Vintage Audio: Measurements of Arcam Black Box 3 mk1

[Blumlein 88]

The 42 dB shift makes sense, but why would it apply to the pass band only and not also to the stop band?

Here is a graph from Stereophile's review of a Project Box S2.

Notice the red level of his - 4 dbFS pink noise is -40 dbFS until the stop band. And how the stop band noise level is about the same as when he is using the 0dbFS 19.1 khz tone in blue. That is what is really happening.

I'm thinking REW reset its 0 db level when you told it to generate 0 db pink noise. And that in fact its 0 level should read -42 dbFS or close to that. When you tell it to generate 0 db tones it is also 0 dbFS. So while done this way it looks different the actual levels have to be like in JA's graph above. You know it can't have 0 dbFS levels at each frequency over the band of the pink noise because that total signal would be well into clipping if it were true. It would be like having 100 hz at 0 dbFS plus 200 hz at 0 dbFS plus 300 hz at 0 dbFS so and so forth with that actually true at every single hz. The total signal would be way too hot.

BTW, it would be a good idea to use - 4db FS for the noise signal like JA does. Pink noise has some variation in level (think peak to RMS ratio) of up to 10 db. Most of it is only a little over 3 db. So doing this at -4 dbFS means rarely would there be any clipping. Done at 0 dbFS clipping will occur now and again. This is most noticeable when sampling the analog result at a different sample rate than the signal is generated or when oversampling is done.

Here is a 44.1 khz white noise 1 second each at -4 db, -3 db, -1 db and 0 db. Then the same resampled to 384 khz. If your ADC and DAC are separate you of course could adjust level at the ADC to mitigate this. But the lower level reduces the odds of inter-sample overs that some DACs don't handle correctly. Red is clipping of course in this image below.

 

[Duckeenie]

I see it as vintage digital audio gear. CD came in 1983, the first consumer DACs AFAIK a few years late. Seen from this perspective a DAC made in 1991 is really old (28 years), ergo vintage.

You're correct of course but the problem with that rationale and the motivation for my original light hearted comment is that if the 90's are vintage then that would make some of around here, well...ancient.

 

[Blumlein 88]

You're correct of course but the problem with that rationale and the motivation for my original light hearted comment is that if the 90's are vintage that would make some of around here, well...ancient.

Please, some of around here are experienced and wise. Maybe not me, but some of us.

 

[Duckeenie]

Please, some of around here are experienced and wise. Maybe not me, but some of us.

I'll have you know I helped my 7 yr old Granddaughter with her homework once.

 

[LTig]

Notice the red level of his - 4 dbFS pink noise is -40 dbFS until the stop band. And how the stop band noise level is about the same as when he is using the 0dbFS 19.1 khz tone in blue. That is what is really happening.

I'm thinking REW reset its 0 db level when you told it to generate 0 db pink noise. And that in fact its 0 level should read -42 dbFS or close to that.

I did not use REW for sending white noise because it does not support different sample rates for in and out. I used the vlc player on my Linux workstation to send the 44/16 white noise signal via the Edirols Toslink output to the BB3. The BB3s analog output fed the input of the RME which was connected to my Windows Notebook running REW with 384/24. I had to use the Notebook because on Linux REW supports only 192 kHz. This is no problem though because I prefer the notebook for measuring because it runs on battery and the RME can do the same, so there is no galvanic coupling. REW actually showed the noise level at around -40 dB and I shifted it up to 0 dB because I didn't know better. So it's all my fault.

And many thanks for the explanation, I think I got it now. I had made one measurement with a series of sinus signals (1kHz to 21 kHz, 1kHz apart). I had it already in the review and then realized that it does not show what I wanted to show so I removed it (but somehow it is still there as an attachment, but nevermind). The point is that when I created it I had to reduce the level of each sinuns to prevent clipping of the sum. And with noise it's just the same, only that there are as many signals as FFT points.

BTW, it would be a good idea to use - 4db FS for the noise signal like JA does. Pink noise has some variation in level (think peak to RMS ratio) of up to 10 db. Most of it is only a little over 3 db. So doing this at -4 dbFS means rarely would there be any clipping. Done at 0 dbFS clipping will occur now and again. This is most noticeable when sampling the analog result at a different sample rate than the signal is generated or when oversampling is done.

Here is a 44.1 khz white noise 1 second each at -4 db, -3 db, -1 db and 0 db. Then the same resampled to 384 khz. If your ADC and DAC are separate you of course could adjust level at the ADC to mitigate this. But the lower level reduces the odds of inter-sample overs that some DACs don't handle correctly. Red is clipping of course in this image below.
View attachment 25752

I adjusted the ADC of the RME such that its display did not show clipping, so I think I was on the safe side. But for future measurements I shall reduce the level.

 

[LTig]

You're correct of course but the problem with that rationale and the motivation for my original light hearted comment is that if the 90's are vintage then that would make some of around here, well...ancient.

It's just a matter of the personal view point, and this changes with age ...

 

[LTig]

Please, some of around here are experienced and wise. Maybe not me, but some of us.

You mean white instead of wise, and that in respect to the color of hair (if any are left)

 

[restorer-john]

Looking at the schematics I cannot see that a lower resistance would change the levels.

Can you post the relevant schematic section here?

What are these lifted leg resistors all about in the left channel?

 

[LTig]

Can you post the relevant schematic section here?

Yep, first the DAC output stage:

And here the analog output stage:

The upper L/R output is the main output, the lower one the aux output.

What are these lifted leg resistors all about in the left channel?

