 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
75.197.122.12
In Reply to: RE: You might posted by E-Stat on June 25, 2025 at 12:41:33
You ask Ralph. It's just another opinion with no curves which makes it even more Irrelevant to my point about GaN.
Follow Ups:
A nice advantage of class D is low noise if all your ducks are in a row.
GaNFETs have no leads- to connect to them they have solder pads. This allows the designer to minimize stray inductance in the design, which results in lower radiated noise associated with switching (caused by parasitic inductances).
We got our amps so quiet in that regard that most tube amps radiate more noise on the AC line and through the air. This is important since switching noise can mess with digital equipment, tuners and the like.
As anyone who runs horns knows, the noise floor (especially with tubes) is a constant battle. Tube amps, even SETs designed with low gain for horn applications, just are not that quiet.
The Class Ds are obviously lower noise and does not take a back seat on the horns (vs SETs or the like) in terms of sound quality. We've been making OTLs that have gotten many awards and accolades in the high end press for decades now; I think the class Ds sound better in every way.
So win win.
What you posted here is true and accurate.
I agree with all you said. I used to design around all those parsitics.
Most modern TI 3255 chip amps (from China) get a THD+N of around 120 db, which is 1,000 times better than the "standard" 90 db S/N spec of power amps using old technology we all came to accept.
I was the first AES member in my State to own a CD player, which was 30 db better than vinyl, even though producers didn't quite yet know what to do with that 4Hz-20Khz. bandwidth and 96 db S/N ratio yet.
I've own Hypex NC-400, UNHEARD at the time, for over 10 years. Now class D has gotten real cheap with the chip amps on horns to high efficiency nerds like us can now afford bigger horns on the amp savings alone! LOL.
a CD player, which was 30 db better than vinyl, even though producers didn't quite yet know what to do with that 4Hz-20Khz. bandwidth and 96 db S/N ratio yet.
As an LP mastering engineer, I learned a few things about LPs causing my prior misconceptions to die an ugly death.
Most LP noise floor is created by the pressing plant. If the pressing machines are properly damped during the cooling process, surface noise is vastly reduced, by about 20dB. QRP in Salinas KS is doing this right now and has for years.
When you set up the cutter head, reducing groove noise is the goal. If all goes well (proper head angle, stylus temperature and so on, assume a lacquer is used for the cut) the resulting groove is so quiet that the playback electronics are the noise floor.
On this account we have the ability to cut LPs with noise floors better than 80dB.
Bandwidth: most LPs have bandwidth past 50KHz. My Westerex system introduced a 6dB rolloff at 42KHz, rendering the system flat at that frequency. I was able to cut 30KHz tones in the leadout groove area and play them back with no worries.
Distortion: Most cutter heads have 30dB of feedback wrapped around them, which does not include the feedback of the cutter amps themselves. The resulting cut is very low in distortion.
IOW, 98% or so of the 'distortion of the LP' occurs in playback, usually due to poor setup and inadequate playback mechanisms, such as poor tonearms. That is the real weakness of the LP format.
I don't contest that digital is excellent. All I'm saying here is that if you want to denigrate the LP you should be armed with facts rather than fiction; it will make your argument stronger.
Accusing me of "fiction," as owner of 1,000 LP's of quantity of noise from high to low, they were all made from 80 db Master tapes on a good day as the source. Also 2,000 cd's and 3,000 DVD's and 500 Blue Rays, including the latest 4K "wow" ones from Hollywood. Dost thou presume too much?60 db S/N ratio was all on could expect from a non linear RIAA curve that made Microgroove records workable even on the crappiest most common systems.
Again this discussion has extrapolated with EMOTIONAL defenses and accusations way beyond the scope of the original text.
I don't care what type of transistor you use in your output devices. The JC-1 amplifier (whose name appeared here as "name dropping" uses many devices in parallel to get output that could be used for arc welding or jump starting a locomotive with the typical "audiophool" tweeter wires of 4 AWG, promoted by snake oil salesmen.
I measure 124 db output in my sweet spot with a pair of 2-way Jubilees in my living room. I was at the Klipsch Factory in Hope for a demo with their Chief Engineer and co-inventor, Roy Delgado.
He was using Professional class D amplifiers (QSC and Crown)even back then, and no one complained about noise he was also using a cheap, $50 cd player as music source.
From the sublime to the ridiculous, all comments are welcome here, relevant or not.
Edits: 06/26/25
It seems to me you are over-reacting.
You are correct with tape, a noise floor much past 60dB is quite difficult. When we were running our mastering operation most of our sources were digital files rather than tape.
