Peachtree GaN 1 Beta

First, very rough draft.....way more to come.....please feel free to contact me about anything that might be added, helpful or incorrect.  This link is going to be sent to reviewers, manufacturers and posted on forums all over the world.....I am spreading the word!

http://tweakaudio.com/EVS-2/The_Audio_Revolution_has_begun.html

Regarding use of MiniDSP Flex for digital domain volume control, the specifications state that the SPDIF digital outputs have a sample rate of 96khz. Sounds like this would be a limiting factor going into the GAN 1 which can accept up to 192khz rates. Am I missing something?

@abd1 Would love to be able to listen to the Cubes.  I'm planning on using the Camerton Binom's for my active setup.   A GaN1 per side driving the Cam's and Purifi woofers.   I'll start with the MiniDSP SHD as the front end and look to see how it performs.

I had the Peachtee GaN400.  It paired nicely with my Rose 150B.   Had too many Class D amps (Purifi, nCore, Hypex) and recently sold it.   I'm using the Rose with a Class D mini GaN 5 and have no complaints.   

Have to say that the simplicity of the GaN1 is brilliant.  Sound wise it betters the GaN400.  Not saying it will eliminate DAC's in the world but I do like the KISS approach.

@ricevs Originally was expecting an active Ravenna system using the Cam/Purifis (from your old buddy at Mivera...).   Never panned out...so this may be a good replacement.   

Regarding use of MiniDSP Flex for digital domain volume control, the specifications state that the SPDIF digital outputs have a sample rate of 96khz. Sounds like this would be a limiting factor going into the GAN 1 which can accept up to 192khz rates. Am I missing something?

Great point.  This could be a limiter for the Flex.   Assuming my SHD has the same sample rate.  Need to check this out...

The Rose outputs 24bit/192KHz and sounded fantastic (Qobuz).  No volume degradation issues with Rose.

When you do not know that EVERYTHING makes a difference then you concentrate on the only things you can see/know. 24/192 might sound a little better than 24/96 into the amp.......but there are so many things that make WAY more difference than that. For instance, the digital cable from the coax jack to the amp board is totally ordinary with ordinary connectors on it. I would remove the input jack board and check underneath that board to see if there is a coupling cap and or 75 ohm resistor there. If there are parts there then replace those parts with better ones and then hardwire  a super sounding cable from that board directly to the amp board (no connectors on either end). If there are no parts underneath the input connector board then solder a super wire directly onto the pins of the input jack.....bypassing the board altogether. This will make much more difference than going between 96K and 192K. Of course, there are other things to try on the amp board.

The power supply to the Flex should be killer....and best if hardwired.

I have not heard the Gan 1 but I have a Gan400 and I am using a Schitt freya+ tube preamp with and it pairs well with the Magnepan LRS+ . I am using the Wiim pro streamer with a Allo nirvana power supply feeding the SMSL do 100 dac . All analog cables are connected via balanced cables . The soundstage is astonishing with depth and the resolution of this system is superb for the price point. I’m waiting for my magnariser stands to be delivered soon. 

Sounds like this would be a limiting factor going into the GAN 1 which can accept up to 192khz rates. Am I missing something?

Yes. Digital audio uses digital 'words' whose resolving power is determined by the number of bits in the word. IOW there is a chunk of data representing a voltage at a certain time. A string of these words makes up the waveform. In a class D amplifier, the on and off states have no such meaning. So a conversion from digital to a pulse stream that has wider and narrower pulses has to occur. FWIW Almost any GaN amp is really switching at 500KHz and beyond.

Ralph, When will your digital amp be available? We don’t need no stinkin DACs, preamps and normal feedback amps!....he he. And the fact that we can use digital xovers and equalitzaion without extra DACs and amps is incredible....way more pure and transparent. The revolution has begun!

So Ralph, are you saying that the data output (to be sent the the input of the Gan 1) sampled at 96kHz would not necessarily be a restrictive limiting factor to the ultimate performance of the Gan -1?

