Thought this new amp from Peachtree Audio deserved it's own thread. Enjoy!!
This Peachtree could be something, it uses the far faster newest "GaN Technology" (Gallium Nitride) output devices, not Mosfets like Class-D uses.
1: has no dac in it per se, just has a (programable?) DSP processor chip to convert PCM to PWM at the end just before the GaN power output stage.
2: is zero feedback
3: no preamp is used
4: and has no buffers
5: has impressive low .004% THD, very good for an amp without any feedback!!!,
6: It just takes in the digital output from your digital source/s (it has no analog inputs).
Addicted to Audio has the older Peachtree line but not this new GaN-1 "digital-amp" yet.
Busisoft the Au importer said they'll be in soon, but yet to have an RRP put on them.
This is Peachtree Audio, and the new GaN Technology with this digital-amp, that is direct digital input from your source/s.
https://www.peachtreeaudio.com/products/gan1
Here's JD ex-Aussie, (gone back to his Pommy roots,) John Darko's very descriptive high praise review on it.
Cheers George
This Peachtree could be something, it uses the far faster newest "GaN Technology" (Gallium Nitride) output devices, not Mosfets like Class-D uses.
1: has no dac in it per se, just has a (programable?) DSP processor chip to convert PCM to PWM at the end just before the GaN power output stage.
2: is zero feedback
3: no preamp is used
4: and has no buffers
5: has impressive low .004% THD, very good for an amp without any feedback!!!,
6: It just takes in the digital output from your digital source/s (it has no analog inputs).
Addicted to Audio has the older Peachtree line but not this new GaN-1 "digital-amp" yet.
Busisoft the Au importer said they'll be in soon, but yet to have an RRP put on them.
This is Peachtree Audio, and the new GaN Technology with this digital-amp, that is direct digital input from your source/s.
https://www.peachtreeaudio.com/products/gan1
Here's JD ex-Aussie, (gone back to his Pommy roots,) John Darko's very descriptive high praise review on it.
Cheers George
Zero Feedback is almost certainly a marketing blurb. It is unfortunate that we'll (probably) never know what type of DSP they are using inside that thing, but I am pretty sure there's a tight feedback loop used.... How tight would depend on their PCM-to-PWM algorithm used.
I bet that thing uses SMPS... here's a little night
https://www.diyaudio.com/community/threads/post-you-smps-noise-spectrum-measurements.394518/
I think the first statement their web page says a lot We Made Too Many! SAVE NOW - 35% OFF!" If they are as good as they say they are then they could never make too many! Please correct me if I'm wrong but I understand GaN Technology to be another MOSFET type that holds less charge so both N and P type switching can be better controlled. This certainly improves reliability but not sure what it does for sound quality in Class D amplification.
I own one of the earlier Peachtree designs, very poor component quality. Came to me with leaky caps and needing a major rebuild even though it was only 18 months old. Somewhat colored my view of the brand.
These GaN amps are open loop- they have no feedback taken after the output filter. As a result, their frequency response is load dependent. You don't have to listen to these to know this is true- it's a result of the design. This was an issue with the Tripath class d amps from some 25 years ago. Class d amp designers figured out that putting the output filter within the feedback loop (all modern class d designs do this) cured this issue. I look at this as simply poor, outdated design given that there is an established solution to the issue that most modern class d amps don't suffer from. Like I said, kind of a joke to market these as using the latest and greatest GaN devices and then put them into a class d 1.0 circuit that is about 30 years old.
On second thoughts... Why do they put so much effort into marketing? I swear the world's marketers have made so much colourful packaging that they've definitely earned some R&R.
They say "no DAC" but the whole thing basically IS a digital to analogue converter. The switcher itself could be the size of a production version of a Raspberry Pi hat. If we're lucky, most of the additional bulk could be power factor correction, which requires a lot of energy storage so the mains doesn't get contaminated with low frequency noise. But I suspect that it's mostly thick chunks of brushed aluminium for no real reason.
Looking at the GaN FETs, they could indeed be interesting because of the lack of body diode that the usual MOSFETs are stuck with. So, inductive spikes from the speaker load could be managed in some interesting ways. But they never talk about this stuff! Honestly, if they have to keep it simple for the lay person who doesn't do STEM fields, they could just say that it has a "surprisingly beautiful sound for the price, have a listen for yourself", blah blah blah.
They say "no DAC" but the whole thing basically IS a digital to analogue converter. The switcher itself could be the size of a production version of a Raspberry Pi hat. If we're lucky, most of the additional bulk could be power factor correction, which requires a lot of energy storage so the mains doesn't get contaminated with low frequency noise. But I suspect that it's mostly thick chunks of brushed aluminium for no real reason.
Looking at the GaN FETs, they could indeed be interesting because of the lack of body diode that the usual MOSFETs are stuck with. So, inductive spikes from the speaker load could be managed in some interesting ways. But they never talk about this stuff! Honestly, if they have to keep it simple for the lay person who doesn't do STEM fields, they could just say that it has a "surprisingly beautiful sound for the price, have a listen for yourself", blah blah blah.
Since it most likely has similar/same DNA as the Peachtree GAN400 reviewed by ASR, you can make your own assessments here.
FWIW, for Class D GaN inspired designs, I’ll take Leo Ayzenshtat’s Orchard Audio Starkrimson over Peachtree fluff anyday. Plenty of AP-555 measurements over on his site. And my build that I did a few years back.
Best,
Anand.
FWIW, for Class D GaN inspired designs, I’ll take Leo Ayzenshtat’s Orchard Audio Starkrimson over Peachtree fluff anyday. Plenty of AP-555 measurements over on his site. And my build that I did a few years back.
Best,
Anand.
The GaN permits to use very low deadtime as low as 4ns is possible. This is what it makes very low distortion not the high switching frequency. On the contrary the frequency of the PWM should be lowest possible as 180khz to get very low distortion.
The problem with open loop is the PSRR. Although in bridged mode it is reduced considerably, nevertheless it needs rock stable noiseless PSU.
PCM to PWM can be delta-sigma.
For $1300 is very good priced compared to boards alone were sold about this price.
The problem with open loop is the PSRR. Although in bridged mode it is reduced considerably, nevertheless it needs rock stable noiseless PSU.
PCM to PWM can be delta-sigma.
For $1300 is very good priced compared to boards alone were sold about this price.
He, on YouTube means, not a standard classD with oscillator, comparator. It is defined as digital amplifier as no analog state occurs except on the output.
I would say that PCM to PWM conversion for class-d amplification is not that new. TacT and TI have products with that principle in their portfolio for more than two decades.
Regards
Charles
P.S.: And I don't like the term "digital amplifier", since there is no suchthing actually.
True but one may not forget the other point of view: As soon as you use feedback (not in this case here) you can take advantage of higher feedback factors with higher switching frequencies.
Regards
Charles
P.S.: And I don't like the term "digital amplifier", since there is no suchthing actually.
Most of the GaN amps out there currently underperform the state of the art class d which does not use GaN. The marketing departments will talk the advantages of GaN (higher switching speed, etc.) but then don't tell you their switching speeds are no higher than regular mosfet class d amps. The Orchard Audio GaN amp is the only one I have heard of that actually competes performance wise with the state of the art, the rest are not close. Bruno has discussed GaN and said basically that cheaper, existing mosfets are able to perform just fine. It's all in how they are implemented and to date, no one has really come up with a clever design that exploits their advantages.
- Status
- Not open for further replies.