Does class A in some way "guarantee" or "makes it more likely", that there will be high R.out?
No. Althou you often see them with little or no feedback which is one of the mechanisms used to reduce Rout. And then SE amps, tube or SS, always run in Class A and tend to have higher Rout.
dave
It can be measured across the speaker terminals (Voltage). It will not be flat but have EQ like in the Nelson Pass article.
Sorry I dont follow, can you elaborate a bit?It can be measured across the speaker terminals (Voltage). It will not be flat but have EQ like in the Nelson Pass article.
Using a single strand of wire is for sure the most efficient, etc., same as the ideal amp is a single strand of wire with gain, but need a collection of differing wire sizes, lengths to find the desired properties, so a good quality power resistor is going to be tough to beat except possibly at extremely low power Vs its max rating, so historically for wide power ranges is to make a resistor grid using whatever the lowest rating required and as many as it takes to get the desired high power rating.i prefer using skinny wire than a fta R
If you measure at the speaker terminals you can see whether the source resistance is having an effect, and how much that effect is. If your amp has a low output resistance then it should measure flat with a speaker connected.
IME = in my experienceThanks, but what are you refering to with IME and BW? If i understand correctly, simply adding a resistor (or this really thin wire which i have installed now) will cause LF shelving, which is supposedly worse than "tracking" speakers impendance?
BW = bandwidth = range of frequencies
Not LF, but mids/HF, i.e. BW above the driver's impedance BW
Hmm, thought I already posted WRT series resistance Vs impedance tracking............
Oh well, once more into the breach.......
Load the driver in a sealed box sim, save and compare to it with 8 ohms of added series resistance to see that all you've done is raise the driver's effective Qts' (weaker motor), i.e. traded (a lot of) efficiency for a wider, peakier ~ 'one note' BW.
Coupled to a pioneer's 8 ohm matching impedance amp (not aware of any simming program that will do it) and the driver's frequency response will be shaped much more like its impedance plot except at a higher! efficiency.
Fast forward to '54 and premium amps of 100 W (peak) with variable DF was now available, so basic amp DF could be somewhat higher (lower output impedance = less distortion) and with the 'ne plus ultra' option of variable DF bass/treble tone controls made it the near enough ideal combo for an overall optimal dynamic speaker performance.
Altec 440 preamp/340 amp combo at ~ $309 (inflation adjusted $3321) was/is quite reasonable even today for this level of performance, but now would have to be hand crafted and presume today's tubes wouldn't be of the same quality/performance, yet may cost 10x more, so just a 'pipe dream' for all but a favored few.
I was very happy with DIY version of Pass labs F2 amp. Its output impedance could be "dialed in", so the 15 ohms output impedance worked very nice with Fostex Fe206e (which otherwise sounded "thin" with "normal" amplifiers)...
Can someone point me to a good/simple online sim tool? Couldnt find anything that I could use.Load the driver in a sealed box sim, save and compare to it with 8 ohms of added series resistance to see that all you've done is raise the driver's effective Qts' (weaker motor), i.e. traded (a lot of) efficiency for a wider, peakier ~ 'one note' BW.
Additionally, can someone point me in the right direction where I could learn how the maths aroud this issue works (alt. explain, if one can spare a moment).
What I think happens is:
-20w of power into 8ohm available from amp
-8 ohm driver gets 20w
-add 8ohm resistor in series
-voltage available halves for the driver, amp "sees" 16 ohms ->power available only 10w
-driver gets only 5w (attenuates mids drastically, while reducing efficiency drastically)
-for bass, at driver res.freq. impedance is 128ohm, with 8ohm resistor amp sees only a minor increase thus attenuation of those frequencies is much less
is this correct in the most general sense at least?
If it is, than using a serious resistor in series like pass suggests (around 47ohm) would obliterate efficiency completely, would it not? Would need much more than 20w right?
It is quite easy to try a resistor in series. Even if it wastes a lot of power, it's not a big deal. Fostex will be happy even with 5 watts 🙂
Great discussion guys, thanks! I have been learning a lot about this in the past few days and now understand the basic principle behind amps output impedance and how it interacts with speaker impedance across the FR. More importantly perhaps, I have stumbled into a place in audio that I always suspected (hoped) existed. Even ordered that book on Current-Source Amp design by that angry Finnish guy. Perhaps my interest in this escalated too quickly giving my lack of basic ET knowledge, but cannot help myself at this point.
In practical terms and in the long(ish) run, I will be looking for an amp that joins most of the philosophies that inspire me. Current-drive, high output impedance, no/low neg. feedback and low price of course😀 But I dont mind a higher price if the value is well represented. Feel free to come up with more suggestions!
