Douk Audio H7 Amplifier Review

[AdamG247]

Or a different wow?

That. I leave it to interpretation.

 

[pma]

My naive idea was, if speaker loads could be simulated based on Amir's measurements, then I would have created a few typical speaker loads as templates and calculated the corresponding amplifier frequency responses with an Octave script to estimate how big the influence of the dynamic load on the amplifier is.

This is not a naive idea, it is fully workable in fact. You get all necessary info from idealized amplifier, its complex output impedance Z1 and complex load impedance Z2. Transfer function is defined by the divider Z2/(Z1+Z2) created from those complex impedances. The formula below show that it is not possible to make a divider from |Z1| and |Z2| only, as was suggested, to get the FR.

 

[sofrep811]

In general I do disregard the cost. But I can't ignore the category a product is in. Should we crown 1000 watt amplifier with SINAD of 120 at $5,000 the top and everything then gets progressively lower score? Clearly that wouldn't make sense and would garner more complaints. In other categories such as DACs, it doesn't cost a lot to achieve very high performance. But with speakers, and amplifiers, there is direct cost in getting higher and higher overall performance. Getting 20 Hz response in a speaker is not going to happen in a $200 bookshelf speaker.

So what I have done is informally created a category of budget desktop amplifiers. In there, amplifiers like this do very well due to very low noise and distortion. The SINAD here is better than 100+ other amplifiers regardless of price! It beats the median for all amplifiers ever tested. All for $140 including all the other features it brings. If I give poor marks to amplifiers like this, where does that leave amplifiers with SINAD of 55 and lots of noise and other issues?

Members are free to ignore my subjective assessment and vote in the poll. That is there to avoid people complaining about my criteria for evaluation.

That makes sense. I was just curious as to what you thought about it.

 

[pseudoid]

One would also need a no load or 1k load FR test as a benchmark if one wants to calculate Rout.

Rout = 8R is a true 'dummy' load that was standardized for the sake of expediency, in laborious manual testing instruments available circa 1970s.
Based on conversation replies, the wanted benchmark may best be defined/calculated in units of "Zout".
Sounds complex but AP was not even in diapers back then.
I recall @amirm discussing a new (and expensive) load-simulator AP add-on rig for his set-up. But I don't recall if that unit was able to switch between different reactive (Zout) loads (aka 'smart' load).

 

[pma]

One would also need a no load or 1k load FR test as a benchmark if one wants to calculate Rout.

Zout of power amplifier is normally measured by current injection into the amplifier output. One gets the complex output impedance, both magnitude and phase. The method is described even in the IEC 60268-3 standard. I described it here at ASR as well. One can measure even very low impedance by this method.

 

[solderdude]

Yes, my remark was about calculating Z out based on FR at 4 and 8 ohm load and then you would need no load as well as per the question.
Not saying this is the best way to do this.

 

[NoxMorbis]

Looks like they have a bad channel in that specific unit? Otherwise, wow, pretty clean and powerful. I was considering this unit or the Aiyima 08 Pro, but this one has no sub out, so that nullified that choice.

Guys, the dual volume is for people who mix music.

What is the difference between Power vs Distortion and Power vs Frequency test, because none of the frequencies match the Power vs Distortion THD+N reading?

 

[C. Cook]

I wonder what the decision making process was for using two volume controls and not a single one? And if there are two channels/volume controls but a single VU meter, which one is the meter indicating?

 

[Kip]

This is a review and detailed measurements of the Douk Audio H7 stereo class D desktop amplifier. It was kindly drop shipped to me by a member and costs US $144.
View attachment 278158

Right away we see differentiation in the form of the nice looking VU meter. Not only that there is a control to set it sensitivity which is a must in my view. The backlight is nice and bright and goes well with that yellow/orange ring. The case is machined aluminum so feels nice and heavy for its size. The switches and controls are surprisingly robust and impart a feeling of quality. In other words, this is in entirely different class compared to typical budget desktop amplifiers. Surprised continue to the back side:
View attachment 278159

Yes, we have real, full size XLR inputs! Of course there is also RCA input but they also added a set of RCA outs. I am assuming that is to connect subs and out but there is no internal filtering.

