engineer?

[shooter]

Does I leading or lagging E(V) matter in audio 
I know in my old life the engineers got all excited, me not so much, since it didn't really help me fix broke things.
I think my brains overloaded with vitamin D

[92Volts]

Phase shift matters if you're using negative feedback. With "local" feedback like speaker to phase inverter, it's not a big deal.

Through many stages, phase shift could be enough to cause positive feedback, at certain frequencies. It is a real issue for hifi for example, where you might want to take NFB from the output all the back to the input to decrease distortion of the entire amp.

[jjasilli]

You're asking if phase matters in audio, which includes not only mono guitar amps, but also hi-fi; stereo guitar amps; and sound management, i.e., mixing boards .  Once you have more than one channel, and/or signal stream, phase matters.  (This includes NFB, as 92volts points out, which is a separate signal stream mixed back in.)

[PRR]

I think you are talking of Power Engineering. *Single* frequency. In that world, lead/lag is shorthand for reactance, which does not directly absorb power in a useful way, but causes power waste from energy sloshing back and forth in long wires, and funny loads on generators and lines.

In audio our frequencies cover 1000:1 range. It is possible, normal, to be leading at one freq and lagging at another.

The ear is not real fussy about phase of single sources. Mixing several essentially identical sources does invite phase trouble.

I found this in my sim. Audio amplifier with cap input and transformer output. The amplitude response is typical for audio: DC and LF is blocked, there is a long flat zone, and then response falls off above the band we promised. Because this is a simple amp with few poles (R-C or R-L pairs), the phase only goes a bit over 90 degrees for any frequency we would care about. And no surprise, on this fairly flat response, the phase is about 45 degrees where the response is significantly down.

[jjasilli]

Shooter, you specifically asked only about the electronic aspect of phase.  But in audio, speaker phasing matters too.  In small venues this is typically ignored; little can be done about it anyway.  But it should be addressed in larger venues.  Bass lags behind higher frequencies.  So hi frequencies need to be electronically delayed; or bass cabinets need to be physically located forward, maybe 14 feet or so to match hi's.  And bass cabs should also be placed so as to cause minimal phase cancellation with bass soundwaves reflected from walls, to reduce room nodes -- places where bass cancels itself out, or adds up.  In very large venues, with multiple speaker systems, control is needed to prevent the echo effect we used to hear in stadiums.

[shooter]

In that world, lead/lag is shorthand for reactance, which does not directly absorb power in a useful way, but causes power waste from energy sloshing back and forth

YES Thanks, when the ? formulated, I couldn't place the context, hence the post

I was "looking" for the wasted energy aspect since I like xSE and efficiency sucks, would "tuning and matching", "greasing the signal chain", somehow get more than a headache.  Since this is a band of freq I suspect not

JJ, I've got the audio signal phasing understanding down enough to know "there's nothing to see, move on" for the most part, I still look at phase input to output and note the phase shift, (my amps average 20-30degres), I also swap wires to give me in phase in to out

thanks again
dave

 

[jjasilli]

I was "looking" for the wasted energy aspect since I like xSE and efficiency sucks, would "tuning and matching", "greasing the signal chain", somehow get more than a headache.  Since this is a band of freq I suspect not
Yes, in terms of wasted energy in the internal signal chain I agree it's not an issue.  We're dealing with voltage gain, not power per se.  We can easily generate more voltage gain than we need.  That changes externally when speakers are driven.  Passive crossovers waste energy; so active crossovers are used.  Vertical speaker arrays concentrate sound.  Every 3dB of SPL saved reduces amp power needed by 50%.  In small venues amp power is cheap enough.  In large venues 10's of thousands of watts can be saved which becomes significant.

