OT modern audio amplifiers

[darryl]

Way OT, but a few of you are still involved in electronic design. Wondering what is hot and new in audio amplifier design- I've built many, but now I want to eliminate the main power transformer if I can by using rectified ac to start with. The option still exists from a single phase supply to create the plus and minus voltages from a bridge rectifier, so for an amplifier I'd have the split supply as usual, but at the higher voltage of +- 160vdc.

Isolating the input wouldn't be that much of a problem, so the problems that do exist are the danger from the output wires being direct connected to ac power, and the requirement for higher voltage transistors for the outputs. Matching the speaker impedance is another issue, but one which can be resolved by making series connections instead of parallel to what would have to be multiple speakers per cabinet.

What I have in mind is to have the amp inside the speaker cabinet, or in other words, powered speakers. I have no problem to ground all the baskets, but there isn't a lot of space between the voice coil and the magnetic structure. The potential for arcing is considerable.

I'm not sure that I'm willing to rely on an optocoupler to pass the input signal and isolate any potential for an accidental connection from the input to a live wire.

I also know that by using a high frequency transformer I can get both the isolation and suitable power supply voltages to run an amp of whatever wattage I design for. Then I wonder why not set this up to deliver a voltage that tracks the output voltage and stays say 10 volts above the requirement at all times. I think this is a derivative of class G, or is it H- don't recall.

Any thoughts on this from anyone?

[beanbag]

Well, you no longer have the transformer's parasitic capacitance and inductance to absorb high frequency AC line noise and spikes.

Also, aren't you now charging your supplies at 60 Hz instead of 120 Hz?

You aren't supposed to connect neutral to ground, so what is it connected to now?

As for the class G or H or whatever, you need to have enough slew rate to "prepare for" a musical transient, yet have enough high frequency PSRR on the regulator / output stage to avoid distortion. IF you can't give the power supply audio signals in advance, it's kinda harder. I haven't looked into this issue much, but that's one of the problems I can think of.

I don't see much point in wasting a lot of power having overly high supply voltages.

Edit: please show a diagram of the "power supply" so I have an idea where the currents flow

[Weston Bye]

...What I have in mind is to have the amp inside the speaker cabinet, or in other words, powered speakers...

I'm not up on what's hot and new in audio systems, but I will apply some old-school reasoning:

A drawback of this arrangement is the audio lines from the preamp or other signal source become overly long, and unless well shielded are more susceptible to induced crosstalk, 60Hz hum, transients, high frequency attenuation, etc.

[topct]

What's hot is tubes.

It's amazing what as little as five or ten watts can sound like coming out of a well designed tube amp.

[Evan]

It's amazing what as little as five or ten watts can sound like coming out of a well designed tube amp.

That is mainly because tubes clip gradually and gracefully. One of my first really well paying jobs was fixing amps back in the heyday of Macintosh and Marantz. Transistor amps were commonly DC coupled and when a transistor blew it would take out the entire string. Bipolar transistors clip hard when they saturate causing a ragged edge to the high rise time signals such as percussion. Tubes just get lazy as they run low on electrons.

The audio business is chock full of superstition and total BS. People claim to be able to hear the difference between recordings made with differing amounts of humidity in the air and other such garbage. Massive silver wires to the loudspeakers is another such hoax.

I recall back then that somebody made a demo recording (on vinyl of course) that had been severely truncated to no more than 10 khz frequencies. This was taken to a big audio show and used to demo new equipment to the audio cognoscenti. The equipment receive wide acclaim for it's crystal clear highs and it's perfect rendering of cymbal crashes, snares and other such sounds.

When it was revealed after the show that the recording had been kneecapped there was a great deal of huffing and puffing and accusing and excuse making by all and sundry.

I thought it was funny as hell because if I played the recording as source material into any system it not only sounded very good but the distortion analyzer revealed why. There was a lot less distortion in the output because the workload of the amp was greatly reduced. All amps introduce distortion not the least of which is phase distortion dependent on frequency. It cannot be avoided.

[Circlip]

"What I have in mind is to have the amp inside the speaker cabinet, or in other words, powered speakers. "

So you've only just discovered what B&O have been doing for the last more than a decade??

Regards Ian.

[airsmith282]

That is mainly because tubes clip gradually and gracefully. One of my first really well paying jobs was fixing amps back in the heyday of Macintosh and Marantz. Transistor amps were commonly DC coupled and when a transistor blew it would take out the entire string. Bipolar transistors clip hard when they saturate causing a ragged edge to the high rise time signals such as percussion. Tubes just get lazy as they run low on electrons.

The audio business is chock full of superstition and total BS. People claim to be able to hear the difference between recordings made with differing amounts of humidity in the air and other such garbage. Massive silver wires to the loudspeakers is another such hoax.