The heart of the unit is TI's TPA3255 single chip stereo class D amplifier with very lofty specs:
View attachment 278160

Wide input range is provided. Included power supply is manageable with 32 volts output with 5 amps of current. If you are going to drive higher impedance speakers and you want to go louder, you may want to up that to 48 volts.

One negative which actually follows some "high-end" gear is dual volume controls. Yes, to increase or lower the volume, you have to crank both of them. I would have wanted to see a single volume control with a balance knob. As is, you can set them once and use an upstream DAC with its volume control (and remote). That is how I tested it by setting the gain to 29 dB and treating it as a power amplifier. You can see the position of the volume controls in the picture for that gain setting.

Douk Audio H7 Measurements
Let's start with our usual 1 kHz dashboard into 4 ohm load using XLR input:
View attachment 278161

The SINAD of the good channel at 82 dB is above average for all amplifiers tested. But for some reason the other channel (in blue) is distorting more which drags the average down some:
View attachment 278162

Zooming in:

View attachment 278163

Noise performance is quite good for class although short of the TI specs:
View attachment 278164

As is typical of budget class D amplifiers, the frequency response varies with what load (speaker) you connect to it in treble region:
View attachment 278165

This means you may hear brighter or softer sound depending on exact impedance of the speaker. Best to use EQ and tailor that to taste.

Distortion is kept very low at 5 watts:
View attachment 278166

The less good channel lost 1 bit here (not shown). So still quite good and more than enough for the application.

Crosstalk is very good for any amplifier let alone a budget one:
View attachment 278171

Let's see how it does in power department starting with 4 ohm load:
View attachment 278167

This is good bit of power and only gets better if we allow 1% THD:
View attachment 278168

I ran this three times as I could not believe it breaking into 3 digit department! 200 watts out of this little box? Granted, the test is short term but still, that is quite healthy.

Switching to 8 ohm we naturally get less power:
View attachment 278169

Note that most speakers are around 4 ohm in bass frequencies where most of the volume is so don't go optimizing for this unless you know you have an 8 ohm speaker in that region.

EDIT: forgot originally to post the power vs frequency:
View attachment 278428

There is only a dB or so impact due to warm up (one channel getting better and the other, worse) so I would just turn it on and use:
View attachment 278172

Finally, there is likely to be some audible pop on both turn on and off:
View attachment 278173

Kind of strange to see the blue channel die out that way. This may have something to do with that channel also underperforming in the dashboard.

I had no luck trying to test the H7 with my powercube reactive loadbox. It caused some kind of interference that would keep it from switching loads. Will need to spend more time investigating what is going on.

Conclusions
The Douk H7 breaks new ground in multiple areas from nicely done VU meter, nice looking case, and very good measured performance. The only issue is if you can live with dual volume controls. I am happy to see such differentiation in this highly price sensitive and crowded market.

I am going to recommend the Douk Audio H7 amplifier.
-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

This is a review and detailed measurements of the Douk Audio H7 stereo class D desktop amplifier. It was kindly drop shipped to me by a member and costs US $144.
View attachment 278158

Right away we see differentiation in the form of the nice looking VU meter. Not only that there is a control to set it sensitivity which is a must in my view. The backlight is nice and bright and goes well with that yellow/orange ring. The case is machined aluminum so feels nice and heavy for its size. The switches and controls are surprisingly robust and impart a feeling of quality. In other words, this is in entirely different class compared to typical budget desktop amplifiers. Surprised continue to the back side:
View attachment 278159

Yes, we have real, full size XLR inputs! Of course there is also RCA input but they also added a set of RCA outs. I am assuming that is to connect subs and out but there is no internal filtering.

