Tube Vs. Solid State

Tube amps simply wont ever measure with better distortion than a proper SS, I'm not sure why it keeps getting mentioned as if it means anything.

 

Rowland have just recently began to offer class d and TPA based amps. when I mention Jeff rowland amps, Im talking about what made him famous and not his new amps offering that have been quite badly received/reviewed. have a look here at a good write up between Rowland class ab designs vs his new class d: http://hifi-advice.com/Amplifier-Technology-ClassAB-vs-ClassD-info.html

in particular that to build a good solid state amp you need to spend as much as you do to build a good tube amp
imo, a no compromise design is either Lateral mosfet or Tubes. "Bipolar output Transistors and vertical Mosfets (Hexfets etc.) are all
extremely temperature dependent in certain parameter.

Tubes and Lateral Mosfets are not temperature dependent or in a way
that stabilises the operation.

Class AB works well as long as the output devices work stable with
temperature, so tubes and Lateral Mosfets. It does not work well with
Bipolar transistors or Mosfets, unless the whole output stage can be
made from a single piece of silicon (chipamp)."

To build a good class ab amp with no compromise such that it has mosfet voltage amp, mosfet output stages and jfet input for example still cost a lot to build. Your modulus is a chip amp right, and those are a sort of niche. very very good design and very cheap. those are the exception in the SS world: lateral mosfet amps are very pricy to build in comparison.

 

I've never encounter a SS amp, where the actual amp stage parts were a significant expense compared to power transformers and enclosure+metalwork.

 

Slowly approaching the BOM cost = amp quality mentality of Head-Case here.

 

In that sense, this is where I find the HD5## family a much better test bed than the 6## family. The 6## start in the 300s and double to the 600s, both of which are high enough that amps with moderate output impedance might not show any irregularities. The 5## has that 5x peak however, and starts low enough to give you all sorts ugly bloom.

 

I'm not seeing any measurements. Only opinion. What am I missing?

[citation needed] Which high end solid state amps have you built where the cost those of a high end (or even average) tube amp?

I suggest checking that claim against a text on semiconductor physics. Ben Streetman's book on the topic is pretty good.

All semiconductors (tube or sand) have temperature effects. In case of the lateral MOS, the tempco works in your favour. I.e. higher temperature -> higher Vgs. On top, the curve is relatively flat, which simplifies biasing.

Tubes have a tempco as well. The THD of my DG300B improves significantly during the first hour of use. It's constant after that.

That's complete bogus. You are correct that biasing a BJT or MOSFET output stage can be a bit of a challenge to get right, but it's not that hard. All the temperature effects simulate, so you can get the circuit dialled in in the simulator before ever turning on the soldering iron. Also, as Douglas Self reported in his Distortion in Power Amps series in Wireless World back in the 1990ies (later released in book form), the biasing can be made considerably easier by using a CFP output stage as this only requires the drivers to track with the bias generator.

It actually doesn't. I suggest building one and finding out for yourself. Or design one in a circuit simulator and calculate the BOM cost. Good output devices are about $5/each. Small signal devices, maybe 25 cents. Resistors, capacitors 10 cents. Maybe $1-2/each for electrolytic caps. The majority of the cost is the power transformer and heat sink. Expect to spend around $200-250 here. So we're talking a few hundred bucks total for a semiconductor amp.
Compare that with a tube amp. Power transformer: $125. Output transformers: $250/each. Right there the BOM cost exceeded that of the semiconductor amp and you don't even have any active devices in there.

Tom

 

The fact that you are repeatedly attempting to discredit someone who disagrees with you rather than adding information that supports your viewpoint automatically weakens your position. It is obvious based on the condescending language you use that you are emotionally attached to your position but incapable of elaborating any semblance of reason to support the strong beliefs you hold.

 

I hope not. At least it's not my intent to imply that. Rather, my opinion is that a competent designer can do wonders with regular parts and they don't need to be expensive.

My philosophy is pretty simple. I deliver circuits built using off-the-shelf parts that provide world class performance. I back my claims of this performance level up with measurements. To me, that's an honest approach. It allows me to sleep at night and feel good about my products. Others are free to disagree, but that's where I'm coming from. To me, "performance" is quantified by measurements, not by personal opinion.

In the end, what matters is that you can enjoy the music the way the artist intended for it to be enjoyed.

Tom

 

He's too busy bestowing us with THE TRUTH man, our weak peasant minds couldn't possibly understand

 

How did you suddenly turn into a Korean woman? What form of trickery is this!?

 

I'm sorry I ruffled your feathers, but when someone claims semiconductors have no temperature effects, they're basically claiming that 2+2 = 5. Am I supposed to agree with them?

If you're curious about my background (educational as well as experience), you can find my (embarrassingly outdated) resume here: Tom Christiansen, Resume (.pdf) I believe I know which way is up.

Tom

 

It's embarrassing that you just linked your resume on a headphone forum, Tom, I have to admit.

 

Embarrassing for who? I'm not embarrassed by my credentials. SteelCanon wanted to start a peeing contest. I'm not into that and I let that be known.

All I'm asking for is a fact-based discussion. Is that OK with everybody?

Tom

 

I understand that some believe/feel that measurements are only part of the story. I'm OK with that.

The facts I referred to were pertaining to whether certain types of semiconductors had temperature effects or not. I hope I'm not required to agree to statements I know violate the laws of physics.

Tom

 

I actually disagree with @SteelCannon 's interjection but I also feel, @tomchr , that it's not always what you say but how you say it. Sometimes one's approach is everything. Food for thought.

Carry on.

 

Yeah, I think my direct communication style comes across as arrogant at times. That's not my intent.

Tom

 

I was pointing out that your style of arguing, where you attempt to undermine others is actually undermining yourself and your own positions.

I wasn't disagreeing with or discrediting you so much as pointing out the weak position you were putting yourself into by being condescending.

I say that as a third party, who is interested in what you have to say.

 

In my experience the only "good" measuring amp which turned out to be extremely meh was the O2. So I must be fortunate that I can pretty safely predict my emotional response to electronics by looking at real life measurements. I've never heard the dreaded solid-state sound and Tom's Modulus86 (when're the new boards out??) was one of the most pleasant sounding amps I've heard; chip, discrete or otherwise.

Oh, and as for tube amps, no-one really builds them for excellent measured performance. If you want good figures, you need to build a PP tube amp, which will NEVER sound like a tube amp. If you want tubes, you go SET, because at the end it's never about just the tubes.