Using Adcom GFA-555s as monoblocks to power magnepan 3.3s in active biamp

@allenf1963 You’re asking questions about a topic on which I’m a bit fuzzy. I studied the design of filters for ‘high pass,’ and ‘low pass,’ performed the basic calculations, and became conversant in ‘first order,’ ‘second order’ and the like; ‘notch filters,’ etc. but I don’t have much practical experience with building, or tuning them. I would have go to my textbooks to get a refresher to go into detail. I’m much more comfortable discussing the sort of filtering that takes place in the power supply, trying to convert AC to DC and eliminating ripple.

That said, I can describe the basic function and components. For a passive crossover, the high pass filter takes advantage of the properties of a capacitor to ‘pass AC.’ The amount of capacitance determines the frequency at which the the ‘highs’ pass to the tweeter. Stages of filtration can be built up and at each stage (or ‘order’) the ‘pass frequency’ becomes more and more selective. That is, the slope of the frequency response curve gets steeper and steeper. I think a first order slopes off at 6 bB per octave, and a second order filter slopes off at 12 dB per octave, but I’m not sure, it may be 3 dB and 6 dB. The inductor is used for a low pass filter, and to be honest, I’m not sure how to describe its operation, but again, first and second order filters can be made quite simply. Resistors are also employed in the design of cross-overs, if you find a ‘vintage’ speaker, there is often a knob or two offering to attenuate either the highs or lows: the potentiometer is serving to ‘trim’ the frequencies at which the filter (cross-over) is effective.

So, yes, cross-overs serve to direct frequencies to the appropriate driver within the speaker (bass frequencies to the woofer, high frequencies to the tweeter, those frequencies that are nether high nor low, to the midrange driver.

The issues with passive cross-overs are at least twofold, (1) the fact that they are passive means that some of the signal’s energy will be absorbed by the cross-over components: the capacitors will want to hold onto their voltage; the inductors will want to hold onto their current, the resistors will restrain current flow and will maintain a voltage across them with a polarity opposing the source as long as they are in the circuit (it’s what they do). An alternative is ‘active’ cross-overs, which must have their own power supply, which ‘massage’ the frequencies without using the signal’s energy, but which may also add to the sound a coloration of their own. I’m not sure what components are used in an active cross-over, but I’m assuming transistors and ICs like op-amps are involved.

If you want to pursue this topic further, you probably need to pm me, as I fear we’ve hijacked this thread. We’re not all on the thread’s stated topic.

@allenf1963 I tried responding to you earlier today, but lost my carefully typed reply disappeared when I tried to exit “preview.” I shall attempt a recreation. Most of the discussion on the ‘Electrical Side,’ as you put it, revolves around Ohms Law and it’s many implications for circuits in general, and AC circuits in particular (music that has been converted to an electrical signal constantly varies in voltage and current similarity to how standard AC voltage varies in phase and current). The AC power which feeds your amplifier is converted to DC fairly quickly, and is then modulated by the tubes or transistors which serve to in an ‘amplify’ at the output the signal which has been input. This amplification process can take place in more than one stage, each stage taking its input from the previous stage’s output. Great care is taken in the early stages to remove noise from the (AC) signal and the (DC) power so that the music coming out is at least as good (if not a bit better) as the signal going in.

The devices used to clean the signal and the power, filter out noise and what might be termed ‘resonances,’ also introduce phase changes themselves; capacitors ‘pass AC,’ and ‘store (or block) DC,’ for current will lead voltage through a capacitor; in like manner, voltage will lead current through inductors (or chokes). There are many factors to consider in the design of optimal circuits, and many parts from which to choose. There is no one clear path to success, but many options, each with trade-offs of their own. Yet each may impart a bit of color to the sound and/or tonality of the whole, or be better at some jobs than others, like powering speakers with low impedances.

Ohms Law states that Voltage (in Volts) = Current (in Amps) * Resistance (in Ohms). A corollary of Ohm’s Law is that Power (in Watts) = Current (in Amps) * Voltage (in Volts). Many permutations of this formula are used to solve for one or the other of these quantities.

In AC circuits, “resistance” can also be defined in terms of frequency as ‘impedance’ generally, and “reactance” specifically, as in ‘capacitive reactance’ (Xc) and ‘inductive reactance’ (Xl) [the symbols for reactance are written italic capital X sub small caps C and italic capital X sub small cap L]. Xc = 1 / 2(pi)fC [where (pi) = 3.146, f = frequency, and C = capacitance in Farads]; and Xl = 2(pi)fL [where (pi) = 3.146, f = frequency, and L = inductance in Henries or Henrys]. You will note that capacitive reactance (Xc) decreases with frequency and inductive reactance increases with frequency.