[ALBATROS1234]

I was taught ELI the ICE man. ELI means voltage leads current in an inductive circuit. E=voltage L= inductive I= current. ICE is current leads voltage in a capacitive circuit.  An inductive circuit can have capacitance and vice versa but it is mostly inductive and the same for capacitive. I would assume an audio circuit is mostly capacitive except maybe for the transformers which are inductive in operation. So by that logic in a transformer voltage is leading in the rest of the main circuit it would be the opposite or lagging as it's usually stated. We learned a phase triangle for reactance etc each side stands for this or that. It's just a way of figuring out what is plugged it where in ohms law and reactance formulas. I don't use it much but we did an awful lot of that in school.

[jjasilli]

Yes, we have 2 trains of thought running in this thread: phase shift; and insertion loss due to reactance. 

[shooter]

phase shift; and insertion loss due to reactance

phase shift is "fine" in my brain, reactance is my curiosity. I'm finalizing my next build and want to "think about" maximizing efficiency, I already know how to "brute force" it just fine, think I've got tweaking the PA stage to optimize max coupling of the OT...at a given F, still weak on impendence coupling calculations stage to stage. 

aside;

Every 3dB of SPL saved reduces amp power needed by 50%

I sold an amp recently, I had a 15" jbl >100db, the buyer really hated "how clean" the speaker was, but couldn't understand why the amp wasn't as loud with his speaker He eventually borrowed his buddies Marshall 1/2 stack, now he wants more amps
moral of the story, matching speakers to a player is just as important as matching the amp to the player

[jjasilli]

still weak on impendence coupling calculations stage to stage. 
Not sure I see a practical problem here, until the power tube(s).  The Rule of Ten's will give proper impedance bridging for a voltage stage.  Your KT-88's are signal hungry, but impedance wise, it seems you have that covered with a CF buffer.

[ALBATROS1234]

when i get a chance ill dig through my books and try and figure out which parts would help and post here. theres plenty of info out there .you get rusty when you dont use it everyday though. they made us learn all that reactance phase angle stuff and it can be quite boring unless you have a reason to use it like tube amp design.

[PRR]

The cost problem in power systems with lead/lag is that: POWER is only when voltage and current are in phase. And billing starts with the Power used.

A pure reactance (lead or lag), V and I are not in phase. The energy is not consumed as power but returned to the generator. It can't directly be billed. 

If power lines had zero loss, this might be OK. (Actually it causes troubles.) However power lines run at 4% to 40%(bad day) loss on all the Energy that goes through. Reactive power goes out and back, double loss. And this causes Real Power (heat) in the wires. And not all gets back to the generator. And nobody is really happy.

The classic BIG offender is large motors, which run lagging at most conditions. Homes historically have not had enough lag (or lead) to care about.

The large-user tariff says there will be a surcharge for large power factor (laggy motors). This is justified by the real cost of over-sizing the lines to handle both the real power and the energy just sloshing back and forth. There's meters for this. The customer can pay the surcharge, or buy opposite reactance (usually capacitors) to hang near his motors and get low power factor (energy slosh inside the plant instead of all the way back to the generator).

None of this pertains to guitar amps. "Efficiency" is OK in a way, but you should use Class D power amps with SMPS. 90% instead of 35% efficient tube amps. What are you really trying to do? Minimize size/cost of power tubes? Bah, bottles are cheap, use more. Minimize the electric bill? Even an SVT won't eat $0.15/hour, and what other fun can be had for that rate? (Book-reading costs more.)

[shooter]

Bah, bottles are cheap, use more.

Ya, there is NO practical app here.  I guess I've "narrowed" it down to; in an OT, If the primary or 2ndary impedance is "miss-matched" audio power changes, is that a reactance force?

[jjasilli]

If the primary or 2ndary impedance is "miss-matched" audio power changes, is that a reactance force?
I think NO is the best answer to that specific question.  If you go from OT 2dary > to pure resistive dummy load, then the primary load is likewise fixed.  There is a pure, simple, linear arithmetic (not geometric, logarithmic) relationship from secondary to primary: a simple multiplication factor. But when you substitute a speaker for the pure resistive load, the speaker coil has reactance dependent on frequency.  That is not "the fault" of the OT, which has no recourse but to reflect back the ever-changing actual impedance of the speaker coil.  Also the power tubes have internal reactance dependent on frequency.  This too is not "the fault" of the OT.

