Digital Amplifier Company

A word about load dependence...

2010-07-23T11:12:00.000-07:00

There is no such thing as "completely load insensitive"...

Also, it takes wrapping feedback around the output filter to REDUCE load dependence, and in our experience, this affects the sonic quality of the amp in a negative way --- no pun intended. In fact that "too much negative feedback" sound (sometimes associated with Class-AB amps) is just what you get with all the phase margin compensation that's required to do so. As long as you're not driving a ridiculously low impedance, a low feedback design sounds so much cleaner than one designed to do better on the bench than into your ears. We can all agree that bench measurements don't tell the whole story. Without writing a book on this topic here, lets just say that we started with a design that was excellent on the bench, then tweaked it over several YEARS to get it to sound the way it does.

I have been designing Class-D amps (and listening to them) for more than 20 years, and these opinions are not based on some fly-by-night theories. There are many misconceptions regarding Class-D technology, and companies that take the risk of designing their own Class-D amps from scratch can easily fall short from inexperience. Same goes for pre-fab module makers. I have seen audio companies employ switching power supply designers that have never designed an amplifier in their life to build their "proprietary amp technology". It is also important to know analog amplifier design in order to make a good "digital" amp. Yes, I know there will be somebody out there to once again point out that Class-D doesn't mean the same thing as digital, blah, blah, blah. I designed my first analog audio amp when I was in grade school, by the way... Designing analog amps is a good way to learn what results in good sonic performance versus good bench measurements. It can be considered a good lead up to digital amp design --- a prerequisite.

-Tommy

2.5mm plug for remote power control

2010-07-07T16:54:00.000-07:00

If you'd like to connect external power control and make your own connection at the amp side, try the CP3-1001-ND solder type plug from Digi-Key.

Here's a link:
http://search.digikey.com/scripts/DkSearch/dksus.dll?lang=en&site=US&WT.z_homepage_link=hp_go_button&KeyWords=CP3-1001-ND

How to use external power control is explained earlier in this blog. Thanks.

2010-04-15T17:49:00.000-07:00

When it comes to preamps, I prefer balanced to unbalanced. My reference tube preamp is an AR SP9, and that's single ended (unbalanced). It sounds very nice with the adapters, but it's a little noisy for my tastes. Balanced preamps are typically quieter end-to-end. A black background is an important advantage to our amps.

Some people define "true balanced" differently than others. To me, it means that all signals are handled differentially throughout. That means double the contacts on selector switches, double the number of pots on volume controls, and double the number of gain stages. Some think that the disadvantage of double the gain stages, even though they work in parallel, is not worth the trouble, so they convert the inputs to single ended. The signals are then processed one per channel until it is converted back to differential (balanced) in the end. Tube units typically do this with a transformer.

You can think of it this way... Every stage of the preamp adds some noise and distortion (even if it's near zero). If you add the exact same noise and distortion on the positive version of the signal as on the negative side, they cancel out. This assumes that the error signal is exactly the same on both polarities, and that's pretty unlikely. Due to that issue, it is twice as important to have low noise and distortion in all gain stages of a truly balanced design, making it twice as difficult to achieve the same result.

In an "internally single ended" design, the signals are converted to single ended before processing. Special handling is necessary for low noise and distortion since you can't cancel out common mode "errors". One unit that converts to balanced at the end of the chain is the Purity One from Bella Extreme with the balanced output option. I'm trying to get one of those to try out.

Remote Standby Operation

2010-04-08T04:43:00.000-07:00

The remote jack on the back of DAC4800A and Cherry amplifiers allows you to force the unit into standby from an external source to save power. Some home theater systems have a control signal output that can be used to control the amp's standby operation.

The input takes less than a milliamp to drive. When the input is below 2V, the amp is placed in shutdown. When it is above 3V, the amp is active. When the input is not connected, the amp is active in the case of the Cherry, or controlled by the user in the case of the DAC4800A. The DAC4800A has a pushbutton standby control, and external control overrides the user setting. The Cherry series of amps does not have a pushbutton standby control.

