The sound of class D compared to S'S or tubes

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm not looking for that much power as this will just drive my woofs in my house system and they are 98db eff. I've just never heard a class D amp. Of course the return policies these days are very good. I could buy it and return it if I don't like the sound.
 
my Crown XLS is good enough for daily listening "fullrange" - not sure how it compares to AB for woofer/sub duty. I've heard a lot of amps, some mediocre in MOSFET and tube - sometimes its a subjective synergy thing. My Klipsch Heresy Is with ancient PIO KBG caps sound excellent with a single ended UL EL34 and a certain DAC - just changing the DAC makes it not as good.
 
Even though I was once convinced that tube was the only way to go - particularly with "Full Range" single driver speakers - since I acquired a TPA3116 a few months back, I hardly miss either of my 2 EL84 P/P, or 2A3 monoblocks, or EL34 SE, and my Jolida 302 EL34 P/P has been languishing on the workbench partially modified for at least 3 years now (too busy building speakers, and surfing the forums in my "spare" time) .

Class D has come a long way since my first Tripath amp ( Kingrex T20) about 5? yrs ago
 
After 20 years experimenting with DIY tube amplifiers I unfortunately have to admit that my latest built with class D L15D-Pro LJM modules provide an overall better sound then my EL84 balans, SE EL84 and SE 6B4G (~2A3 ) amps. Yes, these LJM modules are cheap but great! Even loved above Hypex Ucd180 oem.

However the low input impedance of most class D modules requires special attention! I use Sowter step-down transformers (-10dB step down) as buffer before class D input which transforms impedance 10x and blocks DC signals. Terefore the input Cap at the class D could be removed which improve sound.
 
Class D amps aren't all the same, just like tube amps aren't all the same, but some of the "warmth" that tube amp lovers usually prefer comes from an uneven frequency response caused by the high output impedance of most tube amps, so you might prefer a class D amp that does not include the choke in the feedback loop (and thus has higher output impedance).
 
Hi!
I used the Behringer iNuke 1000DSP as a sub amp. The first one quit working after a few days, but the second one works for 3 years now (with 3 years warranty I saw no problem with this). So be sure to torture it in the first days.

It powered a 2x 12" sub woofer, build into concrete volume under the living room floor and really rocked the house, in the truest sense of the word.

Negative: Noisy fan, even as it seems not to need it, as it stays perfectly cool.

The DSP is not what you would expect from an US$ 250 offer. It is 24bit/96kHz, not only 16bit and the software is very good, mighty but quite logical and you have more options than you ever dreamed of. Not such an antique version as you may find it on many over priced plate amps.
You got to be aware of how the resolution of all digital devices reduces with low input level, but this is not the iNukes fault. This might induce noise if not matched correctly.

As I installed the iNuke in the cellar, I did not bother about the fan noise, but the Denon 3312 AVR sub out had not enough voltage to drive it directly without a hissing noise and some audible hum. A common problem connecting HIFI level outputs to level PA inputs, even as the iNuke does sense an asymmetric input. So I used an active DI-box (US$ 25 Behringer DI20 but think of the right plugs!) to convert the AVR asymmetric output to symmetric (balanced) and cut the ground connection, which reduced hum to zero and left only some minor hiss. This is typical with digital sound stages and low input levels. But it was easily cured using a second low pass filter at 200 Hz with 48dB/oct. That was something I found out playing with it and the many options and can be very important! The main low pass was managed by the AVR with 80 Hz. The result: No noise, even with the ear close to the loudspeakers.
So if you have noise in the loud speaker, there is an easy way to cure this, even as some blame this on the quality of Behringer stuff.
Turning the amp on and of is only a hardly audible "click", by the way.

Good amplifiers dont have a special "sound", but here it depends on the DSP settings what you hear.
The power output, compared to HIFI amps can be described as ground shattering. Loudspeaker control seems to be very good, even as I did not try any other amps with this installation. I used two 4 ohm 12" speakers modified for low resonance frequency, 200w rms/ 600w program and about 88 dB/w/m. The SPL seemed to have no limit in an open 800square foot living room. Definitely loud enough!
The dynamic limiter in watt/ohm are a great help, as you can adjust them to protect the woofers from possible self destruction. If you set them right, at some point the sub frequencies don´t get louder any more, but without clipping or negative audible effects. So even the heaviest explosions on Blue Ray effect channels cant harm you loudspeaker. This is a feature you also find in some of the better, more expensive commercial sub woofers. Good thing if you leave the kids alone. Password protection too :).

The rms output is "only" 2x210 watts in 4 ohms, which on paper seems a little low, but the power it delivers to the loudspeaker is felt like it was twice that much. One could really describe it as hard hitting. Good PA stuff has a lot of dynamic headroom you don´t expect if you are used to usual Hifi amps.
You can bridge it if you have a single sub, even with 4 Ohms. I took the double mono option, so the DSP settings apply for both channels. Again, nice menu.

Here in Germany they sell the 3000DSP for only 50$ more, which gives you a comfortable 2x 620W rms/4Ohms, so if you are mad about power, this might be an other option.