View attachment 25754

Someone has good eyes
Forgotten to mention, that's part of the MK1 mod. One left channel (R10), one right channel (R110). I thought that the additional load of the voltage divider (2 kOhm) for an unused output (aux) was no good idea. With the voltage diver active the output level would be lower by 0.74 dB, but still higher than the standard 2 V (2.29 V and 2.15 V).

EDIT: there's one more mod I did not mention yet ...

 

[Jimster480]

Correct me if I am wrong but.... the performance here is actually not as bad as some other units Amir has tested?

It looks like around 83-84 SINAD? is this not worlds better than the AudioGD stuff?

 

[LTig]

Correct me if I am wrong but.... the performance here is actually not as bad as some other units Amir has tested?

It looks like around 83-84 SINAD? is this not worlds better than the AudioGD stuff?

Yes, it is. It's IMD and Jitter were the problems are (if they were not introduced by my mods).

I've just changed the op-amps back to the originals and while THD and IMD are the same as before S/N got better by about 1 dB. So the op-amp rolling actually made the performance worse insteads of better. The IMD is plagued by spurious signals every 100 Hz which rings a bell regarding the 5V power supply with those hot diodes.

 

[invaderzim]

You're correct of course but the problem with that rationale and the motivation for my original light hearted comment is that if the 90's are vintage then that would make some of around here, well...ancient.

There was a 'news' story about a new 80's themed diner opening around here. I was thinking "Well, that is just stupid. That wasn't long enough ago to be nostalgic for it." Then I realized that the time between then and now was the same as when in the late 80's the 50's diners were popular.

 

[sergeauckland]

There was a 'news' story about a new 80's themed diner opening around here. I was thinking "Well, that is just stupid. That wasn't long enough ago to be nostalgic for it." Then I realized that the time between then and now was the same as when in the late 80's the 50's diners were popular.

As I get older, time seems to get compressed. I see the '80s as recent, whilst in the '80s, the 1950s felt like the Dark Ages.

S.

 

[Blumlein 88]

As I get older, time seems to get compressed. I see the '80s as recent, whilst in the '80s, the 1950s felt like the Dark Ages.

S.

Well were you born in say 1960, in the 80's the 1950's would only be a historical time heard about thru the telling of others. A time in history. While the 1980's would be a time you lived right in the middle of your life. Even if you were born in the mid 1950's it would mostly be a time you didn't have memory of having lived.

 

[sergeauckland]

Well were you born in say 1960, in the 80's the 1950's would only be a historical time heard about thru the telling of others. A time in history. While the 1980's would be a time you lived right in the middle of your life. Even if you were born in the mid 1950's it would mostly be a time you didn't have memory of having lived.

I remember the 1950s in England as a grey and grimey decade, all smoke and bomb sites, so different to the brightness of France or Italy at that time. It was almost as if we didn't get any colour until the mid 1960s. By the 1980s, it was looking back as a Bad Time, and another world.

S

 

[LTig]

As I get older, time seems to get compressed. I see the '80s as recent, whilst in the '80s, the 1950s felt like the Dark Ages.

S.

I once read that the middle of the perceived life time is around 18, meaning the time until you're 18 feels as long as the time from 18 til death. Oh shit ...

 

[Blumlein 88]

I once read that the middle of the perceived life time is around 18, meaning the time until you're 18 feels as long as the time from 18 til death. Oh shit ...

I can sort of believe that.

I have noticed, and read articles to the same effect, that older people get impatient. I've seen the idea presented as it being about percentages of your life. One year for something when you are 10 is 10% of your total life. A long time. 1 year when you are 50 is only 2%. That isn't long, hurry up with things I'm not getting any younger. I've thought about it like reverse half life. Half your life at 50 is 25 years. Half your life at 25 is 12.5 years. Half your life at 10 is 5 years (and some of that you don't remember).

 

[LTig]

I can sort of believe that.

I have noticed, and read articles to the same effect, that older people get impatient. I've seen the idea presented as it being about percentages of your life. One year for something when you are 10 is 10% of your total life. A long time. 1 year when you are 50 is only 2%. That isn't long, hurry up with things I'm not getting any younger. I've thought about it like reverse half life. Half your life at 50 is 25 years. Half your life at 25 is 12.5 years. Half your life at 10 is 5 years (and some of that you don't remember).

Yes, life is logarithmic: audio levels for the ears, lightness for the eyes, even feeling of time ...

 

[SIY]

If old people are impatient, then why the hell are they doing 10 under the limit in the left lane???

 

[LTig]

I'd be replacing the bridge rectifier diodes for what appears to be the 5V digital supply. The board looks a bit toasty... Or did you spill your beer on them?

View attachment 25718

I measured the voltages before (at the diodes) and after (at the small red elcap, 10 µF, left of the big black one, 6800 µF) the voltage regulator (7805, above the small red elcap):

• before: 12.4 V DC with 390 mV ripple of 100 Hz
• after: 4.96 V DC and 150 mV noise, no ripple to see.

The regulator has no heat sink and when I touch it with my fingers it does not feel hot, not even warm. The supply current can't be higher than 100 mA I guess, so it beats me why the diodes (BYD33G, 1.3 A max) could ever get so hot that the laquer on the PCB got burnt.

Anybody here have an idea why Arcam chose to use "Fast soft-recovery controlled avalanche rectifiers" and not just simple bridge rectifier diodes? I can't find them here and the only diodes on hand are 1N5404 (3A) or 1N4004 (1A), both standard rectifiers.