EPC is a producer of GaN transistors. GaN gives higher open loop linearity and higher efficiency. Many of the replies have been ignorantly bashing GaN technology when there was obviously a problem with the setup.
-
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!
Premium Audio Mini GaN 5 Review (Stereo Amplifier)
- Thread starter amirm
- Start date
Killingbeans
Major Contributor
It's probably been returned to the owner long ago, but you can buy one youself from Premium Audio and do your own measurements:
EPC's customer base is confidential.
Killingbeans
Major Contributor
What people don't like, is when a manufacturer presents a technology as a magic bullet and uses it as an excuse for throwing proper implementation and engineering to the wind. Premium Audio Products is not putting on a good show (or so it seems). No data. Just claims and user testimonials. Until we see some good implementations of the technology, you can't really blame people for being sceptical.
daniboun
Major Contributor
I think the main concern here is that Premium Audio has not shown good quality control in its products ... moreover I am the first to have dismantled their amp and from one series to another, the inductors have changed and some other components too .... with different values ...
When I spoke to Premium Audio, all they said was that they had to deal with the shortage of electronic components without checking the values of the inductors....
I have nothing to say about the Mini Gan, which in my opinion sounds great ... I was the first surprised by the ASR results but if I can confirm something: the power supply of my Mini Gan has exploded after two months! This is the first time in my life that I had such a problem with a class D amp) Which makes me say that the quality control is really bad!
Not to mention the fact that Premium Audio had committed to present concrete measures of the Mini Gan on their site .... we are soon in 2022 and nothing has changed) Made in the USA you said ? "Made in the USA" is no longer a guarantee of quality))
Last edited:
+1. Thus like you also said, when the system is worse, there is a bad design/setup...
...and then IMHO, ASR is not there to help bad (system) manufacturers to get better...
daniboun
Major Contributor
The only Gan amp that holds up is here:
I wish Amir could test the Orchard Audio.
Checkout VS the Purifi :
audioxpress.com
I wish Amir could test the Orchard Audio.
Checkout VS the Purifi :
Fresh From the Bench: A Tale of Two Class-D Amplifiers Orchard Audio Starkrimson and Purifi Audio Eigentakt EVAL1
Stuart Yaniger examines new amplifier solutions from two disrupting audio companies. Purifi Audio from Denmark, founded by Bruno Putzeys, Lars Risbo, and Peter Lyngdorf, together with a remarkable engineering team; and Orchard Audio, with a product developed from the creativity and...
restorer-john
Grand Contributor
SIY's review is comprehensive. Not sure why it needs to be revisited.
There is a magic bullet with GaN. It gives higher open loop linearity and higher efficiency. EPC would be happy to help designers extract the amplifier value of GaN.
Based on the article, there was much difficulty in getting the system to operate, much less operate correctly. Perhaps the amplifier instructions could have been written better.
I can't conclude whether the amplifier was bad (damaged or design) or the setup was incorrect. Class D is a credible amplifier manufacturer, so I will assume that their design and characterization was good. We would certainly help them improve if necessary.
If anyone would like help in designing your GaN amplifier, please contact me - [email protected]
Based on the article, there was much difficulty in getting the system to operate, much less operate correctly. Perhaps the amplifier instructions could have been written better.
I can't conclude whether the amplifier was bad (damaged or design) or the setup was incorrect. Class D is a credible amplifier manufacturer, so I will assume that their design and characterization was good. We would certainly help them improve if necessary.
If anyone would like help in designing your GaN amplifier, please contact me - [email protected]
daniboun
Major Contributor
two tests are better than one)
And for the cause: the measurements presented on the Orchard Audio site do not coincide with those presented here:
Amplifier SINAD list ASR + 3rd parties
List Name,THD+N (5W, 4 ohms),SINAD (dB) (5W, 4 ohms),Power (W) (1% THD+N, 4 ohms),Price (USD),Source Topping LA90 (stereo, bypass, low gain),0.00009%,121,90,800,<a href="https://www.l7audiolab.com/f/topping-la90/">L7 Audio Lab</a>,Suggestions, corrections and more info here: Topping LA90 Discre...