Funny how they say nothing in their marketing copy about the load dependent frequency response. So 1980's class d. Most well designed class d amps have solved that issue. Nice concept, poor design.

That’s because it doesn’t have a load dependent frequency response. It’s a great design. Funny how you don’t know anything about this amp.

@donnylovely 

That’s because it doesn’t have a load dependent frequency response.

 

Prove it.

You’re the one that is making the accusation, you prove it. Should be easy for you if you have any idea what you’re talking about. That is if you're not making baseless accusations. 

The proof is right there in the ad copy above- it's run open loop. Show me an open loop amp without a load dependent frequency response.

The ad copy? I’m not seeing what you posted. We’re talking about the gan 1, not your hypothetical open loop amp. You're y the only one saying anything about open loop. Sounds like you don’t have any proof. Just making baseless accusations based on some other random amps that aren’t the gan 1

@donnylovely

If you read the op’s copy and paste ad copy and understood what open loop means you would have your proof. Ball is in your court to show otherwise. I won't be holding my breath as it is clear you are technically in the dark on class d amps.

It says nothing about open loop. You still have shown 0 proof that the gan 1 has a load dependent frequency response. 

@donnylovely

You have proved my point. NO FEEDBACK = open loop. Now you have learned something. You are welcome. Your turn to show how an open loop amp can avoid load dependent frequency response. Good luck.

It says nothing about open loop. You still have shown 0 proof that the gan 1 has a load dependent frequency response. 

Kuribo will never produce because he/she is a troll.  He/She is the same over at Audio Science Review, in the same manner regarding the same things about the GaN 1 amp.

Must be a very lonely man/woman.

Pokémon is indeed very sad and lonely. Angry little creature. Has a crush on this Bruno guy.

You have it backwoods, Open loop = no feed back. Sun = hot, but just because something is hot doesn’t mean it’s the sun. Look inside the gan 1 and see, which you obviously haven’t done. If it was load dependent then why is it rated 200w at 4 and 8 ohms? Or is that what you are talking about? I thought that would mean it’s not load dependent, but maybe I have that backwards? Does dependant mean something else in the amplifier world?

What is so funny is that any load dependency really does not mean anything to anyone except those who look at numbers. If an amp is 1 db down at 20K with a 4 ohm load (so that means your tweeter has a 4 ohm load).....then your speaker will be down about one quarter db at 10K.......do you think you can hear this? Do you think you can hear 1 db down at 20K?

What is so cool about the pure digital amps is that you will ALL end up using them with a digital xover. This is part of the revolution. The digital amp gets rid of the preamp and DAC and analog interconnects and normal amp. But the revolution does not end there. The parts inside your speaker are messing up the sound. By using a pure digital xover you get rid of those parts, you can use a tweeter much lower in frequency to get more transparency by using high order slopes, you can time align, equalize, room correct......all without sonic loss. So, even if your zero feedback digital amp were driving a tweeter with a 2 ohm load and there is some loss there.....well, you just boost the highs in the crossover to compensate......end of story. This is the revolution.....digital amps and multiamping using a digital xover.......YOU ALL WANT THIS. This will give the best sound in the world. LESS IS MORE!!!!!

mbolek, You are the only person in the world who has modded the GaN 1......I think you hinted at changing the coax wire inside......please try this, it should make a big difference.....and eliminate the connectors on the wire, if you can....hardwiing is always best.....and all coax wires sound different (please re read the review of all the digital cables linked to on page 2).....Because one is made from silver does not mean it will give the best sound. Try different ones. I will be doing this when I get one in here to mod.

What is so funny is that any load dependency really does not mean anything to anyone except those who look at numbers

I 100% agree. You either like an amps sound or you don’t. Don’t see why this guy is sweating some bad measurement, which he made up himself. It is kinda fun seeing what BS he comes up with trying to explain the thing he hasn’t measured though. I’m keeping my Gan 1 long term no matter what measurements or lies people come up with.

