Hypex Ncore

Status
Not open for further replies.
@back The spikes I'm talking about are in the time domain. I was trying to explain to Juhleren that these don't happen on speakers that are impedance corrected (such as yours, apparently). The story is this: imagine a speaker with a strong impedance maximum. Now run the speaker with a sinewave at that frequency. Current will be minimal because the back EMF is in phase with the amplifier output voltage and has nearly the same amplitude. Now suddenly reverse the phase at the peak of the sine wave... That creates a current spike which is much larger (up to twice as large) as the maximum current you would get with a continuous sinewave at an impedance minimum. The worst case point is the impedance maximum, not the minimum.
Now what happens when you place a conjugate network in parallel with the woofer is that this network will deliver this peak current, thus making the job easier for the amp. If the impedance measures like a resistor, it will behave like a resistor, no matter what kind of signal you feed it.

@juhleren: the module, as it stands, current-limits at 24 amps.
 
@back The spikes I'm talking about are in the time domain. I was trying to explain to Juhleren that these don't happen on speakers that are impedance corrected (such as yours, apparently). The story is this: imagine a speaker with a strong impedance maximum. Now run the speaker with a sinewave at that frequency. Current will be minimal because the back EMF is in phase with the amplifier output voltage and has nearly the same amplitude. Now suddenly reverse the phase at the peak of the sine wave... That creates a current spike which is much larger (up to twice as large) as the maximum current you would get with a continuous sinewave at an impedance minimum. The worst case point is the impedance maximum, not the minimum.
Now what happens when you place a conjugate network in parallel with the woofer is that this network will deliver this peak current, thus making the job easier for the amp. If the impedance measures like a resistor, it will behave like a resistor, no matter what kind of signal you feed it.

@juhleren: the module, as it stands, current-limits at 24 amps.

it`s clear now.

i agree corrected speakers are easier load.

mine are just inefficient.

julf i have 6mm.
 
At VLF your best bet is to throw more surface area at the problem, not more heat.
Yes, but only if the surface area is backed by more box volume.
my findings are that at 200hz where the xover point is 60% of the power needed is at mid high section.
theory confirms that as i found out that the 50% xover point is 300hz.
That’s probably true if the transduction efficiency is equal across the band, but even using bass drivers with the same nominal efficiency as mids/tweeters, the efficiency at the low end is almost always limited by box volume.

… imagine a speaker with a strong impedance maximum. Now run the speaker with a sinewave at that frequency. Current will be minimal because the back EMF is in phase with the amplifier output voltage and has nearly the same amplitude. Now suddenly reverse the phase at the peak of the sine wave... That creates a current spike which is much larger (up to twice as large) as the maximum current you would get with a continuous sinewave at an impedance minimum. The worst case point is the impedance maximum, not the minimum.

I don’t see the relevance in practice; such a sudden reversal cannot occur in a LP-filtered bass driver - correct?
 
Thanks for the comments, Bruno.

Most often i find impedance correcting filters to be employed in mid and tweeter slopes rather than in woofer slopes. The speaker in question seems to use a another strategy too which is to parallel the 4ohm coil with a low impedance coil to utilize the FS peak rather than just level it and burn the current to heat. Unless you by "a conjugate network in parallel with the woofer" mean this parallel VC arrangement of course :)

Of course even more off topic, but most sane speaker builders would instead use a reflex port tuned to match the in box impedance peak of the woofer thus limit excursion and increase sensitivity while reducing the impedance peak. This would of course shorten the hunt for an adequate amplifier ;-)

What I can´t get my head around is how a 3ohm impedance speaker is forcing a UCD700 in its knees eventhough its sensitivity is only 87dB. The UCD should be able to deliver near full voltage unless something fishy is cooking around in that speakers dynamic behavior -even though it in theory should behave like a resister as which is measures like. We are talking peaks of around 1600W!! I can´t imagine how those drivers can eat that unless most of it is lost in the (highly engineered) filters or it is in fact the speaker that struggles at those power levels...
 
Thanks for the comments, Bruno.