I believe that an American company can send its product to its neighbor Amir) if the Chinese do it ... between compatriots it is all the easier)
Last edited:
daniboun
Major Contributor
Really ? this is credible for you ? Mount a heatsink without worrying about its role?
I hope, without offending you, that EPC engineers its components with more elegance)
- Joined
- Aug 3, 2019
- Messages
- 921
- Likes
- 1,406
No technology, no matter how superior it may be, is immune to the effects of a bad design implementation or poor QA on the part of the manufacturer. It's not all that rare for great or even revolutionary components to be found in a bad end products, especially when designers are unfamiliar with them due to their sheer novelty.
- Joined
- Aug 3, 2019
- Messages
- 921
- Likes
- 1,406
Yes, "obviously" -- but when it comes to a component like a semiconductor, "the setup" includes the end product under test. The design and/or manufacturing shortcomings of a particular product don't reflect on the component's merit -- so stand down from red alert and please refrain for making inappropriate demands. IOW, "don't shoot the messenger" and deal courteously with the (several possible) reasons for the message!
Killingbeans
Major Contributor
Class D? Are you talking about Hypex? They aren't the designers of this product.
You should contact Premium Audio Products (nauseatingly pretentious name, IMO): https://premium-audio.com/
Here's an interesting exchange about GaN tech as it relates to class d audio use from the owners of Purifi:
@Sagittarius : Where is class D heading? Where are further improvements going to come from? There are companies which now use gallium nitride transistors and are pushing the switching speed into the MHz region. Is this the way forward in your view?
Bruno: Well, with the sort of audio performance we’re getting I’d say that we’re asymptotically approaching “perfect”. One could argue that we passed the point of diminishing returns a few years ago already. I’m not saying that a next step won’t have any audible benefits, but in the grander scheme of things, the margin is shrinking.
Lars: That’s if you stick to audio performance alone. Otherwise we wouldn’t have bothered going to class D to begin with. You don’t do that for audio quality. You do that to get better efficiency, make the amp smaller and yadda. And then you get a new set of problems to fix, such as what it sounds like. And then there’s reliability, manufacturability and so on. I wouldn’t say that GaN is going to be the answer to those things, and neither is upping the switching frequency.
Bruno: Well for a given efficiency you could probably increase the switching rate, but if I’m going to shell out as much for a pair of FETs as what you’d normally pay for the whole amp, I’d rather benefit from that in terms of higher efficiency. Of course, not everyone is able to make that choice. I’ve spent my career honing control loops, most audio designers haven’t and so have to rely on simpler control loops. In that case, increasing the switching frequency is definitely helpful to reduce distortion.
Lars: We’re as fanatic about audio quality as anyone else, but because we’ve got feedback down to a T now we’re not forced to resort to higher switching frequencies.
Bruno: If we need to be geeky and I guess that the folks who are going to read this interview can handle that -eh Thomas?- lets grab the specs for the FET in our 400W Eigentakt module and its closest GaN equivalent. So that’s the FDP42AN15A0 (OnSemi) on our left and the EPC2033 on our right. Rdson: 36mOhm vs 7mOhm. Clear win for GaN here.
Lars: It’s also got a higher current rating (24A vs 48A) so if we want to be fair we should be scaling by about 2:1
Bruno: Oh erm well, that’s still a minor win for GaN because after scaling it’d come up at 14mOhm. Gate charge is of course magnificently low (30nC vs 6nC after scaling) so driver losses would be low and you can turn them on fast. GaN also has zero Qrr so you can do that if you want. But the main thing that sticks in my throat here is output capacitance. Our good old FDP42, which is from 2002 mind you, has an output capacitance at 100V of 70pF whereas the EPC device puts in a whopping half nanofarad (or 250 puff after scaling). That means your idle losses will go up, or you will have to increase dead time to allow the output inductor to recover the extra stored energy at its leisure. And isn’t it just idle losses that more or less determine real-life power consumption in full? And if power consumption isn’t something to care about, why not just stick with class A…?
Lars: And high dead time combined with high switching frequency sounds even less attractive. That just increases open-loop distortion.