There is something that may surprise you, if you hold it in your hands for the first time. It is only 6.4 lbs, a weight you may expect from an inexpensive CD-player. Not to be compared with my 250 Watt mono A/B amps, which come at 28 lbs each.
When I finally installed the amp in the cellar, a friend helped me and made jokes about the "empty case" I bought. His face, when I had it all connected and played some heavy DVD really loud for the first time, was worth a million. His jaw dropped and he looked for shelter (Armageddon-Bruce Willis).
Some still seem to be right buy buying HIFI by weight...

The case is not "build like a tank" as some people always boast about there old PA-amps. It is thin aluminum and plastic. Dont step on it! Professionals always carry amps build into protective cases, so this is not a problem. They are thankful for any once they don´t have to carry. We got to get used to Class D.

So if you want go for the iNuke and are not in the PA business, maybe order a DI20, one RCA to 1/4" mono phone connector cable, two XLR male to female connections and don´t forget two Neutrik NL2 FX Speakon speaker plugs, maybe with cable. For a test you can power the DI20 with a 9 volt battery. So you you don´t have to bother with soldering connectors and complicated connection schematics.
Any other DI-Box should work, too. Inexpensive active boxes seem to be better with very low frequencies.
Best would be a asymmetric in / symmetric out pre-amplifier with adjustable level, so you can match any low level AVR to the iNuke´s input for minimum noise. But this has to be matched for your gear and depends on what you can source locally.

In my opinion this amp line is the best powered DSP bang for the buck you can get. If you can buy it with a money back option, even better. But I think you will be surprised, positively.

The opinion of the musicians and pros about the iNukes changed in the last years, now the majority raves about low weight, low price and great sound.
If you read of defects and bad sound, don´t forget many kids and hobby musicians simply don´t understand how to connect and adjust them.

My new iNuke 3000DSP should arrive tomorrow, special x-mast promo sale price 208 Euros including P&P:)

Chris
 
Class D produces a linear response faithfully producing what is delivered.***

No, it usually doesn't. Class D amps often have high FR variances into actual loudspeaker loads. Just like tubes. IMO, the best "triode" amp of all time is the old Panasonic XR55/57 AV receiver with TI PurePath amplification. It sounds just like a $200k Audio Note SET, only it's cheap, reliable, and has much more headroom.

There are a very few designs that do act as you describe: Hypex, Icepower ASX2, Anaview.

Though for systems with active crossovers, pretty much any D amp will work well. Just base the filters on measurements of the drivers driven by those specific amps.
 
there good class d, good tubes and good SS. I use class d andlove it, but it lacks a bit of the goosebump and emotional side of my tube amps, but the class d is more detailed, neutral and punchy and clean bass. Its a trade off until you decide to go big I think :) and pay big bucks.

Tubes is likely king in terms of highs. class D and SS are better at bass. so if you have active system, tube for highs for sure and class D or SS for bass.

But I think the best of the best is still lateral mosfet SS or tubes and I'm not alone to think that.
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
The TPA3116D2 when paired with a good input transformer allows use of the balanced inputs and bypassing the input caps has been reported to be a significant improvement in clarity and soundstage. It is fairly expensive upgrade as the trafo's cost significantly more than the amp. I have heard that this mod transforms the 3116 into a world class amp able to keep up with the best out there at this power level.
 
The TPA3116D2 when paired with a good input transformer allows use of the balanced inputs and bypassing the input caps has been reported to be a significant improvement in clarity and soundstage. It is fairly expensive upgrade as the trafo's cost significantly more than the amp. I have heard that this mod transforms the 3116 into a world class amp able to keep up with the best out there at this power level.

As you say, the transformers are pretty expensive, so quite a leap of faith based on something that you have heard second (or third) hand, and that "has been reported" to be an improvement.

Are there any rational reasons a transformer would be an improvement? Any independent verification?
 
Are there any rational reasons a transformer would be an improvement? Any independent verification?

Well, I use Sowter 9150 attenuator transformers as input/buffer for my class D modules for following reasons:
- adjust amplification (multiple step-down ratio)
- increase input impedance (at-10dB step-down 10 times higher input impedance)
- blocking any DC signals from output source (i.e. during start up my DAC with tube output causes short time DC at output)
- delete input Cap (depending on design)
- high frequency filter
- adapt balance input
-:rolleyes:

Many rational reasons and experienced improvements over other solutions as FET buffers (to increase impedance) and installing better coupling Caps.
 
Last edited:
Well, I use Sowter 9150 attenuator transformers as input/buffer for my class D modules for following reasons:

I can agree with most of the arguments, but there are two I would like to comment on:

- blocking any DC signals from output source (i.e. during start up my DAC with tube output causes short time DC at output)

That is fine as long as the DC is a temporary condition. A transformer is not a good solution if you have a permanent DC offset, as the DC biases the transformer and can cause unsymmetrical distortion.

adapt balance input

Not sure what sort of "adaptation" is needed. The TPA3116D2 has a differential input that can be used as balanced or unbalanced.
 
As you say, the transformers are pretty expensive, so quite a leap of faith based on something that you have heard second (or third) hand, and that "has been reported" to be an improvement.

The improvement works with hand-wound el-cheapo ferrite cored trafos too. IME nothing much to do with keeping or not keeping the input coupling caps though.

Are there any rational reasons a transformer would be an improvement? Any independent verification?

Common-mode noise reduction is my working hypothesis for why trafos improve the sound. If all interconnects were balanced then maybe there'd not be room for improvement.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.