-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Vera Audio Class-D Amp Build Quality
- Thread starter maty
- Start date
-
- Tags
- hypex nc500 vera audio
Armand has conducted the measurements and would know for certain, but it seems like that's the case.
If someone are wondering if it's worth bridging speakers that dip low in impedance, the graphs below show the output power in bridge vs stereo mode with different loads.
What kind of measurements would you like?
- Armand
20Hz to 20kHz distortion and reactive loads.
For multi-channel processors, the center channel performance.
For AVR/Processors SINAD with PEQ and REQ enabled, even if the measurement occurs outside the correction.
These are not necessarily applicable to this site.
- Rich
Here is up to 20kHz. 250W power into 4Ohm Bandwidth 20kHz. We can see the higher order distortion products falling outside 20kHz
This is 200W measured up to 80kHz BW. Now distortion products up to unhearable 80kHz is taken into the equation as well which is somewhat not that meaningful.
Noise performance at 12dB gain and 30dB gain.
There are a few more measurements here, but we are working on this: https://www.vera-audio.com/product.php?product_id=2&fane=3
About inductive load. I have not found any reason to measure this as I have measured the damping factor to be 13000(!) up to 1kHz and over 1400 up til 20kHz. This means that any error on the output will be corrected no matter what.
But many are asking for this, so I might perform it one day.
This is 200W measured up to 80kHz BW. Now distortion products up to unhearable 80kHz is taken into the equation as well which is somewhat not that meaningful.
Noise performance at 12dB gain and 30dB gain.
There are a few more measurements here, but we are working on this: https://www.vera-audio.com/product.php?product_id=2&fane=3
About inductive load. I have not found any reason to measure this as I have measured the damping factor to be 13000(!) up to 1kHz and over 1400 up til 20kHz. This means that any error on the output will be corrected no matter what.
But many are asking for this, so I might perform it one day.
They use the same output modules, but that is all.
Not saying the competitors offer good value, but we argue our price is fair considering:
1. MUCH more powerful power supply. We use 3kW. This power supply has less loss and keeps the amp cooler plus of course that it has much more capacity with the multiple benefits of that.
2. We have our own custom designed input buffer with the following advantages:
* The latest opamp technology with extremely low noise.
* A very accurate gain control system with 3dB steps. Channel matching is guaranteed 0,01dB on all settings.
* Extremely low noise is offered on all gain settings.
* A very high CMRR of guaranteed 94dB @ 1kHz on both channels. Even better at lower frequency.
* Extremely low noise 0,8uV regulators for the drive system of the Purify 1ET400A modules and the buffer section.
3. Bridging capability using only an knob on the back offering a solid 900W into 4 Ohms with maximum 0,0007% THD+N
4. A very sturdy cabinet milled from one piece af alumimium offer no visible seams or screws (exept bottom plate)
5. Instead og just using Purifi's datasheet measurements from the 1ET400A module, all data on our amplifiers are measured using a R&S UPV audio analyzer guaranteeing the performance on the speaker terminals. All amplifiers are delivered with their own measurement document.
6. A microcontroller is continously monitoring the performance of the amplifier, warning about overheating or operation outside safe limits. This system also gives the user warnings when clipping occur (per channel). Ultimately the amp is shut down if any limit is exceeded.
Our goal with this amplifier is to squeeze out as much performance of the Purifi 1ET400A modules as possible and making a versatile small amplifier. Used as monoblocks they deliver huge amounts of superclean power to the most demanding speakers.
Thank you for taking the time to reply to my thread. I have two questions, one for you and maybe one for this forum:
1) Dear ASR members: For mere mortals like me who have their endgame setup in the basement for music listening/home theater, do we really need that much power to be wanting a Bridging capability and all. And then do we need a that powerful of a supply.
2) Armand! I am sure there is a market for the above market but do you have a plans to release a normal version (read economical) with a 2 channel 200Watts into 8 ohms which is enough for some folks.
I am not saying it is overengineered, but will have to think twice to spend twice the money.
One Purifi Eigentakt module, even with a 3kW power supply, doesn't provide a lot of power (about 150-160W in 8 Ohm with super low distortion). How much one needs and what's sufficient is very depended on the speakers and how load you play. We have had a few customers who tried one VA P150/600 RS who ran out of power and would need to bridge to get enough.
I doubt we will release a lower cost unit with Purifi but I wouldn't out rule it completely either. We do however, plan to release a lower cost 4-channel power amplifier with Hypex Ncore. It will also be powerful, though not quite as powerful as the P400/1000.