The heart of the unit is TI's TPA3255 single chip stereo class D amplifier with very lofty specs:
View attachment 278160

Wide input range is provided. Included power supply is manageable with 32 volts output with 5 amps of current. If you are going to drive higher impedance speakers and you want to go louder, you may want to up that to 48 volts.

One negative which actually follows some "high-end" gear is dual volume controls. Yes, to increase or lower the volume, you have to crank both of them. I would have wanted to see a single volume control with a balance knob. As is, you can set them once and use an upstream DAC with its volume control (and remote). That is how I tested it by setting the gain to 29 dB and treating it as a power amplifier. You can see the position of the volume controls in the picture for that gain setting.

Douk Audio H7 Measurements
Let's start with our usual 1 kHz dashboard into 4 ohm load using XLR input:
View attachment 278161

The SINAD of the good channel at 82 dB is above average for all amplifiers tested. But for some reason the other channel (in blue) is distorting more which drags the average down some:
View attachment 278162

Zooming in:

View attachment 278163

Noise performance is quite good for class although short of the TI specs:
View attachment 278164

As is typical of budget class D amplifiers, the frequency response varies with what load (speaker) you connect to it in treble region:
View attachment 278165

This means you may hear brighter or softer sound depending on exact impedance of the speaker. Best to use EQ and tailor that to taste.

Distortion is kept very low at 5 watts:
View attachment 278166

The less good channel lost 1 bit here (not shown). So still quite good and more than enough for the application.

Crosstalk is very good for any amplifier let alone a budget one:
View attachment 278171

Let's see how it does in power department starting with 4 ohm load:
View attachment 278167

This is good bit of power and only gets better if we allow 1% THD:
View attachment 278168

I ran this three times as I could not believe it breaking into 3 digit department! 200 watts out of this little box? Granted, the test is short term but still, that is quite healthy.

Switching to 8 ohm we naturally get less power:
View attachment 278169

Note that most speakers are around 4 ohm in bass frequencies where most of the volume is so don't go optimizing for this unless you know you have an 8 ohm speaker in that region.

EDIT: forgot originally to post the power vs frequency:
View attachment 278428

There is only a dB or so impact due to warm up (one channel getting better and the other, worse) so I would just turn it on and use:
View attachment 278172

Finally, there is likely to be some audible pop on both turn on and off:
View attachment 278173

Kind of strange to see the blue channel die out that way. This may have something to do with that channel also underperforming in the dashboard.

I had no luck trying to test the H7 with my powercube reactive loadbox. It caused some kind of interference that would keep it from switching loads. Will need to spend more time investigating what is going on.

Conclusions
The Douk H7 breaks new ground in multiple areas from nicely done VU meter, nice looking case, and very good measured performance. The only issue is if you can live with dual volume controls. I am happy to see such differentiation in this highly price sensitive and crowded market.

I am going to recommend the Douk Audio H7 amplifier.
-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

I own this unit and a couple of Aiyima A08 pros. I noticed the Douk A7's VU meter is very slow. Even when I compared them side by side with the same audio input, and reducing the bass and treble on the A08 to minimum, I saw the Douks VU move very slow and lethargic.

 

[mhardy6647]

I own this unit and a couple of Aiyima A08 pros. I noticed the Douk A7's VU meter is very slow. Even when I compared them side by side with the same audio input, and reducing the bass and treble on the A08 to minimum, I saw the Douks VU move very slow and lethargic.

Perhaps (?!?) it would be interesting to compare their ballistics to a bona fide VU meter.

The peformance of true VU (Volume Unit) meters was deliberately implemented -- a long time ago -- to deliver, in effect, visual data on average signal levels for recording and for level matching on remote audio via (typically) leased telephone "Lines" (the source of the term "line level", as well).

 

[Kip]

Perhaps (?!?) it would be interesting to compare their ballistics to a bona fide VU meter.