What your question leaves out is that the OT has its own internal reactance, whether there is a matched or mismatched load, as seen by the power tube(s).  Different loads will cause different currents to flow in the OT coils.  Different currents may cause the reactance of the OT to change.  This is the stuff kagliostro wants to measure.

Also: though I get your desire for max impedance bridging in the signal chain, I do not mean to fully endorse it.  E.g., Leo didn't and his amps sound great.  Mismatched impedance bridging will attenuate hi's & weaken signal.  This is actually useful.  Gain stages produce harmonic overtones and possibly excess voltage gain which may be unwanted.  Purposeful impedance mismatching is a way to address that.

[shooter]

Purposeful impedance mismatching is a way to address that.

thx JJ, I'm satisfied, for now
and I'm taking notes for K while his tooth heals
not sure why, but once I'm "at the current build" my brain is formulating the next
I have 4 trannies, in cans, octal base, and zero data, except they came out of a 4 channel mixer, my "next" will be an audio build, hopefully using them 'tween stages?, part of input isolation?, "purposeful mismatch"   

[92Volts]

The loadlines you'd use to evaluate output stages already show the biggest efficiency issues.

Saturation occurs before anode voltage is 0v, wasting supply voltage (and tube dissipation, which is related to supply voltage throughout the cycle). You can minimize this with high plate voltage, the equation to estimate output power at the end of this article illustrates that well: http://www.valvewizard.co.uk/pp.html

Separate screen supplies become necessary for tubes to survive crazy-high plate voltages. As a bonus, decreased screen dissipation improves efficiency.

This idea matches datasheet examples, such as the crazy EL34 example that produces 100W with 800v on the plates but only 400v on the screens.

The only transformer I know of where phase shift is well-specified is this one and maybe others from the same company: http://www.one-electron.com/Trans/UBT2_10.pdf

Within the usable frequency range it's less than 10 degrees

[PRR]

> impedance is "miss-matched" audio power changes, is that a reactance force?

You can show "mis-match" with pure resistances.

You almost NEVER want "match". Best efficiency is for ZERO source impedance. A power example would be my house, a 0.6 Ohm feeder and about a 6 Ohm house: I actually get 90% of the power at the pole. This is an economic decision: 10X the wire would get near 99% efficiency, but would cost more than 30 years of electric bills or 300 years of the lost power. (And not my loss: the meter is on MY end of the wire.) (OK, there's other ways to improve my wire, but all not worth it.)

With tubes, there is both series resistance and shunt resistance. Both are expensive to reduce, far more expensive than another gauge of wire. The tube-book suggestions ARE good guides-- the old guys were no fools, and their jobs depended on tubes doing well so they sold well.

[shooter]

the old guys were no fools

No they weren't, I would never have made it where I did without them. One in particular, Master Chief Jahn. He was In charge of my last Navy school, Mk51-D, 50weeks, 8 students, 2 day-day instructors.  Everything up til that school was mostly math n theory, now all practical, meters, scopes, spectrum analyzers, and 2 broke Radars, EVERY day! Master Chief, learned troubleshooting on River boats in V.N., after the bad guys blew them out from under him n crew, he'd get the systems back up n do it all over.  3 of us students went to the same ship, 6 mths later Jahn shows up and we finish out our careers, his 30yrs, mine 6.  He was my "LEO" mentor.  To date, my longest, continual system troubleshooting, 39 hours, I burnt out 2 other techs, then Master Chief stepped in, and said "lets fix this thing", handed me my 106th cup of coffee, In my plastic smiley face foot cup I got from mailing in cereal box-tops , we nailed it about 8 hours later, I went to bed!!

