor point, were some heavy dampers; a clamp, and weights on a stub of cable (two above, two below).

We found we could whack the cable with the dampers by getting the damper weights swinging on their cable stubs.

The sound generated started at a very high frequency (short but increasing section of the cable (like 2 inches thick steel) vibrating, sort of a long yeee-ooww-oohhhhhh sound.

It was quite amusing.

Eventually we got the cable excited at a very low frequency, maybe 1 Hz, and then heard something start banging around up at the top of the tower. There was no way to stop the cable (thousands of pounds, tower height around 440 meters).

Saw the transmitter man come out of the hut looking up and scratching his head, and we left.

https://www.google.com/maps/@27.840...4!1s_4fCJn1_FvTg6n8rHh4yBQ!2e0!7i13312!8i6656

Bummer, design change, nothing to bang on the cable now. Has some kind of roller damper 100 feet up.

https://www.google.com/maps/@27.840...4!1sahqBYGmaItUq8b3XWV7EFA!2e0!7i13312!8i6656

I've tried to duplicate the effect on a guitar string, it's probably there, but it just isn't the same.

 

[rebbiputzmaker]

Well yes and no. A complex waveform with many different sines making up the composite signal will mean each of those frequencies are lower in level. So the IM products will be lower in level and well mostly they'll drop below the noise floor.

As for the 24 bit noise floors well, I've yet to make any recordings that are even 20 bit. Actually I don't think I have made one at 16 bit. More common is 10-12 bits worth.

Just to get a handle on the group thought process here.

So, the transient nature of music, the the peak rise in energy, (and the tax on power supply delivery this may cause) decay, the harmonic overtone complexity etc, can be mimicked by steady state test tones?

 

[Wombat]

Just to get a handle on the group thought process here.

So, the transient nature of music, the the peak rise in energy, (and the tax on power supply delivery this may cause) decay, the harmonic overtone complexity etc, can be mimicked by steady state test tones?

Read-up on Fourier Analysis.

 

[rebbiputzmaker]

I had the opportunity to play with a support cable on a TV tower (1974 or so)

Near the anchor point, were some heavy dampers; a clamp, and weights on a stub of cable (two above, two below).

We found we could whack the cable with the dampers by getting the damper weights swinging on their cable stubs.

The sound generated started at a very high frequency (short but increasing section of the cable (like 2 inches thick steel) vibrating, sort of a long yeee-ooww-oohhhhhh sound.

It was quite amusing.

Eventually we got the cable excited at a very low frequency, maybe 1 Hz, and then heard something start banging around up at the top of the tower. There was no way to stop the cable (thousands of pounds, tower height around 440 meters).

Saw the transmitter man come out of the hut looking up and scratching his head, and we left.

https://www.google.com/maps/@27.840...4!1s_4fCJn1_FvTg6n8rHh4yBQ!2e0!7i13312!8i6656

Bummer, design change, nothing to bang on the cable now. Has some kind of roller damper 100 feet up.

https://www.google.com/maps/@27.840...4!1sahqBYGmaItUq8b3XWV7EFA!2e0!7i13312!8i6656

I've tried to duplicate the effect on a guitar string, it's probably there, but it just isn't the same.

 

[rebbiputzmaker]

Read-up on Fourier Analysis.

That was not my question!Nor is it the same. smh

 

[amirm]

So, the transient nature of music, the the peak rise in energy, (and the tax on power supply delivery this may cause) decay, the harmonic overtone complexity etc, can be mimicked by steady state test tones?

Mimic? You can do one better: you can create test signals with infinite energy which would never be capturable in real life. The J-test is one such signal by the way.

 

[rebbiputzmaker]

Mimic? You can do one better: you can create test signals with infinite energy which would never be capturable in real life. The J-test is one such signal by the way.

How does that stress power supply/ power deliver system? ?

 

[amirm]

How does that stress power supply/ power deliver system? ?

What? I thought you knew the answer to that when you asked the question.....

 

[Wombat]

That was not my question!Nor is it the same. smh

No, silly. It is where you will find the answer.

 

[RayDunzl]

So, the transient nature of music, the the peak rise in energy, (and the tax on power supply delivery this may cause) decay, the harmonic overtone complexity etc, can be mimicked by steady state test tones?

From my own observation, analysis of a sweep tone creates a display with (in my humble case) virtually the same output as an actual step or impulse sound recorded by microphone.

See the Impulse and step response calculation from sweep vs actual in-air recording. https://www.audiosciencereview.com/...1-r2r-dac-measurements.2324/page-7#post-67355

How does that stress power supply/ power deliver system? ?

In my case I played the sweep (semi steady state) , and I played the square wave and I played the impulse (transients).

The agreement of the "calculated from sine sweep" vs "recorded from real playback", surprised me.

If it's this close in the air with my little analysis toys, I see no reason for it not to be excruciatingly closer/more accurate using real tools, to look at other aspects of the input vs output.

 

[Wombat]

What? I thought you knew the answer to that when you asked the question.....

So much for all of that inferred knowledge and experience which after several requests to be elucidated upon has not been proffered. Somewhat telling. Courage of conviction issue??