The plug is a stabndard 2.5mm size barrel type, typically used by "wall warts".

A word about High Power Amps

2009-11-24T20:17:00.000-08:00

Is more power better? We believe it is, within reason!

More power usually comes with a few compromises, such as zero-crossing distortion, transient distortion, and noise. We worked for decades to find a solution to these issues and think we've found it. Doing so is not cheap, but it doesn't need to be outrageously expensive either.

A quick word about zero crossing distortion... At low levels, high power Class-AB amps have real issues controlling a reactive load. Imagine driving a car where the middle 10 degrees or so of steering wheel rotation has no effect on the direction of the car.

Regarding noise, the more power output, the louder the same noise figure (dB) with respect to maximum power sounds. There's no magic, so you need really good SNR to sound good at low levels with high max power.

Another concern is peak output delivery. If you can put out 100A into 4 ohms momentarily (micro- or milli-seconds), this is equivalent to 40,000W! The path from the power supply to the speaker needs to be as direct as possible to provide clean peaks.

This is just touching on the subject, but hope it clears up a few issues that are commonly misunderstood.

Our DAC4800A and Cherry amps address zero crossing distortion. Our design inherently has 0% zero crossing distortion.

Our DAC4800A and Cherry amps address transient distortion. Our design uses output devices capable of more than 100A output.

Our DAC4800A and Cherry amps address noise. Our design achieves more than 110dB SNR. Just try the "ear to the speaker" test...

And... We can also deliver more than 1,200 total watts continuous power.

So, next time you find yourself discussing high power amps, remember that it can be DONE RIGHT!

Equipment and Music mentioned on our review page

2009-01-04T20:25:00.001-08:00

Here is the link to our review page:
http://www.digitalamp.com/DAC_REVIEWS.htm

Here are a few lists of mentioned items in categories...

Speakers:
Dennis Murphy
Canton Karrat
Apogee Caliper Signatures
ACI Sapphire
Magnepan MG1.6
Gradient Helsinki 1.5
B&W 805s
SPTech Minis
Von Schweikert VR-4SR MKii
Von Schweikert VR4JR
VAC PA100/100 amp
Joseph Audio RM7si
Quad ESL 57
Quad 22L2
Ohm5 MKii
Advent lG
Bose sub/sat (unknown model)
Hafler 300 Series

Amps:
Pioneer/Luxman Hybrid Amps
Denon
Parasound
NAD
Manley Stingray Integrated Tube Amp
Conrad Johnson CAV 50 Integrated Tube Amp
Cary Slam 100 Mono Blocks
AVA Insight 440H
ICE-based H2O Audio S250 Signature
Exodus Audio UcD400
Aberdeen Signature TacT S2150
PrimaLuna tube integrated
Channel Island D200s
Sunfire Signature 600 watt
Aragon 2004
Pass Labs
Nuforce
Nakamichi PA-7 (stasis)
DAC 4800a
DAC4800A
Cherry
DAC/Cherry

Preamps and D/A:
MacIntosh tube pre-amp
Bel Canto pre-3
NAD C160 preamp
Jeff Rowland Capri
Bel Canto PRe3
Red Wine Isabella DAC/PRE
VAC Standard MKII preamp
Promitheus TVC rev 3 passive preamp
DACT CT-1 (volume control pot)

Cables:
Theils and Silver Cabling
Balanced KCI Gold Cable
Black Sand Silver Ref V Power Cord
VandenHul carbon fiber interconnects
DIY 8’ Axon cables

Misc Equipment:
Shanling T200A SACD Player
Audio Technica AT440MLa Cartridge
Adcom GFT-555 II Tuner
Rotel 1072 CD Player
Resolution Audio CD55 Player
Misc Terms:
Pro-Ject Perspective Carbon
Tube Box SE
GIK, 8th Nerve panels
Reality ICs
Black Sand PCs
PRAT, pace rhythm and timing
CLDjam
1.5" thick, braced walls on heavy 24"steel stands
350w into 8ohms and almost 700w at 4ohms
15A breaker
XLR inputs
binding posts
removable IEC plug
750VA torroidal transformer
1500VA transformer
thermal, shorted, voltage and over current protection
standby/operate switch