The peformance of true VU (Volume Unit) meters was deliberately implemented -- a long time ago -- to deliver, in effect, visual data on average signal levels for recording and for level matching on remote audio via (typically) leased telephone "Lines" (the source of the term "line level", as well).

I agree, but this doesn't come from expecting much. Both units use the same chip so it's interesting how they behave differently.

 

[mhardy6647]

I agree, but this doesn't come from expecting much. Both units use the same chip so it's interesting how they behave differently.

ahh, I see.

 

[pseudoid]

Perhaps (?!?) it would be interesting to compare their ballistics to a bona fide VU meter.
The peformance of true VU (Volume Unit) meters was deliberately implemented -- a long time ago -- to deliver, in effect, visual data on average signal levels

What have you achieved w/your reply? You made me spend time and got me sad while doing it!
Ballistics? Not "bullet-proof VUmeters" but a meter's pointer counter-balance (damping) act.

Old fogy-engineers get nostalgic when they see a VOM (Triplett? Simpson?).
In early 1970s, "DMMs" (and the 'DVM' before it) were not yet on workbenchs.

Question: Is it made in the U.S.?
Answers:
1)Yes, In 1985, Simpson Electric was purchased by the Lac du Flambeau Band of Lake Superior Chippewa Indians. One of the last brands to stay USA. << May 2, 2016
2)All Simpson Electric Company meters are still manufactured in Lac du Flambeau, WI. They are not made in China. << June 20, 2021
3)No it's not. It use to be. Now it's made in China. << January 25, 2021

 

[mhardy6647]

Old fogy-engineers get nostalgic when they see a VOM (Triplett? Simpson?).

I resemble that remark! Well, other than not being an engineer.

 

[NoxMorbis]

Thank you @amirm,
Some years ago, I had purchased a ClassD amp which arrived with a 32Vdc adapter, but the webpage had stated that it was a 48Vdc adapter.
When inquired about the discrepancy I was told that the TI's TPA3255 life-expectancy significantly reduces with DC@48V.
Is this true?
I also don't know whether the use of a higher input voltage results in any significant changes to your measurement results (re: @32Vdc).

From what I have read, the sweet spot is 36V.

 

[NoxMorbis]

It matter to me, 1 dB is audible. It shows load dependency. Combine this with the fact that the amp didn't play nice with the reactive load Power cube test, not a good look.

It doesn't reach +1dB until 19k at 4 Ohms and -1dB at 30k and 8 Ohms. You can hear that? At 18 years old, most people can only hear 15–17 kHZ, after which they can't hear jack, and by 50 years, it's down around 12k. I'm just wondering?

 

[Holdt]

It doesn't reach +1dB until 19k at 4 Ohms and -1dB at 30k and 8 Ohms. You can hear that? At 18 years old, most people can only hear 15–17 kHZ, after which they can't hear jack, and by 50 years, it's down around 12k. I'm just wondering?

Those deviations are with resistive loads, right?
This can be several dB's with complex loads at other frequencies depending on phase and impedance.

 

[arancano]

OOOOOoooo and there are OP amps to roll au plenty. (j/k.)

hi, i just got the h7. does it use op amps? do u know the type? have u had experience with sparkos op amps? how in the world do u take the lid off the case. i've gotten it partly out but it seems i'm not doing it right. before i break something i wanted to check with someone who has already done it.

 

[arancano]

Dual volume? I feel like this is some joke from a Rick and Morty episode... but it was dual-bag bro!! This is a terrible design design, one that made me vote headless panther. Very poor design decision.

I disagree. It was a great design decision. I get why u don't like it but there are many i believe that do. once i adjust for the room i don't change the volume since i control it from the preamp.

 

[arancano]

Amir, I got it and love it. Questions: hi, does it use op amps? do u know the type? have u had experience with sparkos op amps? how in the world do u take the lid off the case. i've gotten it partly out but it seems i'm not doing it right. before i break something i wanted to check with someone who has already done it.