I could go on, but I fear either losing you or boring you (or both). I was where you are twenty years ago when, at age 49, I went to Community College and earned an Associates of Science degree in Electrical Engineering Technology. I have been, and still am, active on the faculty as an adjunct instructor. I encourage you to seek out training in the following two topics ‘DC/AC Circuit Analysis’ and Electronic Devices.’ Textbooks on both topics by one Thomas L. Floyd informed my study (and still does). There are other avenues these days, including interactive online seminars and courses. A YouTube channel, ‘Mr. Carlson’s Lab,’ offers to train you in the field by working your way through repairs of various radios, amps, and test equipment while teaching fundamentals and techniques. YouTuber ‘XRayTonyB’ walks you through extensive repair and restoration of vintage HiFi gear, and another fellow in Arizona likes to repair vintage, tuned guitar amps. His name escapes me at present, but he too walks you through the schematic, explains how the circuit works, and also teaches you some rather crafty techniques to restoring the cabinets as well. Both he and Mr. Carlson don’t hesitate to ‘improve’ on circuits they find with known problems, which I also find very interesting.

To reiterate, I found formal education necessary to answer the questions I was coming up on in my work; plus, I was just plain curious. There is math involved, up to and including Algebra II and trigonometry. While it takes forms of calculus to truly solve some of the equations on the AC side, you can get by with some instructions for using special functions (imaginary numbers comes to mind). I found the exploration of the topic well worth my time; I’ve been involved in electrical/electronic pursuits ever since.

Good Luck

(By-the-way, this write-up only faintly resembles what I wrote this morning at 5:30a, I guess I get a bit loquacious in the afternoon (and it’s not even 5:00p yet).

@allenf1963 If I may comment on @petaluman ’s answer, in terms of Ohm’s Law, let us say you have an old-time filament-type flash light. If you Ohm out the bulb, it shows continuity, just as a speaker coil does, but when you turn your old-time flash light (or ‘torch’ as the Brit’s are wont to call it) on, the filament lights up and gets warm, the warmth is energy (power) being dissipated. The bulb (or more correctly lamp’) is rated in Watts because of the energy (power) it is releasing, mostly in the form of heat, and only secondarily as light. [Light Emitting Diodes (LEDs) emit more light than heat, but that’s another story.]

Now, suppose that this is a magical flash light and capable of illuminating the space in any color of the spectrum and beyond, shining different colors like a loud speaker reproduces different frequencies of the audio spectrum. What @petaluman is saying is that this ability to ‘play different frequencies’ is not without cost. As the wavelengths get longer (that is, frequencies get lower), it takes more power to generate them. For light waves, this means that red light is harder to generate than blue or violet; or, more to the point, that bass is more difficult to generate than treble. In fact, when really low bass notes are generated, unless the amp and speaker are designed for it, the speaker appears as a near short to the amp. Remember, to an Ohmmeter, it IS a short, it is only because the wire is wrapped in the form of a coil, and has AC current running through it at a particular frequency that it exhibits a resistance, or, more properly ‘reactance’ (XsubL). So the speaker coil gets warm and expands, possibly shorting, the fuse, if installed blows, the output transformer (if part of the amp’s design), heats, shorts, and blows, any integrated DSP chip starts complaining and sending messages to the screen before it blows —and the music dies. In our flash light analogy, running the red light all the time burns the batteries up more quickly than light of higher frequencies.

A disclaimer, the flash analogy is used for illustrative purposes only, I have no idea how much energy it takes to produce light of a specific color, although, since red LEDs were produced well before the relatively recent white LEDs I may have it backwards. I was trying to emphasize the longer wavelengths of bass compared to treble.

Hope this helps.

@allenf1963 I forgot the math. Since P = I * V, and since V = I * R, then P = l * (I * R). When you turn the flash light on, and the bulb heats up, it’s resistance increases, and you find the source voltage across the lamp (don’t try this with an LED). If it takes more power to produce red light than blue or violet, then either the resistance has increased and more voltage is required, or the resistance has decreased allowing more current to flow for the same amount of voltage ‘driving’ the circuit. In the case of a speaker’s impedance (XsubL) drop producing the lower bass frequencies, because the speaker offers less resistance at the lower frequencies (XsubL = 2(pi)fL), more current passes through the circuit for the same amount of voltage driving the circuit. (With semiconductors it is easy to enter a ‘death spiral’ in which more current generates more heat which draws more current, which generates more heat, etc. the semiconductor offers less and less resistance until it finally burns itself up). Unless there is a microprocessor monitoring the circuit’s power usage, or appropriate and well-placed fuses in the circuit, part destruction is a real possibility where impedance ‘droops’ occur in the bass region, or if you hook up 4 Ohm (or 2 Ohm) speakers to an 8 Ohm tap (at least in theory, many folk recommend trying your speakers on different taps, just to see if they sound better). In practical terms, it isn’t going to make much difference unless or until you turn the volume up and start to drive the amplifier into clipping or at least start to use significant power (unless there is a short in your speaker or connections).