Music:
Dire Straights "Love over Gold"
Cowboy Junkies "Trinity Sessions", "Mining for Gold" (Margo Timmins solo), "Misguided Angel"
Portishead "Dummy", "Mysterons"
Peter Murphy "Deep", "Marlene Dietrich's Favorite Poem"
Yello
Everything but the Girl
Peter Gabiel "Shaking the Tree"
Monster Magnet "God Says No"
Radiohead "Kid A"
Delerium "Poem", "Aria"
Paganini Caprices
Bach's great organ works
Rob Wasserman’s Duets "Angel Eyes"
Ani Difranco’s "Garden of Simple" and "So What"
Ben Harpers "Fight for your mind" title song, as well as "Burn one down" portrays the drums with excellent speed and timbre. Impressive!
The Trinity Session by Cowboy Junkies
Patricia Barbers Modern Cool, "Company"
Red Rose Music vol 1 track 6 (twenty-third psalm, Shane Cattrall)
The Belfast Harp Orchestra, Earth Water Wind and Fire
McIntosh Test Disc "Voices of Angels" (Track 5)
David Manley recordings #8 Cantos de los Gitanos
Diana Krall
Tsuyoshi Yamamoto (piano)
Boston Pops
Cincinnati Orchestra
Pink Floyd
Eva Cassidy "You take my breath away"
Rebecca Pigeon "Spanish Harlem"
"The worlds Greatest Audiophile Vocal Recordings" produced by Chesky records
Red Rose Recordings "In a sentimental mood", "Misty"
George Winston "Autumn"
The Grande Piano Concert by Rick Wakeman
Red Rose "Recitative in Scherzo for Solo Violin" (track 4)
Ottmar Liebert + Luna Negra's "Viva", "Buddhas Flower" (track 2)Stravinsky's "Firebird Suite"
Tchaikovsky's "1812 overture", "Fanfare for the Common Man"
Dagda "Druids in the Glen"
Pink Floyd "Dark Side of the Moon"

Slew Rate of DAC4800A and Cherry

2008-10-08T19:48:00.000-07:00

The DAC4800A and Cherry amplifiers are "switching amps", so technically, the internal slew rate is that of the output stage (about 7200V/uS). The output filter of the amp removes the switching wave and yields an output slew rate of about 20V/uS (with an 8 ohm load). This is a respectable number, but keep in mind that it is limited by the output filter and increases as load impedance goes down.

The frequency response goes up to 60KHz on the DAC4800A and Cherry amps. This is a more valuable number than slew rate to many.

The sound of these amps can't be summed up with measurements (even though they are excellent) since they are designed from the ground up to deliver the best in music reproduction, not just test results in a lab.

Line Conditioners?

2008-09-30T19:21:00.000-07:00

The DAC4800A and Cherry are designed to tolerate large AC mains fluctuations so you shouldn't need any line conditioning (save some money there). In fact, these amps even tolerate brief brown outs, playing through them without any noticeable effect.

About Speaker Cables --- They Don't Need to Cost a Fortune!

2008-09-25T19:28:00.000-07:00

It comes down to resistance and inductance. If you can solder, buy 12 gauge or 14 gauge (if the run length is less than 15 feet) wire and get terminations from Digi-Key. By terminations, I mean banana plugs or spade lugs. I definitely recommend soldered connections instead of crimped for long life. I agree that this is a typical area for spending too much when it really doesn't make a difference to the sound or your system. This comes from an engineer with 20 years experience, not a marketing firm.

Audio Review Still Hasn't Responded

2008-09-25T19:11:00.000-07:00

We tried to add the DAC4800A amplifier (see http://www.digitalamp.com/4800a_amp1.htm) to Audio Review so we can get customers to add reviews there. However, there is no selection for adding a new manufacturer. The listing shows up with "unknown" as the manufacturer. Here's the link:
http://www.audioreview.com/cat/amplification/amplifiers/unknown/dac4800a/PRD_419903_1583crx.aspx

We wrote to them, but no reply. We posted to the help board, but the post was removed.