It's clear that the meaning and implications of load dependent frequency response are over the heads of many here. It's of no use debating with those who don't even understand the issue at hand, yet argue any way. Pure ignorance.

And, lol, no, power at a given impedance is not the same as frequency response at varying impedance. But you do keep proving my point that many here have no understanding of the technical underpinnings of these amps yet argue on and on without a clue what they are arguing. Bravo.

Still waiting for an example of an open loop amp without a load dependent frequency response, or any proof whatsoever that this Peachtree GaN amp is such.

@ricevs I will send you my GAN1 amp after I check it out for a week or 2. I am expecting delivery of the GAN1 before the end of this year.

You can take your time with it since it will take 3 months from time of order of my speakers to delvery.

@kuribo Pure ignorance would be you complaining about some "issue" that no one cares about and you not even knowing if the gan 1 has the so called issue or not.

Pure ignorance, would be judging an amplifier before listening to it.

Don’t need to know anything about the technical aspect to enjoy its great sound. I guess you only need to know the technical aspects if you want to be a dick about it and rain on everyone’s parade.

Example, gan 1, where is your proof that it’s load dependent?

You must be delusional. You haven't given any proof about the gan 1. 

Who's delusional- the guy talking facts or the guy in denial who literally bought the koolaid, too dense to understand the technical explanation? Ever heard the word "bias" before?

I can see I am talking science to a zealot. I would have to be deluded to continue. I'm not wasting anymore time. Enjoy your 1980's amp.

@ricevs I did swap out the stock coax.   Added an AECO RCA and used the Illumati hook up wire.   Nice cheap mod and believe it is another improvement.  

Sad man obsessed with crapping on a thread for an amp he’s never heard calls someone else a zealot.   The irony….  Why does hifi attract such weirdos?

Back on topic, any ideas on what type of speakers work best with this amp?

How do you do the @ thang..?.....you know with their name afterwords

mbolek,

The dirt cheap iluminati cable was one of the better digital cables 25 years ago.  I cannot imagine that it sounds anywhere near as good as a serious cable today.  We have better conductors, better dialectric material, better standing....we can cryo, better damping, etc.  I make my own digital cable by hand using a twisted pair high purity Cardas 24 gauge bare copper wires that is then put into a cotton sleeve and I twist the the hot and ground together.  I listen to the wire off the spool for best sound direction.  This is the first wire I will try inside the amp when I get one in here.....I will try some other expensive wires too......including silver/gold combos.  I find the 75 ohm thing not nearly as important as other things I have tried.  Back in the late 90s I bought a Goldmund digital cable that was suppose to be baked in an oven so it would be super pure 75ohm.  I made a twisted pair cable out of Nimtec high purity copper in a teflon sleeve in 10 minutes that blew it away.  My cable was just a twisted pair....with no mind to whatever impedance it was.....lots going on here that people do not understand.

@wise Santa B why do you expect delivery next week? I am waiting too.......

Thanks for this thread folks I NJoy it! U R All WAAAY above my pay grade

@donnylovely   "...complaining about some "issue" that no one cares about"

Seriously? This is certainly something to care about, if you have any clue about amps.

Until these new wonders can run my Acoustats, or any difficult load, they're not worth a dime. Heaven forbid manufacturers post their 2-ohm ratings!

I dare anyone owning this amp to send it to ASR. Yes, I know they don't give a damn about sound quality. But they do give a damn about the flip side of the audio coin: measurements.

@kuribo the guy in denial, which would be you. You haven't given any facts. 

@art_boston I stand corrected, only 2 people care about. I only care about sound quality. I'll never send my personal stuff to those jokers at ASR. They can buy their own gan 1 and be smug about their measurements. 

@art_boston

Seriously? This is certainly something to care about, if you have any clue about amps.

He is clueless, don’t waste your time. The FACT is open loop (zero feedback) amps are load dependent - the output filter interacts with the load, that is one reason why modern designs use feedback after the output filter. Simple physics which can't be ignored, as much as one tries.

@singintheblues

Back on topic, any ideas on what type of speakers work best with this amp?