Tape was used as a master simply because it was easily re-recorded. LPs have vastly superior specs to tape. You really only get that when direct to disc or using digital source file.
60 db S/N ratio was all on could expect from a non linear RIAA curve that made Microgroove records workable even on the crappiest most common systems.
While this is commonly true, one must be careful to not conflate anecdote with the actual medium. Its a common mistake. I recommend you do what I did; get yourself an LP mastering system and work with it. You'll find what I said is correct. Like I said, I had misconceptions about the LP (poor noise floor and so on) that died an ugly death.
don't have to be so abrasive.
Curves?
You really don't have to be sensitive to someone who's posting the truth of device behavior.
Science Trumps Emotion.
I ask a simple question and your response consists only of posturing.
Let's try again.
To what "curves" do you refer?
I posted a curve of a simulation of the most demanding of POWER amplifiers.
In this case it was a bass horn, but subwoofer horns are more demanding of power.
If you can reach 125 decibels with less than 30 Watts, I'd say that pretty much negates the use of GaN transistors.
The likes of which were presented as an OPINION with Zero data, so I responded in like kind, with a contrary opinion of them being too expensive and overkill.
What more do you want from me at this point? You have clearly made yours.
If you can reach 125 decibels with less than 30 Watts, I'd say that pretty much negates the use of GaN transistors.
As I mentioned earlier, a nice advantage of GaNFETs is lower noise. In the statement above a fairly high efficiency speaker is suggested; keeping the noise floor down is a pretty big deal.
Yes it is, but I have had dead quiet for over a decade with modified Hypex NC-400's and now TI 3255 based chip amps.
How much expensive overkill beyond what is already dead quiet do we actually NEED?
I just got rid of my 44" Epson printer and even 1080P resolution files can make a great looking 12x18 print. Wanna buy a 61 Megapixel Sony which is also expensive overkill?
Also, they are 400 Watts, not 100 Watts and don't cost $6,000 you claim to be "not that expensive."
Like the man from Vermont was once asked: "How's your wife."
His reply: "Compared to what?"
How much expensive overkill beyond what is already dead quiet do we actually NEED?
GaNFETs just are not that pricy! Keeping in mind that most MOSFET class D amps need a heatsink of some sort, the fact that GaNFET heatsinks are not as expensive also plays a role.
As a result most tube amps people might put on horns are more expensive than our class D GaNFET amps.
We didn't do our class D amps for low noise though. We did it for sound quality; low noise was a nice side benefit. Its easy to hear how they sound better than tube amps on horn speakers.
I'm of the opinion at this point that tube power amps are on borrowed time as class D continues to make inroads, especially in the musical instrument amplifier (guitar) market.
I agree about tubes, which are like fixed tone controls in comparison to linear behavior of modern class D.
I have no issue with your choice of output devices. I also designed a heatsink before, years ago, so not one is going to "instruct" me on thermal behavior of devices. My HYPEX NC-400's don't have heatsinks, and don't use GaN transistors.
If someone else here wants to do further name dropping, I have Carver Mead's Cell phone number, since I worked with him 26 years ago. His knowledge goes way beyond all this Audio stuff, and is responsible for much of modern electronics technology.
So now that you admit you have a selling interest in an Amp whose devices you claim are not that expensive, $30, that means $300 for each device when marked up to the consumer level inside a product. Do you use more than oner per amp? LOL.
I measured 124 db output in my living room in my Jubes with $40 TI chip amps and a 19 volt, $10 Laptop power supply, 6 years ago. That was good enough for me.
Good luck with your sale.
You're off by a factor of 3 or so, at least for the part we use. Nor do we mark up that much!
So its not overkill on a horn system. Our class D amp is rated 100 Watts into 8 Ohms, on my speakers which are 16 Ohms they make 50 Watts, which is plenty when the speakers are 98dB. Not overkill. They replaced a set of our fully optioned M-60 OTL amplifiers, which have gotten nice reviews and awards in the high end press. I think the class Ds sound better and they cost less too.
Well then. Are you a USA company? send a link to my email.
OK, so I had to dig down into the site to find this "Wonder" Class D amp, buried among the inferior valve offerings, by your own admission.
This site avoids commercial adverts, which is why I asked for a link via Email. Thank you for that.
However, I had to click onto a review to find that the amps in question are 100 Watts per monobloc at a cost of $6,000 vs. my 10 year old Hypex NC-400's that have 4 times the Power at 1/4 the price for two monoblocs.