Does anyone know somebody in charge at Audio Review? We would like to get on the manufacturers list. Locking us out is just not fair. Thanks.

Why Our Amps Sound Better

2008-09-25T19:01:00.000-07:00

Some of our customers have told us that they love the sound of our digital (technically "Class-D") amps, but they don't like the sound of other digital amps (from other companies). It might be the output filter or feedback design. People often criticize me for saying that feedback wrapped around the output filter of a Class-D amp can have a negative impact on the sound. Actually, this could be your answer!

By the way, fake class designations like "Class-T" or "Class-Z" are actually Class-D. The definition of "Class-D amplifier" is "an amplifier that uses its output devices as switches". This is a pet peeve of mine.

Thanks for hearing me out.

Switching Amplifier Output Ripple Doesn't "Fold Down"

2008-09-24T22:21:00.000-07:00

Regarding "folding down", this is called aliasing and is explained by sampling theory. Here is a basic explanation:
http://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem#Aliasing

Aliasing only happens when the waveform is sampled, which is not the case for a switching amplifier output driving a loudspeaker. The time domain output waveform of a switching amplifier (after the output filter) has switching ripple superimposed on top of the audio. The ripple is ultrasonic, so it is not heard. The mechanical limitations of loudspeakers do not allow this superimposed waveform to enter the room, not that it would matter anyway, as long as the switching frequency (Fsw) is high enough (>100KHz). Most switching amps change states (switch) at a rate of 350KHz or greater. Higher switching rates cause additional EMI and lower efficiency, and lower rates are harder to filter. The power dissipated in the loudspeaker by the ripple is typically a few milliwatts.

In order to accurately measure the performance of a switching amplifier, a filter that removes most of the ripple is usually placed between the amplifier under test and the measurement device. Such a filter requires flat response out to 20KHz or greater, but a rapid fall-off by the 100KHz range. Attempting to measure a switching amplifier without such a filter can drive the measurement equipment nuts as it tries to sync to the amplifier's output ripple instead of the audio signal.

Links to DAC4800A and Cherry Amplifier Information

2008-09-24T22:09:00.000-07:00

Here's a link to the DAC4800A page:
http://www.DigitalAmp.com/4800a_amp1.htm

Here's a link to the Cherry page:
http://www.DigitalAmp.com/Cherry_amp2.htm

Here are links to reviews:
http://www.DigitalAmp.com/DAC_REVIEWS.htm
http://www.TrueDigitalPower.com/ (Amazon)
http://www.TrueDigitalAmp.com/ (Amazon)

Damping Factor

2008-09-24T22:04:00.000-07:00

We received a question about damping factor. Both the DAC4800A and Cherry were tested for this at 1KHz, which is higher than normal for a damping factor test. Into 8 ohms, it measures around 200 (Rload/Rout). All the way down to 20Hz, the measurement was about the same.

There has been some controversy regarding the real meaning of damping factor, however. Tube amps are known for their sweet sound but usually have a low damping factor due to high output impedance. We believe the natural open loop performance of the amp comes into play as damping factor lowers. In most cases this occurs as frequency rises. With good power supply design and strong output stages, damping factor can mean less to the sound of an amp.

The proof is in the pudding, so there's no substitute for listening.

Output Resistance Measurement Trick

2008-09-24T21:37:00.000-07:00

Here's an easy measurement trick to find output resistance, but you need an accurate audio analyzer to use it:

Load the amp with 24 ohms and change the input voltage (sine wave) until you get exactly 24 volts out of the amp. Remove the load and subtract 24 from the new output voltage measurement. This is the output resistance in ohms! 24 ohms was picked to increase accuracy, but you can do the same thing with 8 ohms and 8 volts. The voltage measurement should have 3 digits after the decimal point for this to yield an accurate measurement.

Knowing the output resistance, you can calculate damping factor (DF = Rload/Rout). Note that output resistance can change over frequency.