I doubt we will release a lower cost unit with Purifi but I wouldn't out rule it completely either. We do however, plan to release a lower cost 4-channel power amplifier with Hypex Ncore. It will also be powerful, though not quite as powerful as the P400/1000.
A Vera Audio P150/600 RS is being shipped out today to Amir for a review. It's also 120V compatible now. Will officially release the amplifier soon in another thread. It's will be offered with the option of silver finish besides black.
The brushed surface becomes more visible with silver finish.
A P150/600 RS is in line to be tested by Amir.
facefirst
Member
Love the look of this. Balanced inputs only means I can't try one as my Naim Uniti doesn't have XLR outs...
restorer-john
Grand Contributor
It is meaningful and something the Class D proponents simply want to brush under the carpet. The requirements for amplifier specifications as they stand are THD and power specified across the audible bandwidth. That includes harmonics of up to 20kHz and requires a proper a measurement bandwidth of ~200kHz to take in up to the 9th or so harmonic. Class Ds already need a filter to prevent slew rate limiting issues in the front ends of analysers, but they don't need a chopped off measurement bandwidth if their switching frequencies are ~500kHz in order to artificially inflate their THD figures for marketing purposes.
You have a R&S analyzer. Use it. Show the honest plots, not the cherry picked, best ones.
Let's see a full rated power (not just 200W) and a decent bandwidth. Then, and only then, you can fairly compare these Class D amplifiers with all other amplifiers that have been rated properly since 1974.
Last edited:
First, we do not show cherry picked plots!
The first plot in post #85 is done at 250W into 4 Ohms which is pretty close to rated power of: "270W with maximum 0.0007% THD+N @ 1kHz".
That plot is bandwith limited to 20kHz and show us what we actually can hear. Maximum audible distortion is 0.0012% around 5kHz
I can produce any plot you request as we have nothing to hide.
Her is a plot I made today from a random sample from the production line.
270W power into 4 Ohms load at 20kHz with the instrument configured to measure everything up to 250kHz. No filter of any kind is installed between the amp in the instrument.
It is very important to understand what we are looking at here. In my experience many people do not understand what 0,035% THD+N at 20kHz actually mean. For some reason they think that the (barely audible) 20kHz tone is somewhat distorted and will "sound" different, but that is not the case. The reason for that is that the distortion occurs at frequencies above 40kHz and will be completely inaudible unless you are a bat.
* First, we have the power. The test tone is 20kHz at 33V. This equals 272W (33^2/4=272W).
* To increase accuracy, a filter inside the instrument mutes the 20kHz test tone with a very narrow band filter before the THD is measured, but we can still see remains of it at 0,15% (50mV)
* Then we can see all the harmonics at 40kHz, 60kHz and so on. The highest one below 200kHz is the 3'rd order at 60kHz with 0.03% (10mV). The second order is only at 0.004% (1.3mV)
* Note that there is absolutely nothing in the complete range of human hearing below 20kHz. Not a single thing above 0.0005% (0.17mV).
In fact, any normal human would hear exactly nothing when feeding a speaker with this signal since most of us has lost the ability to hear 20kHz a long time ago. Not to mention the next signal at 40kHz at only a very small 0.004% fraction of the 33V 20kHz signal.
(Do not try this on a normal speaker as it will melt the tweeter in seconds )
I do not understand how you realistically can think that a x'th order harmonic distortion at 43kHz, 75kHz or 150kHz can affect sound quality? We cannot hear it and the levels are so low they cannot damage or heat up tweeters either.
We could of course add a warning in the manual that if you are looking for an amplifier with very little distortion at inaudible frequencies you should look elsewhere.
The first plot in post #85 is done at 250W into 4 Ohms which is pretty close to rated power of: "270W with maximum 0.0007% THD+N @ 1kHz".
That plot is bandwith limited to 20kHz and show us what we actually can hear. Maximum audible distortion is 0.0012% around 5kHz
I can produce any plot you request as we have nothing to hide.
Her is a plot I made today from a random sample from the production line.
270W power into 4 Ohms load at 20kHz with the instrument configured to measure everything up to 250kHz. No filter of any kind is installed between the amp in the instrument.
It is very important to understand what we are looking at here. In my experience many people do not understand what 0,035% THD+N at 20kHz actually mean. For some reason they think that the (barely audible) 20kHz tone is somewhat distorted and will "sound" different, but that is not the case. The reason for that is that the distortion occurs at frequencies above 40kHz and will be completely inaudible unless you are a bat.
* First, we have the power. The test tone is 20kHz at 33V. This equals 272W (33^2/4=272W).