Bruno: In applications like motor controllers and high density SMPS GaN and SiC devices are a breakthrough, mind you. It’s just that audio is this weird application where average power is very low and where dead time actually affects performance.
Lars: And GaN is going to mature so this picture is bound to shift at some point. Just not now.
Bruno: True. On the other hand, silicon is doing the same. More recent devices are getting frightfully close to GaN. Sadly they only come in SMD packages that require fairly expensive methods to get the heat out. Like most GaN devices. It looks terribly ancient, but the good old TO220 package is still a neat compromise in terms of performance for the price.
Lars: It’s just a game of tradeoffs. The fact that we’re using normal parts, and the reason why we chose to do so doesn’t make for a sexy story. We all like to believe in a magic bullet but when you’re an engineer you have to make choices based on tangible grounds. So the sexy story we’re trying to push about Eigentakt is that we think it’s a bloody clever design.
@Sagittarius : Where is class D heading? Where are further improvements going to come from? There are companies which now use gallium nitride transistors and are pushing the switching speed into the MHz region. Is this the way forward in your view?
Bruno: Well, with the sort of audio performance we’re getting I’d say that we’re asymptotically approaching “perfect”. One could argue that we passed the point of diminishing returns a few years ago already. I’m not saying that a next step won’t have any audible benefits, but in the grander scheme of things, the margin is shrinking.
Lars: That’s if you stick to audio performance alone. Otherwise we wouldn’t have bothered going to class D to begin with. You don’t do that for audio quality. You do that to get better efficiency, make the amp smaller and yadda. And then you get a new set of problems to fix, such as what it sounds like. And then there’s reliability, manufacturability and so on. I wouldn’t say that GaN is going to be the answer to those things, and neither is upping the switching frequency.
Bruno: Well for a given efficiency you could probably increase the switching rate, but if I’m going to shell out as much for a pair of FETs as what you’d normally pay for the whole amp, I’d rather benefit from that in terms of higher efficiency. Of course, not everyone is able to make that choice. I’ve spent my career honing control loops, most audio designers haven’t and so have to rely on simpler control loops. In that case, increasing the switching frequency is definitely helpful to reduce distortion.
Lars: We’re as fanatic about audio quality as anyone else, but because we’ve got feedback down to a T now we’re not forced to resort to higher switching frequencies.
Bruno: If we need to be geeky and I guess that the folks who are going to read this interview can handle that -eh Thomas?- lets grab the specs for the FET in our 400W Eigentakt module and its closest GaN equivalent. So that’s the FDP42AN15A0 (OnSemi) on our left and the EPC2033 on our right. Rdson: 36mOhm vs 7mOhm. Clear win for GaN here.
Lars: It’s also got a higher current rating (24A vs 48A) so if we want to be fair we should be scaling by about 2:1
Bruno: Oh erm well, that’s still a minor win for GaN because after scaling it’d come up at 14mOhm. Gate charge is of course magnificently low (30nC vs 6nC after scaling) so driver losses would be low and you can turn them on fast. GaN also has zero Qrr so you can do that if you want. But the main thing that sticks in my throat here is output capacitance. Our good old FDP42, which is from 2002 mind you, has an output capacitance at 100V of 70pF whereas the EPC device puts in a whopping half nanofarad (or 250 puff after scaling). That means your idle losses will go up, or you will have to increase dead time to allow the output inductor to recover the extra stored energy at its leisure. And isn’t it just idle losses that more or less determine real-life power consumption in full? And if power consumption isn’t something to care about, why not just stick with class A…?
Lars: And high dead time combined with high switching frequency sounds even less attractive. That just increases open-loop distortion.
Bruno: In applications like motor controllers and high density SMPS GaN and SiC devices are a breakthrough, mind you. It’s just that audio is this weird application where average power is very low and where dead time actually affects performance.
Lars: And GaN is going to mature so this picture is bound to shift at some point. Just not now.
Bruno: True. On the other hand, silicon is doing the same. More recent devices are getting frightfully close to GaN. Sadly they only come in SMD packages that require fairly expensive methods to get the heat out. Like most GaN devices. It looks terribly ancient, but the good old TO220 package is still a neat compromise in terms of performance for the price.