Preamp Output Impedance and Amplifier Input Impedance

2008-09-24T21:16:00.000-07:00

We received some questions about DAC4800A (and Cherry) input impedance, which is 10K. Here was our response:

It's simple... You get more noise with higher input impedance. Going from 10K to 100K might cost as much as 3dB, but since stock SNR is over 114dB, there's room. We don't do this stock because most preamps have a 600 ohm or lower output impedance, more than 10:1 (the typical "rule of thumb") to drive the amp. We opted for better SNR. The current of the preamp's output driving 10K is VERY low (microamps), so that's not really a significant factor. We can modify amps for 100K input impedance when requested, but there is some rework required.

We don't recommend preamps with AC coupled outputs, although these are few and far between. AC coupling between the preamp and amp creates a high pass filter that can rob you of low bass response.

A Few Words About Output Current

2008-09-24T21:14:00.000-07:00

The output devices in the DAC4800A and Cherry amps are capable of pulsed drain current of more than 150A! However, the output stage cycle-by-cycle current limit is set to about 30A instantaneous.

Our Amps are NOT based on Ice Power Modules

2008-09-24T18:12:00.000-07:00

We DON'T make amps with Ice Power modules! I replied recently to someone's "just another ice power amp" email comment. It takes a lot more amp design experience (20 years for me) to design high performance, high efficiency, high power amps from the ground up than to put off-the-shelf modules in a box with a power supply. Something like 90% of the digital amps out there are like that, so we're special. We designed the guts. We build the guts. We are the guts.

Digital Amplifier Company amps are built with proprietary patented technology that took more than 10 years to develop. It's a novel approach to modulation and control that allows closer-to-the-rails output swing while holding noise down as far as 120dB from max output. Really low THD is also possible with this technology, down to 0.0006% on the bench.

We are customer centric because we appreciate having the privilege of serving them. They are the reason we exist, and there's nothing quite like having customers repeatedly tell you that your design is the best they have ever heard. That makes all the hard work that goes into each amp well worth it. Thanks again.

A Few Words About Buying Vintage Loudspeakers

2008-09-24T17:51:00.000-07:00

Watch out for speakers with drivers you can't replace. Also, take a good look at the condition of woofer surrounds, dust caps, and the like. It may work well and sound good, but for how long? I burnt out one woofer each on two of my speaker systems, one DIY and the other a well known high end name, and couldn't replace the burnt drivers with exact replacements. One was luckily fixable with a "close replacement", and one lays dormant (Focal dual coil woofers). Of course, when one side blows out, you want to replace both sides. Crossovers are easier to find replacement parts for, that is, if the components are marked with the value (air coil inductors are notorious for no markings). Also check if the drivers are original. Good luck.

REAL ADVICE ABOUT PICKING AUDIO POWER AMPLIFIERS

2008-08-22T23:22:00.000-07:00

I'm always weary of reviews from magazines. There are hardly any bad ones in fear of "burning bridges", getting sued, or losing a potential advertising client. No magazine in their right mind will admit to this. In fact they will deny it tooth and nail. Same deal for "product rewards"; many are bought and paid for indirectly.

Reviews that simply show measurements and no opinion whatsoever can usually be trusted, but are hard to find. Measurements can't tell you everything, but can tell you a lot. Manufacturers rarely falsify or embellish measurements due to fear of getting caught. However, there are some tricks that can make measurements look better, such as rating power at 10% THD, but this is also very rare. The most trustworthy source of how something sounds (beyond measurements) is actual customers. However, people have a tendency to like whatever they bought, so watch out for reviews that have no comparisons to anything else.

Regarding well recognized brand names, don't believe that just because a brand is well known and has a long history that they only make spectacular products. I had a WELL KNOWN (maybe the MOST well known amp brand of all) amp and was explaining my total disappointment with it to a friend. He said something like "I can't believe you bought one, everybody knows that amp sucks. It sounds just awful and was a complete failure for them". He also said that despite the bad sound of this design, they sold a TON of them.