I recall back then that somebody made a demo recording (on vinyl of course) that had been severely truncated to no more than 10 khz frequencies. This was taken to a big audio show and used to demo new equipment to the audio cognoscenti. The equipment receive wide acclaim for it's crystal clear highs and it's perfect rendering of cymbal crashes, snares and other such sounds.

When it was revealed after the show that the recording had been kneecapped there was a great deal of huffing and puffing and accusing and excuse making by all and sundry.

I thought it was funny as hell because if I played the recording as source material into any system it not only sounded very good but the distortion analyzer revealed why. There was a lot less distortion in the output because the workload of the amp was greatly reduced. All amps introduce distortion not the least of which is phase distortion dependent on frequency. It cannot be avoided.

have you ever hooked up larger wires to a set of speakers and then thinner ones and then checked the respones on a scope to check the different in destortion levels at same volumes leves , a scope although only as accurate as its made will show it if there is a difference, as for people being able to hear the difference , weather you think it BS or not there are people that can hear the differnece in the changing of different weather conditions..

just about anyone can hear the difference in the same sets with only one change being done paper cone main speaker to a plastic cone everything else the same the only change being the cone it self..

speakers have about 50% of the quaility of sound, that comes from an audio system, another large factor in the over all sound quaility is the current that comes out of the amp to the speakers, NAD made a 35 watt per channle amp that sounded awesome, i had such and amp the were made on low watts but hi current out put, as was harmon cardon, another amp i had it was 50 watts per channel the harmon was very nice and delived very nice base response on the low and hi end the NAD was only nice on the lower end of the scale, i have also owned an apine system as well it was nice on hi's but lacked good base , but it was loud , speakers like EPI where made as a monitor speaker and were not very good at hi volumes but on the lower side sounded nice, ESMs were nice as well low end but not hi , cewern veage sucked all the power you could feed them they sucked over all but they could take a beating, my AR's were very nice on low to mid range applications as were energy's..

speakers do make a large difference even when used a on poor end amp like kenwoods, the sony amps were pretty good , sansui was all about good low to mid but on the hi end clipped like crazy. harmon cardon amps very good low to mids, the marantz amps were great as well..

hi watt low current amps are junk, low watt hi current amps were the better way, how soon you can get the speaker to its starting watts or power draw as some say is very importante to its over all sound quaiilty that it can produce, same as a good set of speakers helps also as well.

bigger is not better in the amp world..

take 10 brands of different amps and change lets say the Energy monitor speakers these were very popular in recording studios, now any moron can hear the difference in the sound quaility from one amp to the next with all volume and tone controls set the same for each amp.

destoration as we call it or crappy sound will start to kick in at different times to , if your speakers are drawing more then your amp can give them them THD kicks in rather fast, now speaker wire guage can also have alot to do with this to not only for the destortion kick in but also for sound qauility, iif you are starving the speakers using to thin of a wire you will get bad results so a thicker wire say 3 times thicker will deliver better sound and a more even , why is this well its becuase you are now allowing a larger area for the current to flow so its not being choked out. hard to belive or get your head around but its true in speaker wire bigger is better expcialy on the hi end your drawing more on higher volumes so you need the thicker wire other wise you can cause your amp to over heat and things will start to blow, back in the day my self and a few others were extreamest in the auido world and we went threw alot of equipment before we were happy with our stuff, even the speaker cabnet dicates sound responses but the larger the speaker the more of a hi current producer amp you needed, watts does not amount to mcuh if you dont have the current to push thoes watts also you needed good speaker to handel the hi current and the large gauge wires,,, so yes you can hear the differences, one set of speakers to the next or even changing the wire guage or perhaps you just need a low watt hi current amp with good speakers ,.

driving alot of bass into speakers is ok at low levels but not good at higher levels , the bass control draws alot of voltage or amps from the system and when your speakers are also drawing or wanting to draw alot more then you have available now you will get distortaion sooner can over heat and cook your amp pretty fast to , and so on,,

some guy want good bass good high's but dont care much for the mids ,, so a decent 2 way speaker is a good idea but they also want it at low levels so you turn up the base advance the treble a bit, use a set of speaker that have a port hole or bass reflex hole in them this create bass at lower volumes alone so when adding more bass at low volumes is good and really rockes the house, at a higer volume you dont want bass reflex speaker, how come well you then over push the speakers and fry them, they are allow to move in and out to easy on bass reflex enclosers, to they are not ment for hi volumes , a sealed encloser wiht a hier watt and larger diameter speaker are better suited for hi volumes and can be push harder as well,, by allowing the speaker to much freedom the cone pops and the speaker blows and it does not take much to do this,

bass reflex systems offer better qauility sound on the lower end , where as sealed systems can not offer this so easy,
paper cone systems are good but the poly cones do offer more better rounded sound over all. the cone ring this is the ring that holds the cone to the speaker fram makes a difference in that the under bubble ring offeres better control over the above bubble style ring, foam rings are of lower gread in some cases over the paper holding sytem , so many things control , when an amp starts to sound bad, the good news is most amps will start clipping at just after 1 oclock on the dial if all settings are set at 12 oclock bass treble and mids, add more bass you clip your amp sooner heat builds up faster and so on..