Lars: It’s just a game of tradeoffs. The fact that we’re using normal parts, and the reason why we chose to do so doesn’t make for a sexy story. We all like to believe in a magic bullet but when you’re an engineer you have to make choices based on tangible grounds. So the sexy story we’re trying to push about Eigentakt is that we think it’s a bloody clever design.
- Joined
- Oct 11, 2019
- Messages
- 1,852
- Likes
- 2,772
No one was bashing GaN technology. Rather, an implementation of it by a single manufacturer did not measure well. I think you would be better served to contact that specific manufacturer to discuss what you feel went wrong so that manufacturer can improve their product rather than posting about it here.
- Joined
- Oct 11, 2019
- Messages
- 1,852
- Likes
- 2,772
Now THAT’S what I meant.
daniboun
Major Contributor
Speaking about implementation....
Would like to see a NewClassD @asr and see if a good SiCFET implementation worth it )
"Launched in 2020 the Singularity 3 Class D modules use the newest technology, and are among the best sounding amplifiers in the world. Even in competition with Class A amplifiers. This is leading edge technology. The switching devices in other Class D modules are regular MOSFET's but we use SiCFET's instead, which are about 10 times faster, giving better resolution in the top region, and less switching noise. Due to the latest enhancements in SiCFET technology, these are even significantly faster than the competing GAnFET technolgy in this power class. This may change as technology progresses though.
The audio circuits, such as buffer and modulator are both made with low noise japanese discrete transistors, and thin film precision resistors, to optimize sound quality, beyond what is possible with integrated circuits."
Would like to see a NewClassD @asr and see if a good SiCFET implementation worth it )
"Launched in 2020 the Singularity 3 Class D modules use the newest technology, and are among the best sounding amplifiers in the world. Even in competition with Class A amplifiers. This is leading edge technology. The switching devices in other Class D modules are regular MOSFET's but we use SiCFET's instead, which are about 10 times faster, giving better resolution in the top region, and less switching noise. Due to the latest enhancements in SiCFET technology, these are even significantly faster than the competing GAnFET technolgy in this power class. This may change as technology progresses though.
The audio circuits, such as buffer and modulator are both made with low noise japanese discrete transistors, and thin film precision resistors, to optimize sound quality, beyond what is possible with integrated circuits."
I bought a stereo amp from Premium Audio to run my ridiculously inefficient Stax F81 Electrostatic speakers. I liked it enough to ask Tom to make me mono blocs. They run cool and with zero issues, sonically they best my Rogue Medusa my NuPrime mono’s and Ghent Audio mono’s. More detailed less haze than any class D amp I have used. Bass is exceptional with a smooth extended top. Midrange is a little more forward than other amps but that ok. Value for money is outstanding. I understand measurements are important, but if I measure some of the tube amps I love I think I would be just as shocked. I say let your ears be the judge.This is a review and detailed measurements of the Premium Audio Mini GaN 5 Stereo Class D power amplifier. It was kindly sent to me by a member and costs US $799 (recent price increase).
The GaN 5 comes in a compact enclosure with plenty of ventilation at the cost of decent looks:
View attachment 157253
Beside the sole power button and blue indicator, there are two red LEDs that turn on when you power the unit on and then go off. Strangely, when the amplifier goes into protection, they do not light up and getting the unit working again requires a power cycle. I like to see protection circuits that self recover when the extreme condition goes away.
The back panel shows binding posts that are too close together for easy fitment of my large banana speaker cables. But I like the inclusion of balanced XLR inputs:
View attachment 157254
As you see, there are also jumpers to change the gain setting although the range is not very large (26, 28 and 32). Most of my testing was with medium gain which is as shipped.
I had trouble getting the amplifier to function initially. It would simply not output anything. I played with grounding as I have seen some very low end amps cause AP to not be able to capture their output due to high level of noise (despite my use of AES-17 filter). This didn't work at first so I switched inputs to RCA. That did not work either. After some fiddling and switching back to balanced input, the amp all of a sudden started to work. Not sure how much of this is interaction with Audio Precision or design issue. Either way, it is an exception to the rule of 95% of amplifiers I have measured.