Most often i find impedance correcting filters to be employed in mid and tweeter slopes rather than in woofer slopes. The speaker in question seems to use a another strategy too which is to parallel the 4ohm coil with a low impedance coil to utilize the FS peak rather than just level it and burn the current to heat. Unless you by "a conjugate network in parallel with the woofer" mean this parallel VC arrangement of course :)

Of course even more off topic, but most sane speaker builders would instead use a reflex port tuned to match the in box impedance peak of the woofer thus limit excursion and increase sensitivity while reducing the impedance peak. This would of course shorten the hunt for an adequate amplifier ;-)

What I can´t get my head around is how a 3ohm impedance speaker is forcing a UCD700 in its knees eventhough its sensitivity is only 87dB. The UCD should be able to deliver near full voltage unless something fishy is cooking around in that speakers dynamic behavior -even though it in theory should behave like a resister as which is measures like. We are talking peaks of around 1600W!! I can´t imagine how those drivers can eat that unless most of it is lost in the (highly engineered) filters or it is in fact the speaker that struggles at those power levels...

going with bass reflex would require double box volume.

the efficiency of the speaker is 87db with 2.83 volt so 1 watt will be

84db.

on top of all these my ucd700s are powered a bit at the low side 75volt

so they can`t deliver the full 700watts but something like 600-630 watts.
 
back,
I deliberately said sensitivity and not efficiency. But yes peaks in the area of 1600W (70V@3R) is only translated into acoustic output with an efficiency about 84 (if the speaker was 4ohm, which it isn´t) so we are really talking about 83dB if we are arguing over a trifle here :)
A 1000W peak should still give you 113dB peak. If not your speakers are probably compressing due to heating of VCs and motor systems.

If you find the UCD2k with more grunt and worse distortion to sound better than the UCD700, isn´t the nCore the wrong way to go as it has both way lower distortion and lower power specs too?
If the nCore is happy with 1,5R loads you may find them usable in a bridged config, but remember that also nCore like UCDs seem to perform lower distortion at higher impedances hence you may not achieve much by bridging into so low impedances...
Hmmm, its probably wisest to go with Bruno´s advise here as your speakers seems out of the ordinary and very very different from what i have experience with...

cheers,
 
going with bass reflex would require double box volume.
.

Not really if the woofers were designed for reflex usage (as most are today). Comparatively you will get more output in a smaller box if both woofer, box and tuning is designed properly(hence higher sensitivity and efficiency).

Using a high Qt driver for bass reflex will of course not enable you to use a smaller box but with a larger box you will also go way lower. Closed boxes were the thing back when woofers couldn´t get the Q down. Its religious of course, some just believe in closed boxes and some don´t...
 
back,
I deliberately said sensitivity and not efficiency. But yes peaks in the area of 1600W (70V@3R) is only translated into acoustic output with an efficiency about 84 (if the speaker was 4ohm, which it isn´t) so we are really talking about 83dB if we are arguing over a trifle here :)
A 1000W peak should still give you 113dB peak. If not your speakers are probably compressing due to heating of VCs and motor systems.

If you find the UCD2k with more grunt and worse distortion to sound better than the UCD700, isn´t the nCore the wrong way to go as it has both way lower distortion and lower power specs too?
If the nCore is happy with 1,5R loads you may find them usable in a bridged config, but remember that also nCore like UCDs seem to perform lower distortion at higher impedances hence you may not achieve much by bridging into so low impedances...
Hmmm, its probably wisest to go with Bruno´s advise here as your speakers seems out of the ordinary and very very different from what i have experience with...

cheers,

113db is the max they can give with ucd2k.

if i try more the woofers exceed their xmax.

ucd700 sound better than ucd2k except bass and very loud levels.
 
Which is the same as saying that more power won´t really help you much then....:D


i don`t understand why we continue this conversation.

i don`t need more power i have plenty.

i need ncore with the same power but they will not be available.

if i can afford 4 nc400 i will buy them.

if not i will keep what i have.

after a couple of years (if i will still have a job)i will build something from

scratch and it will be fully active.
 
Status
Not open for further replies.