weather conditions can and do alter the sound of your system , just as an eq and no eq do , and yes you can hear it , improving your sound can be just a change in speakers or getting rid of the EQ in the present speakers and same goes the oter way and eq can also enhance your quaility of sound as well. amps with low watt hi current out put are the best , hi watts low current amps not a good choice,,

my 50 amp per channel harmon cardon hi current amp was better and louder and cleaner then my 1500 wattt per channel sansui amp even on the set set of speakers the harmon cardon was better., also matching speakers properly to the amp you are using is also a good idea. also the ohms the speakers run in must be able to be used on said amp to as well as using proper sized speaker cable for amp and speaker confiquration...

tube amp or moderen amp really makes not alot of difference..

can you hear a change in sound on your sytem with differnt weather , answer is yes if your not tone deaf..
bigger speaker wires do make a good difference to big is not good to small is also not good...quaility of your speakers tweeters mids if you have them and so on all play factors in how things sound but 50% is still in the speakers for sound quaility..

good speakers right size cables and a good low watt hi current amp to get thoes speakers to fire up makes all the difference, as does the ohms the speakers run on and weather your running bass reflex enclosers or non bass reflex enclosers, size of the cabnets and how the curtians are done in side insulaton or no insulation , so many factors affect it all ..

[J Tiers]

EVERYONE is doing the amp in the speaker thing now, it's hot in the pro audio biz, and has been.

The tracking supply is patented, several times over in different versions, I believe... Yamaha did it so long ago that the patent is probably nearly run out, if not gone. A couple of the very very high end audio companies have done that, either in "analog" or in a switched version. Yamaha was switched.

The direct connection to the mains was tried..... At least one company was at NAMM many years ago with one. They were there ONE time, and never seen again.
You cannot do that in any practical way. You found one reason, insulation of the speaker is insufficient. And making the speaker cone out of 100% UL94V0 materials is impractical, as is sufficient prevention of access. It would be far cheaper to isolate the supply.

The hot item is class-D. Inherently lower distortion is one good reason. The distortion means are most all low-level mechanisms, as well as those due to switching speed and "dead time". The air is thick with patents concerning the matter.

Very high efficiency is another good reason. And if you can get "green" efficiency WITH very good distortion, it's a win.

As for tubes, Evan is reasonably close to the truth about the distortion differences, but there is a LOT more past the "conventional wisdom" in that matter. I happen to know rather a lot about that, and have studied it over a long time period, long enough that my* patent ran out several years ago.

It isn't just "graceful clipping" or "running out of electrons"......... it is exactly what you MEAN by that..... the character of the harmonics produced, and the rate at which they increase, etc, etc, etc. And the great inherent difference between tubes and solid state.

*it obviously belonged to my then employer

[BobWarfield]

If one amp in speaker is good, several are better. Do the crossover in the digital domain, send the bits to high quality D/A at the speaker, one amp per driver. Meridien was doing this years ago and it sounded great. Though that is ultimately the path I took on any of my own systems.

I'm a big believer in blind testing no matter what the component, though it can be painful to arrange it sometimes. Forget what anyone says is best, you are the arbiter. Get no more than 2 or 3 tracks on your source material and make sure you really know them well. Haul them around and do a ton of listening. You'll be surprised at what you can hear, but you'll also be surprised at what seems to make no difference whatsoever.

As far as tubes go, a sweet tube preamp will provide the friendly clipping and you can then send that on to your Krell, Levinson, or whatever (I have a Krell and Bryston amps) to get more watts. Lots of speakers need a lot of watts to sound good. My Martin Logans are not especially efficient, for example, nor are my B&W Nautilus speakers. I've had Klipsch corner horns (no longer) that sounded great on just a few watts.

Lastly, a lot depends on what you listen to and exactly when it was recorded. Things recorded in the heyday of tubes sound better with tubes (though the tube preamp to transistor amp gets you most of the way there). A lot of classical jazz is in this category. I was talking to a recording engineer about this one time and his response was, "Yea, those old recording engineers knew they were gonna lose a lot of high end so they boosted it in the original mix. Sounds harsh if you don't play it back on tube gear."

I first experienced that listening to some original Buddy Holly recordings. They just didn't sound good except on tube gear. If you're a rocker, or listen to later stuff, you probably want to hear it on transistor gear.

Cheers,

BW

[J Tiers]

For an eye opener, look here

http://www.talkbass.com/forum/showthread.php?t=617982

At the first post, with the pictures

1000 watt amp, NO VISIBLE HEATSINKING