[jjasilli]

I have 4 trannies, in cans, octal base, and zero data, except they came out of a 4 channel mixer, my "next" will be an audio build, hopefully using them 'tween stages?, part of input isolation?


Sounds like these are mic input transformers.  If so, they are meant to match Lo Z mic's to the hi impedance of a vacuum tube input stage.  Isolation is not the primary purpose.

They are not suitable as interstage trannies.  Transformer coupling of gain stages is obsolete, being replaced with the resistive-capacitor coupling we are all familiar with.  There are a number of good reasons for this. 

[shooter]

Transformer coupling of gain stages is obsolete,

ya, so am I
I only built 2 PP amps, both sold, both got nice reviews, both use inter-stage trannies as the PI, Soooo much simpler to install n wire. Even the price is ~ same.

Lo Z mic's to the hi impedance of a vacuum tube

isn't that true with tube coupling also, tube out lo Z, tube in HI? 

I get the tried n true, I'm not trying to re-invent anything, I "borrow" all my good ideas
And if I can't find a good idea from the past that used inductive coupling 'tween stage for audio, I won't use them, but they look cool, I can paint them, I have a BB,

[jjasilli]

ya, so am I I only built 2 PP amps, both sold, both got nice reviews, both use inter-stage trannies as the PI, Soooo much simpler to install n wire. Even the price is ~ same.


Yes , I was hoping to pull someone's chain! 


There's nothing wrong with interstage trannies, but there are factors.  In earlier, more primitive times, OT's maybe were good for 100 Hz to 8,000Hz.  Also back then, tubes were pretty low mu.  Interstage trannies could assist with voltage boost.  Power tubes were big bottle triodes good for only a few watts output; and needed large input voltage, like 115V.  With only low mu tubes to drive them, tranny coupling could also boost that drive voltage.  The bad news: trannies are big & heavy compared to modern hi mu preamp tubes; full frequency range trannies are expensive; and they generate magnetic fields which can pollute the signal stream.  OTOH, no caps needed so phase stays better aligned which is important to hi-fi people. 

I agree that probably the best place to use such a tranny is in lieu of a PI.

Mic input tranny used as interstage tranny: "Audio transformers are used where the impedance is not naturally what is needed for best performance. For low-impedance pickups or microphones working into tube grids, the transformer steps up the minute voltages (in this case impedance is incidental) to achieve better margin over tube noise. Sometimes interstage transformers are used for essentially the same purpose. The limit to what can be done occurs when the impedance connected to the primary is stepped up to such a value that it adversely affects performance in some other way." http://www.vias.org/crowhurstba/crowhurst_basic_audio_vol2_072.html

* You'd have to do the math to see if the impedance ratio of your mic transformer is useful in an interstage circuit. 
* I honestly don't know if the mic tranny could handle preamp tube current.
* The frequency bandwidth of a vintage PA mic tranny might not be ideal for hi-fi.  It only needed to be good enough for the narrow bandwidth of the spoken human voice.
* I am intrigued by the idea of low mu tube stages with interstage trannies.  It should sound very pure for hi-fi.

[shooter]

It should sound very pure for hi-fi.

We'll see, I just ordered the PT for my current build, but the next one will be hifi-ish. 

[jjasilli]

Here's an interesting article on interstage trannies:  http://www.ankaudiokits.com/agrove_interstage.html


It does tend to confirm, by implication, that a vintage mic tranny is probably not a good choice for this application.

[shooter]

Good read, thx, bookmarked for the future!

[PRR]

> with tube coupling also, tube out lo Z, tube in HI?

Small tube's output is many K-Ohms. Microphone is close to 100 Ohms. Small tubes will not drive mike windings.

The grid side of a mike transformer never has even 1V across it (preamp grids won't take more). Your power tube grid probably wants a lot more signal.

Mike transformers can be sold to studio gear DIY guys.

[Ed_Chambley]

Great thread.   


Learned a lot of new stuff.

[shooter]

Small tubes will not drive

Relegated back to the oldie moldy box it is, thanks