As the name indicates, this amplifier uses the new Gallium Nitride transistors (GaN) which has some advantages over classic Mosfet transistors. All else being equal, it can produce higher efficiency and better performance at expense of higher cost.
Mini GaN 5 Measurements
As noted, I started my testing in medium gain which produced near nominal 29 dB which is my standard for amplifiers. XLR input was used per introduction:
View attachment 157255
Well, this is disappointing. Distortion is fairly high causing SINAD which is the sum of distortion+noise to be worse than a number of budget desktop amplifiers:
View attachment 157256
Noise performance by itself is not good either:
View attachment 157257
I had trouble getting the amplifier to go to full power for the right graph. It would shut down at just 0.7 volt input. Strangely, later on it accepted higher levels of input. I think the impulse nature of this measurement was causing it trouble. So maybe it can produce higher SNR at higher power but still, none of this is competitive.
On the other hand, crosstalk was very good:
View attachment 157258
Multitone performance was also reasonable:
View attachment 157259
What is not reasonable is high dependence on speaker load when it comes to frequency response:
View attachment 157260
This is just not proper. Clearly the output filter is interacting with the load.
Power measurements were quite puzzling:
View attachment 157264
Notice the pull back due to some protection kicking in, likely the power supply rather than the amplifier. I could not remotely get the power rating they specify using medium gain. I thought maybe low gain does better but then the pull back occurred at even low power (blue and green)! Beside lack of power, distortion is also quite high for this price class and category. We are talking about poor Audio/Video Receiver amplifier performance!
Switching to 8 ohm tamed the output a bit but power shortfall is still significant:
View attachment 157268
This being the second amplifier I have recently tested with such power shortfall, I started to doubt my instrumentation. So I pulled out my Purifi reference design and measured it. It tracked the dashed line in the above graph from a year ago 100%. Someone had posted this from their website:
600 watts per channel??? Into 8 ohm? This is the spec from the manual:
View attachment 157269
How on earth did they measure 600 watts when the spec is 200 watts? Strangely, that graph is nowhere to be found on their site. There is some talk about a larger amplifier of which is a shrunk version? Maybe that is the one that produced higher power but why does it say "mini" in the title of above graph?
FYI, this amplifier has the green board which I understand to be the newer design. Maybe the older red one had more power. Hard to say.
Anyway, I tried to run my max and burst power:
View attachment 157272
In Burst mode, the amplifier would keep going into pull back and so I could not compute anything for that. Likewise the THD+N vs frequency and power could not be run. As the frequencies got lower and lower, the point where the amp would pull back would also shrink to lower and lower wattages. What I could get did not show great transfer function.
Conclusions
While I leave the door ever so slightly ajar to some interaction with Audio Precision, the measurements seem to indicate an amplifier that has both design issues and inability to meet specifications. The fact that the company took away the only measurement graph makes it suspect that maybe they had have modified the design and selling something different than what they originally intended. As it is, it is a waste of GaN technology as it brings nothing to the party but high expense.
I can't recommend the Premium Audio Mini GaN 5 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/
It would have been nice if the manufacturer of NewclassD could present relevant and clear measurement data instead of what is on their website now. There are many words, but few data. So yes, an ASR test is needed.Speaking about implementation....
Would like to see a NewClassD @asr and see if a good SiCFET implementation worth it )
"Launched in 2020 the Singularity 3 Class D modules use the newest technology, and are among the best sounding amplifiers in the world. Even in competition with Class A amplifiers. This is leading edge technology. The switching devices in other Class D modules are regular MOSFET's but we use SiCFET's instead, which are about 10 times faster, giving better resolution in the top region, and less switching noise. Due to the latest enhancements in SiCFET technology, these are even significantly faster than the competing GAnFET technolgy in this power class. This may change as technology progresses though.
The audio circuits, such as buffer and modulator are both made with low noise japanese discrete transistors, and thin film precision resistors, to optimize sound quality, beyond what is possible with integrated circuits."
Similar threads
- Poll
