Well, that is two changes at once, but if you simply changed from high impedance amp to low impedance amp, the difference would be exactly related to the woofers impedance curve.
Goint the other way, if you know the system response for constant voltage (low Z) and you imagine the impedance curve is an equalizer curve with a bump at resonance and a continuous rise at high frequencies (inductance rise), then that is exactly the difference you get by moving to constant current.
Constant current gives an input voltage that traces the impedance curve.
David S.
I meant with voltage drive. With current drive, I completely agree you'd get a peak in the response where it ramps up the voltage to try and put the same current through.
This would explain the use of Zobel networks along with valve amplifiers (aren't they more current drive?), otherwise the driver's inductance would lead to greater output at higher frequencies.
Not a real problem with voltage drive, which most speakers seem to be designed around.
Yes, this is my major concern (or I should say confusion).
A. boost by line level or digital (EQ) -> low Z amp -> low Q speaker
B. low Z amp -> high Q speaker
C. high Z amp -> low Q speaker
All three can be tuned to have similar (or identical) frequency response. I've tried case A & C, but not B (because I don't have such woofer).
They operate each 'stage' in different ranges, e.g., huge voltage gain variations along the amplifications in case A. It 'overcomes' the LF rolloff by applying high voltage into the high impedance portion of the woofer (under 100 or 80Hz normally).
OTOH, high Z amp (intrinsically) delivers relatively higher current into the woofer in its high impedace LF range without severely amplified front end voltage signal.
And in the comparison between case B and C, I think the major difference would be in the motor. Weaker motor of the high Q woofer might probably in more trouble in the magnetic field modulation (by the VC), if no other mean is taken to overcome this.
Eventually they must get identical "brake force" to have the same Q (if all other factors being equal). In case C (low Q driver), the fixed magnetic field is strong, while the VC portion (of the electro-magnetic brake) is tuned 'softer' by the output impedance of amp. In case B (high Q driver), the magnetic field is proportionally weaker in the whole motor. So they operate each part of the motor differently (, thus different distortion characters).
These are my understanding. Please correct me if something wrong.
And as mentioned, I can't measure them and I don't know what to measure. I just like the sound of case C better. Of couse I can be wrong.
I always wondered if driving a single speaker with a current source would make the speaker more linear. Remember that flux is proportional to current, not voltage, and flux is what moves the cone.
Bobby Dipole
I always wondered if driving a single speaker with a current source would make the speaker more linear. Remember that flux is proportional to current, not voltage, and flux is what moves the cone.
Bobby Dipole
A good question. I did some measurements of woofer distortion with an old Fisher tube amp with variable output impedance. I used a compressor loop to keep response flat, for a fair comparison. There was no difference with bass distortion for varying output impedances. There was, however, lower distortion in the midrange in the region where flux modulation in the ferrite magnet has an effect. Constant current gave lower distortion.
David S.
Hi,
I would point out that a amplifier with high output Z will have to produce
pretty much the same voltage swings as for cases A, for C. For C it is a
lot more efficient if Z is active rather than an external resistor, e.g. see :
Variable Amplifier Impedance
I agree that in the end the "stronger" motor should be better / cleaner.
Case B I'd say is best when you don't want to boost bass, valve stuff ?
rgds, sreten.
Back to the OPs question:
The vast majority of amplifiers try to emulate a perfect voltage source. They take in a voltage from a source (FM, CD, iPod, whatever) and amplify it. The amplified voltage is put across the speaker's impedance. Voltage across impedance creates a current flow through the voice coil [google "speaker cross section"]. Current flowing in a magnetic field results in a force. In this case the current flows through the coil, which moves, and since it is attached to the cone-the cone moves!
For car, home, and pro applications it is still easier to provide a high voltage drive than it is to supply high current. A cheap amp may deliver 20 volts = 100 watts, but only into a pure 4 ohm resistor test load. If asked to deliver 20 volts into a 4 ohm capacitive or inductive load, it may clip or blow up entirely. It takes a much beefier amp to maintain the output voltage into all different loads.
The vast majority of amplifiers try to emulate a perfect voltage source. They take in a voltage from a source (FM, CD, iPod, whatever) and amplify it. The amplified voltage is put across the speaker's impedance. Voltage across impedance creates a current flow through the voice coil [google "speaker cross section"]. Current flowing in a magnetic field results in a force. In this case the current flows through the coil, which moves, and since it is attached to the cone-the cone moves!
For car, home, and pro applications it is still easier to provide a high voltage drive than it is to supply high current. A cheap amp may deliver 20 volts = 100 watts, but only into a pure 4 ohm resistor test load. If asked to deliver 20 volts into a 4 ohm capacitive or inductive load, it may clip or blow up entirely. It takes a much beefier amp to maintain the output voltage into all different loads.
I want to thank all of you for all your answers.
This has all been very informative and please forgive my silence- used to have a hard time shutting me up
(I did a long post recntly but it all got lost)
I've been designing an amplifier, and really wanted to learn and see what the "speaker" side of the world look at this...Seeing as this is what I'll be driving.
I have made a couple of amp circuits that result in two different outcomes. The mosfet seems to drive the speaker with more volume whereas the BJT with the same voltage doesn't seem to drive the speaker as much.
However, when I put my multimeter to the speaker wires I got very little voltage...mV's. I couldn't get an amperage (bad multimeter).
I have researched this and the most I've gotten (before here) is 2.23V put through an 8ohm speaker results in 1Watt of power. I have 90db speakers, and that should be giving me 90db of volume at 3' away.
That is not happening.
Again, thank you and I will frequently respond to your answers.
This has all been very informative and please forgive my silence- used to have a hard time shutting me up
(I did a long post recntly but it all got lost)
I've been designing an amplifier, and really wanted to learn and see what the "speaker" side of the world look at this...Seeing as this is what I'll be driving.
I have made a couple of amp circuits that result in two different outcomes. The mosfet seems to drive the speaker with more volume whereas the BJT with the same voltage doesn't seem to drive the speaker as much.
However, when I put my multimeter to the speaker wires I got very little voltage...mV's. I couldn't get an amperage (bad multimeter).
I have researched this and the most I've gotten (before here) is 2.23V put through an 8ohm speaker results in 1Watt of power. I have 90db speakers, and that should be giving me 90db of volume at 3' away.
That is not happening.
Again, thank you and I will frequently respond to your answers.
Last edited:
In practice, when working with OB, the voltage swing requirement is not that much. Far less than you might think.
Eventually the speaker driver on OB only needs that little power to reach Xmax (and give you more distortion than you need).
My OB bass is driven by LM3886 chip amp provided with +/- 41V rails. In a test, it rushed over Xmax @ 20Hz (and gave me distortions more than I can swallow) for a short while. No clip at all, the chip amp stayed stone cold. No sweat.
So it's the woofer gives up first. (a pro sound 18"er rated 500W)That's a 20 Ohm (active) output impedance driving a 0.3 Qts woofer on a small OB.
I think there'd be only very very few case(s) that really need both very large voltage swing and very high output impedance (closer to pure constant current). That's sick and should be abandoned no matter what
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.