So, assuming there is Zero Audible difference (no proof or curves required, it's subjective, like 99% of EMOTIONAL Testing in Audio) Your "don't cost much" statement would receive a Paul W. Klipsch BULLSHIT button if he were still alive to assign one to that post!!!
Bluster is never a way to win a debate.
What you didn't read were the reviews. Perhaps anecdotal evidence like that means nothing to you, but FWIW the comments we get from reviews are strikingly similar to those from our customers, which can be seen on our feedback page. When lots of people say the same thing over a period of years, those anecdotes take on more weight.
Yes, its more expensive than your old Hypex, which is an excellent design.
What we've found is that with a lot of Hypex and other designs from the godfather of class D, Bruno Putzeys, is that things like the input buffer and power supply have an enormous impact on the way the circuit will perform. The Hypex modules were sold as modules only so someone had to buy them and put them in a box to make them work. If they skimped on anything the resulting amp would not perform as well as it could.
I think that is an enormous part of why class D has not had good acceptance in high end audio. I've met plenty of people who have said they will never try a class D amp again. Its not the fault of class D or the module; just poor execution which prevented the module strutting its stuff.
So yes, our stuff is more expensive (still a lot cheaper than the tube amps to which it is superior) but you have a nice chassis that survives shipping with ease, a properly built and protected power supply and the input buffer integrated into the module; all easily serviced and updatable.
If the cost is all that is stopping you that's no worries on my part. I've noticed that the best horn systems (of which IMO the Klipsch are not an example) aren't cheap either.
Bluster has to do with the spoken word, in a loud manner, which is not possible with text, unless it's printed with all caps, which I use for EMPHASIS to drive my point. No actual shouting involved.
"Anecdotal evidence" (translated is just one persons OPINION). It is the antithesis of widespread FACTS, which is all I stated with Personal Experience, measuring cost and SPL of actual OWNERSHIP and via financial commitment. If my universal point of "Large, Expensive Horns" only needing good little chip amps that are CHEAP and good enough to get to 124 db output, for less that $100 total cost is lost on a "High Efficiency" group as this, then y'all are in the wrong forum. For those who read with their neighbor's room temp IQ at play, while ignoring the actual WORDS that I typed, this is my anecdotal evidence and my opinion.
I'm sure that the product you designed is a great performer, no issue with that at all, even though I have not heard or measured it, you would not have released it otherwise. But don't tell me how "low cost" it is using overkill technology and claiming such, which at that point PWK will rise from the dead and give your his famous yellow button. It's all relative to PRICE/PERFOMANCE ratio. You'll be more successful at AXPONA nest year where, the rich and technically ignorant buyers, who pay $1,000 a foot for speaker cables will be your willing victim.
All of your pointed "parasitics" are minimized with micro length interconnects on Silicon. No one can BS me about what those are because I was a world renown PWB designer in my other life, where moving electrons at 1 gigahurts (pun intended) meant your were limited to an interconnect of 9 inches or less with damping resistors and parallel capacitors on the inputs as filter. I also had experience with Ferrite Beads to tame the ringing of FAST logic from Farichild Shottky 5 volt swings. Thank God those are part of the anals of history now unknown to young engineers of today. Also, no one can BS me about the performance of copper, aluminum heatsinks since I have a product I designed 40 years ago that still works, before Surface Mount Technology took over. And yes I wrote the very first Article for the SMTA Newsletter as an early adopter, and wrote portions of the early SMT discrete land patterns standards for the IPC.
I still build big horns, listen to Music, movies on 7.4.2 Atmos and use ALL chip amps to do it. But it did not happen in one day.
People do bluster on keyboards all the time, as you have shown. All caps, just so you know, is considered shouting; to emphasize a word, on this site you can use html to provide emphasis, or bracket the word **with asterisks**.
Mr. Klipsch will not be coming back or doing anything , regardless of what I do or don't do, so no yellow button. In the meantime, the amps with which our class D effectively competes are usually a multiple of the cost of the class D.
There are cheaper class D amps out there of course. Each with its own means of a corner cut. If you want a decent chassis and power supply for example, that costs something. We found that if you use a switching supply, it had better well be designed for the application else it will hinder performance. Using such a thing would allow us to make the amp much less expensive, but custom built SMPS are not practical unless you can order in 1000+ quantities. So we use a toroid.
"the amps with which our class D effectively competes are usually a multiple of the cost of the class D."
Therein likes the truth. Apples to Oranges.
I was comparing class D chip amps and my 10 year old Hypex NC-400's
Class D chip amps and Class D Discrete. You clearly stated that your ONLY class D was far superior to the best of your OTL Valves. No argument there.