In the short run I will try experimenting a bit with resistors and/or simple networks. Can you say a few things (or point me in the right direction), what sort of specs for the resistor should I be looking at with my 2x20w into 8ohms (voltage source) amp? I calculated roughly that my hairline speaker wire (26awg 5m) presents around 1.5 ohms of resistance, which in turn "equalises" the FR by ca. 2db or so between the mids at 8ohm and Fr (resonance fr.) at 128ohm.
I would like to try perhaps and 8 ohm resistor, that should 1/4 the power available (5w), which should bring ca. 6db "equalisation" and go from there. What do I need to be careful about shopping for one? What is the power rating that is suitable? 5w, 10w, 20w? Do i need to worry about (parasitic?) inductance of a resistor? Does it make sense to go for high quality 20w mundorf resistor, vs something generic (in terms of sonic degradation)?
Help very appreciated!!
In practical terms and in the long(ish) run, I will be looking for an amp that joins most of the philosophies that inspire me. Current-drive, high output impedance, no/low neg. feedback and low price of course😀 But I dont mind a higher price if the value is well represented. Feel free to come up with more suggestions!
In the short run I will try experimenting a bit with resistors and/or simple networks. Can you say a few things (or point me in the right direction), what sort of specs for the resistor should I be looking at with my 2x20w into 8ohms (voltage source) amp? I calculated roughly that my hairline speaker wire (26awg 5m) presents around 1.5 ohms of resistance, which in turn "equalises" the FR by ca. 2db or so between the mids at 8ohm and Fr (resonance fr.) at 128ohm.
I would like to try perhaps and 8 ohm resistor, that should 1/4 the power available (5w), which should bring ca. 6db "equalisation" and go from there. What do I need to be careful about shopping for one? What is the power rating that is suitable? 5w, 10w, 20w? Do i need to worry about (parasitic?) inductance of a resistor? Does it make sense to go for high quality 20w mundorf resistor, vs something generic (in terms of sonic degradation)?
Help very appreciated!!
I wouldn't worry too much about anything except power handling at this stage considering we're talkng speakers.. and I'd get a few values of simple wirewound resistors.
Is my maths correct when I assume i need 2-3W (say 5w) power handling spec for resistors between 2-8ohm, given that the drivers minimum impendance is 8ohm and amp's power rated at 20w into 8?I wouldn't worry too much about anything except power handling at this stage considering we're talkng speakers.. and I'd get a few values of simple wirewound resistors.
It doesn't hurt to get higher power handling than necessary. It is somewhat uncertain what they'll actually dissipate.
10w ok? something like these? Sorry for being anal, but want to make sure I'm not missing something obvious (to you, but not me).It doesn't hurt to get higher power handling than necessary. It is somewhat uncertain what they'll actually dissipate.
I have 2 ways I approach a situation like this. Either can be OK.
a) Use an oscilloscope to watch while I listen.
b) Make an educated guess, then monitor for signs of heating. Feel the resistor after listening to something louder than usual. Keep the resistor where you can see if it releases smoke. Keep it elevated from flammable surfaces.
a) Use an oscilloscope to watch while I listen.
b) Make an educated guess, then monitor for signs of heating. Feel the resistor after listening to something louder than usual. Keep the resistor where you can see if it releases smoke. Keep it elevated from flammable surfaces.
Ok, I'll give it a go. Also today I have seen something, that i cannot un-see. A smaller sibling of the driver in question, deiven direcly from a Mojo dac/amp:
I happen to have the mojo2 as my main dac and now i'm just extra curious to try this. Rob Watts, Mojos chief designer even wrote this:
Though, output impedance is also vanishingly low and bringing resistors in might kill the little power it has.
But this did get me thinking about more powerful headphone amps, some of which have quite great designs and high output impedances. Some probably work even as "propper" current-source amps?
I happen to have the mojo2 as my main dac and now i'm just extra curious to try this. Rob Watts, Mojos chief designer even wrote this:
5. Mojo is designed to drive loudspeakers. You will be amazed hearing it fill the room with beautiful sound using efficient 8 ohm horn loudspeakers.
Though, output impedance is also vanishingly low and bringing resistors in might kill the little power it has.
But this did get me thinking about more powerful headphone amps, some of which have quite great designs and high output impedances. Some probably work even as "propper" current-source amps?
So you are saying 12,5 W for an 8 ohm resistor, given that fostex are 8 ohm and the amp is rated at 20w into 8 ohm, correct?ROT for industrial power apps = resistor rating 2.5x min. power supply rating, so see no reason why it's not valid at audio power rated apps.
- Home
- Loudspeakers
- Full Range
- Best Amp & cables for Fostex FE206En BLH