 

That’s the gist of the problem, isn’t it, with amps that have load dependency? For modern designs which avoid that issue it isn’t a thing. Try something with a flat impedance curve that is basically a resistive load. Good luck.

@donnylovely

Maybe this will help...

From:

https://sound-au.com/articles/pwm.htm

Class D Audio Amplifiers - Theory and Design

6 - Output Filter Design

The output filter is one of the most important parts of the circuit, as the overall efficiency, reliability and audio performance depends on it. As previously stated, a LC filter is the common approach, as it is (theoretically) lossless and has a -40dB/decade slope, allowing for a reasonable rejection of the carrier if the parameters of the filter and the switching frequency itself are properly designed.

The first thing to do is to design the transfer function for the filter. Usually, a Butterworth or similar frequency response is chosen, with a cutoff frequency slightly above the audio band (30-60KHz). Have in mind that one of the design parameters is the termination load, that is, the speaker impedance. Usually, a typical 4 or 8 ohm resistor is assumed, but that would produce variations in the measured frequency response in presence of different speakers. That must be compensated for by means of proper feedback network design. Some manufacturers simply leave it that way so the response is strongly dependent on the load. Surely a non-desirable thing.

7 -   Feedback

As I have stated previously, timing errors can lead to increased distortion and noise.  This cannot be skipped and the more precise it is kept, the better the design will perform.  Open loop Class-D amplifiers are not likely to satisfy demanding specifications, so negative feedback is almost mandatory.

And there you have it...

@dolfan It could be ignorance on my part. I saw the blurb on Peachtree web site that said they have a new batch being built for Dec 2022. I will give them a call to see if they can put some SpeakON speaker termination for me. That way I can do a real apples-to-apples comparison with my Benchmark gear.

I will use the GAN1 amp for my KEF Blade 2 Meta (ordering in Feb 2023) until I get funds for a more powerful amp for the Blades. I was wondering what to do with the GAN1 amp afterwards. I realized that the GAN1 has the protentional to be a great headphone amp for my RAAL SR1a earphones. I want to get a headphone amp for the bedroom and the GAN1 could be a perfect low box count solution. I bought some very short used Audience AU24 SX speaker cables a few minutes ago for that purpose.

As a frame of reference, the best 2 channel gear I had on the RAAL SR1a earphones cost over $20K and I sold all of them since it was not as good as my RAAL VM-1a amp. The VM-1a only works on the SR1a and not speakers so it has some advantages in design.

So, I now have 2 comparisons to do with the GAN1.

1) Benchmark stack vs GAN1. I will have to downgrade my speaker cable for this comparison to Audience Conducter SE unless Peachtree can do SpeakON termination for me.

2) CODA 07x preamp | KRELL DUO175XD amp | Musetec 005 DAC | Audience FrontRow speaker cable vs GAN1. I still remember how this sounded since I listened for hours each night for over a year. I will use the just purchased Audience AU24 SX speaker cable which is one notch below the FrontRow so it should be a close approximation.

@wise Santa B

Hoping we both get ours next week. They told me last week I would hear sumn about testing and or shipping but nothing so far.

u have nice systems

Mine will be Gan 1, Node, Tekton Perfect Set, and Rythmik FV15HP. It's a TV and backround music set up.

I need a coax cable. Where would u start?

BTW, the GaN 1 digital board measures almost the same frequency response as the board in the GaN 400 (analog input board). It is one db down at 20K using a 4 ohm load. This will give less than one quarter db attenuation at 10K......no one can hear this. However, with a 2 ohm load it is 2db down at 20K.....so it will make 10K down one half db.....maybe you could hear this. OMG this is so horrible!!!! (Kuribo speaking here). There is maybe one person in 100 with a speaker that MAYBE you could hear the rolloff.......I don’t hear much above 10K myself. There are a couple of electrotats that have very low impedance at 20K and there are a handfull of regular speakers that have a 2 ohm tweeter load.......but very, very few. Just go to stereophile mag and read the measurements of various spekakers they have reviewed lately....only the Kef Blade 2 goes a little under 4 ohms at 10k.....all the rest are above 4 ohms......just not an issure.