My ONLY point was that it's not necessary to use GaN trasistors to get great perfomance on all the horns I built and/or owned in the last 10 years. Any bass horn will require more power than treble horns, which I used Jubilees in my living room with Chip Amps to prove my point.
Everything else just got out of hand and I'm done arguing with people that are don't contribute anything (not you), or have products to sell (you).
Over and out. Have a nice day.
You are absolutely correct.
We don't use GaNFETs so our amps can work on horns or any speaker for that matter. You can use MOSFETs and they will work perfectly fine.
My position is if price of the output devices is the issue, GaNFETs are not just not pricey enough to be a concern in the design. They simply aren't that expensive.
We use them because we can get less switching noise. The speed of them is irrelevant- although they are a lot faster, they need a bit of an inductive kick to get them to shut off. So the deadtime considerations are really similar to MOSFETs. So its really mostly about the radiated noise. The On resistance is lower too, so they don't need nearly the heatsinks that MOSFETs do. But if you're not pushing the amp hard (as in your case) that's not a concern. In our amp how that works out is you can run the amp at full power into four Ohms all day long and its no worries for the amp; it just sits there and within seconds after reducing the power the output devices are stone cold.
I guess if was my Automotive Electronics background rearing it's head.
I that market of 300,000 modules/model year, if you save $1 per assembly they almost raise as statue in your honor.
Not being an Audio Amplifier Manufacturer, I can only know the consumer level Price and NOT the development and productions costs of non-automotive volumes.
You have been relatively accurate in your comments here. No problemo.
Besides, there are no absolutes, Everything is relative, especially in audio and music reproduction!
"or have products to sell"Yup. Tom Danley (here known as tomservo) immediately comes to mind. He never misses an opportunity to promote himself.
*********
We are inclusive and diverse, but dissent will not be tolerated.
Edits: 06/30/25
This is where I think you are wrong about Tom, especially without long term evidence, like have, and everyone else here.
He's "just the facts" kind of guy, and has helped tremendously in helping educate us all in the complexities of horns and sound reproduction, as has Bill Fitzmaurice, and a few other participants here.
He has never promoted himself, I have read pretty much all of his posts to constantly upgrade my quest for further education. All he he does is SHARE real information based on his experiences with horns and drivers and none of it has EVER been promotional, other than his business partner named the company after him (an honorable thing to do). I assure you that they are both high integrity individuals.
That is a weak accusation that is beneath you, so I will assume this is a "glitch/faux pas" on your part indicative of anything further.
Peace, Love, Beatles
Yeah...exactly like Ralph does. And Tom does occasionally allude to the fact that he is in "the business" just as Ralph does. Just go look at the thread at the top of "Speakers" here right now. The only reason I have ever heard of the SH 50 speaker Tom builds is because he brings it up in his response to the OP. And you know what? I'm perfectly fine with it, just like I am with Ralph's oblique references because it is obvious to me that in both cases the postings of these two are maximally well intentioned and helpful and minimally inclined to self promotion. Would that all of the participants in AA were of the quality and as informative as these two guys are.
Maximally well intentioned, indeed. Good post.
I posted a curve of a simulation of the most demanding of POWER amplifiers.
Repeat request: kindly reference that. Not sure what you're talking about.
The likes of which were presented as an OPINION with Zero data,
Ralph can most assuredly provide you with data and more importantly, strategy with his approach which is different than others.
Why so abrasive?
I'm just hard to convince of theoretical applications of overkill devices.
No one got mad at Paul Klipsch for passing out Bullshit buttons, so why does everyone jump on my case and call me names for doing the same?
Water of a duck's back.
back to the question advanced a couple of posts ago. Quelle surprise!
Kindly reference your notion of "curves". WTF is that?
LMG3100R017
ACTIVE
100V 1.7mΩ GaN FET with integrated driver
Look at all the curves and see if it can apply to class D amp cheaper than mosfet.
.
Have Fun and Enjoy the Music
"Still Working the Problem"
This thread makes me smile. :)
I don't know anything about FETs, or diodes, or capacitors, or any of that stuff.
But my buddy John Curl can, and has, give me a 15 minute lecture on the properties of FETs. I'm tellin' ya what, there's more to 'em than just a pretty case. That's all I know, or remember.
:)
*********
We are inclusive and diverse, but dissent will not be tolerated.
Look at all the curves
of exactly what and where ? WTF does that mean?
It's a no brainer as to what's cheap.
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
