Current Drive in SS

[Derek Balonek]

This is probably going to be a bit long, but I think the results could be interesting. Most (all?) SS amps use voltage drive, that is, they put out a constant voltage regardless of speaker impedance. Current varies with impedance and that affects power. At lower impedances more current flows and there is more power and vice-versa. Pure voltage drive is seen as having an output impedance of zero.

Looking at the impedance curve of an average speaker, it has a resonant peak in the low frequency and then rises later on due to inductance. The curve doesn't spend as much time below the nominal impedance and this means it is often putting less power into the load at various frequencies.

In comparison, current drive pushes a constant current into the load and voltage varies with impedance. At higher impedances the voltage rises and this puts more power into the speaker. So looking at the same speaker impedance curve, there is more power into the speaker at resonance. Pure current drive is seen as having an infinite output impedance.

So why does this matter to us? SS amps operate with a near-zero output impedance and tube amps have a defined higher output impedance. So tube amps operate closer to current drive. The advantage is a more responsive, alive sound. The disadvantage is reduced speaker damping, so the speaker has greater effect on tone.

So, why not set up an SS amp to run on current drive? It is possible and in fact by using a hybrid setup an amp can have a variable output impedance. Varying the amount of current drive can affect box tuning snd even extend bass response if done well. There is also the bonus of sounding much closer to a tube amp. A few people have said that the defined high output impedance is a major part of the tube sound. With people who want to emulate tube amps with SS tech, the choice should be obvious.

As far as I can tell though, voltage drive is the standard in SS amps. Why? It's not too difficult to change.

 

[seamonkey]

Sure, do it. Just add some resistors to the output of the SS amp.
The frequency response will then tend to follow the impedance curve of the speaker.
Simple ohms law at work here.
And you'll get a nice smiley face curve that would be much easier with EQ.
For ancient history research the Carver Challenge. Adding resistors to the output absolutely convince audiophiles in blind tests.

 

[Derek Balonek]

I know it's not a challenge to set up, but why don't any major manufacturers do it? For personal use I'm going to work on a small MOSFET power amp with variable impedance in the future, but I'd like to see a bigger company try it.

 

[Passinwind]

I know it's not a challenge to set up, but why don't any major manufacturers do it? For personal use I'm going to work on a small MOSFET power amp with variable impedance in the future, but I'd like to see a bigger company try it.

Nelson Pass has done so, for one.

Check out this blog and the followup one, if you haven't already done so. I am not at all a fan of tube power amplifiers for bass in my particular playing contexts, or I would already built up one of Mr. Broskie's circuits by now...

 

[seamonkey]

I know it's not a challenge to set up, but why don't any major manufacturers do it? For personal use I'm going to work on a small MOSFET power amp with variable impedance in the future, but I'd like to see a bigger company try it.

Buy a four or so of these and wire them in various series parallel until you hear what you expect.
http://www.amazon.com/Electrical-Chasis-Mounted-Aluminum-Resistor/dp/B005GIRXRE

No amp company would do this since simple EQ does the same.

 

[Derek Balonek]

Nelson Pass has done so, for one.

Check out this blog and the followup one, if you haven't already done so. I am not at all a fan of tube power amplifiers for bass in my particular playing contexts, or I would already built up one of Mr. Broskie's circuits by now...

Cool. Thanks for the link. I've seen a few ideas on Rod Elliott's site too, that's where I first heard about current and voltage drive. Why don't you prefer tubes, if you don't mind me asking?

Buy a four or so of these and wire them in various series parallel until you hear what you expect.
http://www.amazon.com/Electrical-Chasis-Mounted-Aluminum-Resistor/dp/B005GIRXRE

No amp company would do this since simple EQ does the same.

Not really, matching the curve with EQ is quite a bit harder than one would think. It doesn't match the responsiveness of a defined impedance, and the effect is more than just EQ. Damping response is effected, and you can actually modify box tuning. EQ is kind of similar, but not close enough.

 

[seamonkey]

No, it's easy.
Here's an old paper that shows the difference in EQ
Invalid Link Removed

 

[Derek Balonek]

No, it's easy.
Here's an old paper that shows the difference in EQ
Invalid Link Removed

The sample size is a tad small there, and they specify that they are untrained. I assume a musician with experience would be able to tell better.

They also do not specify how they matched frequency. Doing so would require a parametric at least, which is a bit more than simple EQ. Standard bass/mid/treble probably wouldn't get close.

 

[seamonkey]

They were all guitar players.
The paper was IEEE peer reviewed, take it up with them on the sample size.