So, thinking you need feedback and op amps and DACs and Preamps and analog cables so you do not have this very slight problem that practically no one will hear is simply silly. But, some people make mountains out of mole hills.....hey, I get excited for positive things.....some people get excited to put things down.

I like UP myself. I say Yes to life.....and Blessings to all......how about BLESSESS......Beautifully Love Every Single Soul Every Single Second

Speaker loads are rarely purely resistive nor are they flat.

@ricevs I figured the the Illumati would be better than stock. Maybe, maybe not. It is a damn skinny wire. Added a couple of pics to my profile.

I may try using some Triple C copper as the interconnect.

Kinda cool that a $1500 box can have so much potential. It is dead quiet, no heat and sounds better than my reference Purifi. I stopped trying to justify Class A many years ago....

@yyzsantabarbara If Andrew can’t add speakons for you, have Ric swap out. Not more than $10...if you choose not to decide, you still have made a choice.

It probably is much better than stock......and you do incredible soldering and work.....looks better than my stuff.  Very professional.

@dolfan I do not know much about digital cables, but I do know when some companies do cables right. The 2 companies I trust are Audience and WyWires. I have not spent too much time trying other cables and have just stuck with these 2 because they work for me.

Saying all of this I just went to the WyWire web site and ordered the Silver cable. These cables are 40% off the price you see on the web site.

S/PDIF Digital Audio Cable Litespd – WyWires

I asked Alex to make me a BNC to RCA wire to eliminate an adapter. Hopefully, that was a good choice.

Has anyone tried connecting a turntable using the 3.5mm input on the Bluesound Node and played it on the amp? How does it sound comparing to your pre/power amp? 

 

Has anyone tried connecting a turntable using the 3.5mm input on the Bluesound Node and played it on the amp? How does it sound comparing to your pre/power amp? 

So far, the only one I know that has done this is Andrew Clark, president of Peachtree Audio.  Here are his remarks copied and pasted from Peachtree's website:

A few weeks after exhausting my library of demo tracks and personal favorites, I hooked up a turntable to the NODE. (My turntable has a switchable built-in phono preamp so I can connect to the NODE's line in.) I don't know what I was expecting, but I was pleasantly surprised by what I wasn't hearing. It sounded crisp and clear without any background noise or humming. This is great. I'm not hearing anything that makes me believe this input was an after thought or compromised because its a dual input (analog or optical digital audio) or because it uses an ADC inside the NODE. I have ZERO ISSUES recommending that people hook up their turntable to the NODE and enjoy their vinyl on this system this way! It sounds great! I even programmed my turntable as a preset, so when I want to spin vinyl, I just touch the first dot on top of the NODE and lower the tonearm and I'm listening. Easy.

I might consider adding a phono preamp to boost my turntable output a bit. (My turntable's built-in phono preamp has no gain adjustment.) Albums are mastered at much lower levels than today's digital files so when I want to crank up the volume on my vinyl I can sometimes max out the NODE's volume and still be left wanting more. Some of my older CD files (ripped to my computer years ago) have a similar issue. They were mastered at levels significantly lower than today's norms so even when the NODE is set to maximum output, I sometimes could use more. Fortunately most of the earlier recordings have been remastered at some point in the last 20 years, so that helps with the gain/level issue. If my primary listening was vinyl or digital files mastered before say 1990-2000, I would definitely use a phono preamp with gain settings (or higher max ouput than my turntable's built-in preamp) and re-mastered digital files. Or perhaps some other way to affect gain in the digital domain like Roon, Foobar, etc.

Hope this is helpful!

Thanks kitsap2, Since so many people in this forum like vinyl just wondering have anyone of you had tried it and what's the result.

@mbolek those are some nice looking mods. Much higher quality parts it looks like. Have you noticed a significant change in sound quality?