Thats the amplifier input impedance, put output impedance there which is probably much less than 1 ohm, basically a short circuit.
Yeah I also suggest that you concentrate on the regular use of passive crossover as filters and what the effect is on frequency response. Perhaps check back to this sideline on later date.
When you are ready with the xo I'm sure there is expert feedback available if you want feedback on it and post it here. I'm not too familiar with passive crossovers, but many others are.
Yeah as TNT already wrote its resonance in the driver, physical property of the driver. You can and should manipulate the response in crossover to make the frequency response of the system nice. There are probably many ways to do this, including some variations of the low pass filter or perhaps additional shelving filter or a notch filter, perhaps something else.
If you wish to play with the "advanced impedance manipulation" this would be one spot to do it 😀 The resonance is kind of an amplifier in the acoustic domain and any current through voice coil around the resonant frequency is emitted (amplified) acoustically by the resonance, including distortion current due to the driver motor non-linearities. Ideally you'd want to make the filter so that it reduces current through the voice coil, a filter that is in series with the driver. Such filter would reduce any current through voice coil at the resonance including any distortion generated by the resonance, or any other distortion current generated in the driver motor like hysteresis. Filter scheme with greatest series impedance would be the lowest distortion alternative. Some other filter could could do the same equalization of frequency response but instead increasing the series impedance for the driver it could lower it, like a shunt circuit which would route some amplifier current past the driver.
Simplest examples of the two are series notch (in parallel with the driver) and a parallel notch (in series with the driver). Series notch (in parallel) would probably show up as higher distortion in acoustic domain than a parallel notch (in series with the driver) because of difference in the series impedance, difference in distortion current.
You can "see" which has lower distortion by checking the impedance from drivers perspective, the filter that has highest impedance around the 1-3kHz would have least distortion current, least acoustic distortion. I'm not sure if a notch is best way to do it, perhaps others know simpler solutions to make the frequency response happen.
It could be that there is not much difference so don't lose your night sleep over it 🙂
Last edited:
Is it not the shape of the bass driver impedance peaks that mostly matters ? Amps are powerfull enough in the mid and heigh...
What about the width of the base of the impedance peak and its heigth too at seing bass driver datasheets ?
What about the width of the base of the impedance peak and its heigth too at seing bass driver datasheets ?
Ahh, i missunderstand....output impedance on my amp is 0,08 ohm and rose very slightly, to 0.11 ohm at 20kHz (Steroephile mesaurement notes)
If i put 0,08 ohm at voltage source it shows this
And when i short the tweeter like this, 0,9 ohm at 1700 hz
read this, https://www.edn.com/loudspeaker-operation-the-superiority-of-current-drive-over-voltage-drive/
Don't get stuck on the title which is provocative but the main content of the article is nice. Simple to understand maths and all.
Last edited:
Hi,
yeah not sure if AMT benefits any of if. The Audax woofers with the impedance blip could so try having the generator in series with the woofers.
Last edited:
A modern approach to system design is based on the present on high precision, transparent EQ as I see it. Everything else is wooden wheel and horse carriage 🙂
Once we have this, sensitivity and frequency response of drivers become non-critical and we can concentrate on more important aspects like avoidance of non correctable anomalies - like high Q resonances (like the one in the Audax) and directivity. Even better if we have one amp per driver i.e. x-over also in DSP and not only EQ.
//
Once we have this, sensitivity and frequency response of drivers become non-critical and we can concentrate on more important aspects like avoidance of non correctable anomalies - like high Q resonances (like the one in the Audax) and directivity. Even better if we have one amp per driver i.e. x-over also in DSP and not only EQ.
//
I would contend that flat (or flatter) impedance and fewer phase variations are an audible advantage in a loudspeaker design. There is no disagreement that frequency response is of primary importance, acoustic phase overlap through the crossover is extremely important, etc. However, consider what demands are placed on the amplifier when the impedance jumps to twice the R and when there is phase warp. Twice the R means double the voltage swing and phase departures from 0 demand more current. If you assume 20 db music peaks, And if you have 88 db/W speakers aiming to hit 100-105 db peaks, the numbers become very large.
Constant load impedance is important for amplifiers with a few tenths of an ohm or more of output impedance. These are tube amplifiers and some no-global-feedback amplifiers. Your Krell will be unaffected by variations in load impedance.
Ed
Ed
Yes, with that impedance curve try to minimize the series resistance of the speaker wires.
The resistance is directly proportional to the round-trip wire length, and also is inversely proportional
to the cross sectional area. So the longer the wire, the thicker it should be for a given series resistance.
re:'do you mean that the components valeu can fluctuate"; "the component "leave" a bit of its "signature"- no, there's no magic here, pure science only
Time to read up on the basics: https://sound-au.com/lr-passive.htm
Time to read up on the basics: https://sound-au.com/lr-passive.htm
I asked if "the component "leave" a bit of its "signature", and you answer
"no, there's no magic here, pure science only
Time to read up on the basics: https://sound-au.com/lr-passive.htm"
Why people buy expensive caps and coils then? 😉
It would have been so nice Lojzek 👍
Use REW and my laptop, but don't understand exactly how I mesaure impedande in REW....Seems complicated.
But i will make 1 meter mesaurements with calibrated mic in the auctual room, and post it here tomorrow.
Use active filter for both woofers and MTM, woofers get "everything" under 340 hz, and MTM "everything" over 340 hz with a passive filter crossed at 2K.
Woofers and MTM each has a stereo amplifier, one for left right MTM and one for left/right woofers...both Krell
For now i have trained to "try understand" passive filter designing, and what changes where regarding to komponent and placement.
And have worked on 2 different MTM filters buy mesaurements, and use a bad responemesaurement and manufacturer's impedance.
Curves in VituixCAD do not fit well with my SPL Trace file, but have been some "benchmarks", even if vituixCAD looks like **** it mesaures decently.
Strangely, it makes a whole 3-4 dB difference on mesaurements over the hole frequency range
Regards John
You wrote this yesterday, but i diden´t understand what you meant with the words " rather they affect how the crossover components react"?