Try some resistors, that was your original questions. Look up current source and see how to change a voltage source into a current source.

 

[seamonkey]

Why don't you prefer tubes, if you don't mind me asking?

That's a myth that spread by some of the folks on my ignore list.

I use tube sound all the time.

 

[Derek Balonek]

Sorry, that was directed towards Passinwind. I am going to make an attempt at it, based off this design. I was curious as to why it wasn't too common, since it seems to have some useful benefits. I was also hoping to add a bit of info to the pieces of the 'tube sound' that aren't commonly mentioned. Most of the focus is on soft clipping, while I believe the output impedance to be the greatest effect.

 

[BassmanPaul]

Derek your link doesn't appear to work. I am familiar with Nelson Pass's work. He's written more than a few articles for Ed Dell's publications in the past.

 

[Passinwind]

Cool. Thanks for the link. I've seen a few ideas on Rod Elliott's site too, that's where I first heard about current and voltage drive. Why don't you prefer tubes, if you don't mind me asking?

I actually love tubes, especially for guitar. For my EUB and fretless BGs (I don't own any fretted BGs) my DIY tube preamp already gives all of that feel that I can possibly use though -- in fact I've had to seek out some new playing situations just to make it make sense. In general, growl and grit and bloomy sustain are at the other end of the spectrum from where I tend to want to be, but all are beautiful things for lots of music I rarely if ever play.

DIYaudio.com has a bit of discussion on this too, FWIW. Nelson Pass has a subforum there. Even though I don't see it as a bonus in my playing world at the moment, I find it pretty interesting. I noticed that Rod Elliot says he doesn't hear a huge difference, if any. I've seen many opinions that conventional ported loudspeakers are not the ticket with current drive too. That seems to mirror the experience and/or preferences of some tube amp players, interestingly enough.

 

[Derek Balonek]

Derek your link doesn't appear to work. I am familiar with Nelson Pass's work. He's written more than a few articles for Ed Dell's publications in the past.

Huh. It worked on my end. Best I can suggest then is search 'variable impedance' and 'Elliott Sound'.

 

[Derek Balonek]

I actually love tubes, especially for guitar. For my EUB and fretless BGs (I don't own any fretted BGs) my DIY tube preamp already gives all of that feel that I can possibly use though -- in fact I've had to seek out some new playing situations just to make it make sense. In general, growl and grit and bloomy sustain are at the other end of the spectrum from where I tend to want to be, but all are beautiful things for lots of music I rarely if ever play.

DIYaudio.com has a bit of discussion on this too, FWIW. Nelson Pass has a subforum there. Even though I don't see it as a bonus in my playing world at the moment, I find it pretty interesting. I noticed that Rod Elliot says he doesn't hear a huge difference, if any. I've seen many opinions that conventional ported loudspeakers are not the ticket with current drive too. That seems to mirror the experience and/or preferences of some tube amp players, interestingly enough.

Cool. In a lot of situations I could see bloom and sustain not fitting in.

I've heard that it's better with sealed cabinets of smaller size. Since amplifier damping is low, there needs to be more mechanical damping to keep it from being floppy.

 

[fdeck]

The reason for using voltage drive is that a speaker is a voltage device. What I mean is that a mainstream speaker has reasonably flat SPL response in its passband, as a function of input frequency, when the input voltage is held constant. So, voltage at the speaker terminals is a better analogue of the audio signal, than current.

When talking about tube amps, it's important to note that guitar and bass amps tend to follow somewhat different design principles. AFAIK low power guitar amps are often designed with no negative feedback. Higher power guitar amps, and almost all tube bass amps, have negative feedback, which lowers their output impedance considerably. So it's hard to generalize about what tube amps are actually doing, unless you're considering the full system with negative feedback. Also, if negative feedback is in use, then it might be dicey to stabilize a current drive system against oscillation when driving a complex load.

As for adding resistance to the output, it would seem to be preferable to simply add a voicing filter in the small signal chain, e.g., between the preamp and power amp, where parts are cheap and don't waste power.

 

[teemuk]

Most (all?) SS amps use voltage drive, that is, they put out a constant voltage regardless of speaker impedance.

Most SS bass and HiFi amps perhaps.

Most SS guitar amps use a combination of current and voltage drive.

Most tube amps also use a combination of current and voltage drive. That's one of the reasons for what the negative feedback loop is for.

Manufacturers could easily use the same thing that is used in guitar amps but they don't (customer input). Perhaps the current drive isn't all too ideal for bass amplification (flabs and farts out, etc.) All in all, generic SS rigs also seem to be far more preferred in this application than what they are in guitar amplification.