* To increase accuracy, a filter inside the instrument mutes the 20kHz test tone with a very narrow band filter before the THD is measured, but we can still see remains of it at 0,15% (50mV)
* Then we can see all the harmonics at 40kHz, 60kHz and so on. The highest one below 200kHz is the 3'rd order at 60kHz with 0.03% (10mV). The second order is only at 0.004% (1.3mV)
* Note that there is absolutely nothing in the complete range of human hearing below 20kHz. Not a single thing above 0.0005% (0.17mV).
In fact, any normal human would hear exactly nothing when feeding a speaker with this signal since most of us has lost the ability to hear 20kHz a long time ago. Not to mention the next signal at 40kHz at only a very small 0.004% fraction of the 33V 20kHz signal.
(Do not try this on a normal speaker as it will melt the tweeter in seconds
)I do not understand how you realistically can think that a x'th order harmonic distortion at 43kHz, 75kHz or 150kHz can affect sound quality? We cannot hear it and the levels are so low they cannot damage or heat up tweeters either.
We could of course add a warning in the manual that if you are looking for an amplifier with very little distortion at inaudible frequencies you should look elsewhere.
As discussed in the previous post, there is no point in measuring things that are clearly not audible. That is why all audio analyzers have bandwith limiting so we can measure what is actually audible.
Another ting that looks less favourable in class D amplifiers than in A/B amps is the THD+N measurement when doing a 250kHz BW measurement. That is because the "N" part (the noise) is included all the way up to 250kHz.
Yes, class D amps are noisy as heck compared to a good class A/B amp. The thing is, -the noise frequency is so high that we cannot hear it. Nor can the speakers reproduce it for that matter.
Below is a plot I made today at 33V into 4 Ohms. (272W). 250kHz bandwith and no filters of any kind. The screenshot was taken when the signal was at 4kHz at the second run.
Note that this measurement takes about 2 minutes to complete, so the amp is continously delivering 2x 272W into my load that is now quite hot with its fans at full blow.
We can se a relatively poor THD+N of about 0.01% up to 10kHz, and rising to about 0.03% at 20kHz.
But what is that noise? Well, it starts slowly to rise from about 40kHz, and rises to about 75dB below the signal at around 300kHz. That is about 6mV.
In other words 99.99% of the 0.01% THD+N is noise no human can register.
I see no point in including noise that is clearly inaudible and at very low levels? Remember, this is an Audio amplifier.
Any other measurements you would like John?
Another ting that looks less favourable in class D amplifiers than in A/B amps is the THD+N measurement when doing a 250kHz BW measurement. That is because the "N" part (the noise) is included all the way up to 250kHz.
Yes, class D amps are noisy as heck compared to a good class A/B amp. The thing is, -the noise frequency is so high that we cannot hear it. Nor can the speakers reproduce it for that matter.
Below is a plot I made today at 33V into 4 Ohms. (272W). 250kHz bandwith and no filters of any kind. The screenshot was taken when the signal was at 4kHz at the second run.
Note that this measurement takes about 2 minutes to complete, so the amp is continously delivering 2x 272W into my load that is now quite hot with its fans at full blow.
We can se a relatively poor THD+N of about 0.01% up to 10kHz, and rising to about 0.03% at 20kHz.
But what is that noise? Well, it starts slowly to rise from about 40kHz, and rises to about 75dB below the signal at around 300kHz. That is about 6mV.
In other words 99.99% of the 0.01% THD+N is noise no human can register.
I see no point in including noise that is clearly inaudible and at very low levels? Remember, this is an Audio amplifier.
Any other measurements you would like John?
The input buffer on the Vera amplifiers are fully capable of accepting an unbalanced signal. Just use an adapter like this:
Never mind, haters gonna hate just the same. This has been going on for months/years.
restorer-john
Grand Contributor
No, that's utter BS. Audio analyzers have a minimum of a 200kHz bandwidth for THD measurements and have had since forever. You know it and I know it. Why? So we can capture actual harmonics- not cut them off! In addition, you'll have an A-WTD and fc-400Hz hum and noise filter. 'Audibility' and rated THD specifications are two completely different things.
Your amplifer's THD is poor above 10kHz at anything near full rated power.
Rate your amplifiers THD from 250mW to rated power across the audible spectrum. You know, like the FTC amplifier specifications actually require. Nothing else is meaningful, justifiable or defensible.
Yes!
Actual, honest, compliant with FTC regulation specifications. I'm so sick of "best case", "bandwidth limited" ,"let's change the rules to suit our product" specs or test results.
Last edited:
I'm curious which loudspeakers you have (or anyone else for that matter) that require anything near full rated power above 10khz, and does not melt the tweeter in the process?
Similar threads
- Poll