One important figure is signal to noise ratio. Anything less than 100dB with respect to rated power should be avoided like the plague. There's no reason in this day and age that you can't design just about any type of amplifier at least that good. Noise is the enemy of good sound! If there is an SNR measurement at 1W, look for at least 85dB. This is A-weighted. Unweighted SNR is not as meaningful of a spec since the human ear has frequency dependent sensitivity curve. The "ear near the tweeter" test is actually quite useful if you have a single tweeter that's no more than 2" long in any direction. Listen not only to how much noise there is but also the character of the noise. Quality amps are dead quiet or close at idle.

Look for amps with THD+N below 0.01% for a good portion of the power curve and at all frequencies. This is especially important at low levels, so look for THD+N well under 0.1% at 1KHz, 100mW. Unfortunately, many manufacturers don't provide this spec, and if they do, it's probably on a graph (THD+N versus power), not in text, and sometimes you need to interpolate. Look for amps with direct coupled output. Output transformers are fine for distributed audio in your local supermarket sound system. They saturate at low frequencies.

Another one... Damping factor under 50 is a red flag. If damping factor or output impedance (reciprocal of this divided by the load impedance is damping factor) is not mentioned, it's probably poor. Also, as damping factor goes higher and higher it means less and less. I doubt any human being can hear the difference between an amp with a damping factor of 300 and one with a damping factor of 450 that's otherwise the same.

Regarding the "sound" as perceived by any reviewer comparing two or more amps, if the comparisons aren't double blind, they aren't super valuable. The mind simply can't remove bias reliably otherwise. Single blind is much more meaningful than non-blind. You might also be surprised how many so-called "golden ears" fail double blind listening tests (essentially can't tell the difference) when the amps are similar! It's funny to hear the excuses when you call them on it. This is especially amusing with aftermarket modifications. My theory is "buy something that was designed right in the first place".

Anyway... Regarding output power, watch out for manufacturers of anemic amplifiers that tell you their 30W amp is great for driving your 10-driver per channel tower speaker system. You need adequate power to drive your speakers through peaks unless you never "turn it up". Big speakers need big power to play loud and clean at the same time. I recommend a few hundred watts minimum per channel unless you're driving a pair of bookshelf speakers in a small room. Also watch for amps with lots of power but bad specs otherwise. Beware of amps with a spec called "instantaneous peak power" of something similar. This is technically twice RMS output power at rated output right before clipping. If you see this rating and it's NOT twice rated power, there's something fishy going on. This spec is used deceptively so beware of its use either way.

Regarding inputs, when possible, insist on balanced inputs or both unbalanced and balanced. Balanced audio is inherently better due to built-in noise cancellation, and if you can afford a good preamp with balanced outputs, use one for your primary audio system.

Here's something I wrote regarding pro amp measurements more than 3 years ago:
http://www.livesoundint.com/archives/2004/dec/power.pdf
Yes, I worked in pro sound for several years, but learned a lot. I know lots of high-end people think anything pro audio is junk. They are just wrong.

INTRO

2008-08-17T00:15:00.000-07:00

I'm the founder of Digital Amplifier Company.

On our web site, we explain that we don't go to trade shows or pay for standard print or web advertising. Do you have any thoughts on this?

This post is to get feedback about our products. If you have auditioned or purchased either of our current products, please give us a shout.

This post is to get feedback about our products. If you have auditioned or purchased either of our current products, please give us a shout. The DAC4800A is a 1000W amp and the Cherry is a 1200W amp. They are both 2-channel units and have very respectable specs (<0.009% distortion, near 120dB SNR). Some have said that our amps are the best they have ever heard regardless of price. Reviews are posted on our site and on Amazon.com (we sell there too).

Please let us know if you have heard one of our amps or if you have visited our web site (www.DigitalAmp.com). I'd like to get some feedback ASAP, especially concerning comparisons with other high end amps. We plan to introduce some new stuff soon as well.

I'm offering my expertise regarding digital (Class-D) amps since I have been designing them since the mid 1980s. If you have questions about these designs, let me know, and I'll try and answer them.

If you are only a casual music listener who thinks low bit rate MP3s on your iPod are the best in audio, please refrain from commenting!

Thanks very much.