Which Fc of the XOVER will suit better???

[endrek]

Hi, i just build a
221L Bass Reflex box with Fane Colossus 18XB
140 L aprox. Bass Reflex box with Double 12'' (Beyma 12MI100) + CD (Beyma CD755Nd)+ Horn (TD595)+ Filter (Beyma FD212, Fc 1,2KHz).

The sub hasn't got a filter in it, so i use my dBx DriveRack to filter it, and the other signal i send it to the Top box with a HPF.

This is my first box design/build and i'm newbie. I would like some advice about the Fc frequency.
I am thinking of cuting at 100Hz as a lot of people told me.... But now i look at the FRF and it looks "nicer" the FRF if i cut at 125Hz...
Also i use the dBx to make HPF at 100Hz to send the signal to the "Top box".

what do you think??

by the way...the big fall of energy of the "top box" is another thread i have to start..., because i think is something about diference of path, that cancel energy, blabla..., because if move the microphone change's of frequency, blabla....but dunno how to solve it.... ( JUST IN CASE SOMEONE STARTS SAYING ITS TOO BIG THE LOSS OF ENERGY BLABLA)

green line XOVER LPF 100HZ // pink line XOVER LPF 125Hz // red line Top FRF HPF 100Hz



thx for all your help.

gracias from Spain.

[weltersys]

Originally Posted by endrek

Hi, i just build a
221L Bass Reflex box with Fane Colossus 18XB
140 L aprox. Bass Reflex box with Double 12'' (Beyma 12MI100) + CD (Beyma CD755Nd)+ Horn (TD595)+ Filter (Beyma FD212, Fc 1,2KHz).

The sub hasn't got a filter in it, so i use my dBx DriveRack to filter it, and the other signal i send it to the Top box with a HPF.

This is my first box design/build and i'm newbie. I would like some advice about the Fc frequency.
I am thinking of cuting at 100Hz as a lot of people told me.... But now i look at the FRF and it looks "nicer" the FRF if i cut at 125Hz...
Also i use the dBx to make HPF at 100Hz to send the signal to the "Top box".

what do you think??

by the way...the big fall of energy of the "top box" is another thread i have to start..., because i think is something about diference of path, that cancel energy, blabla..., because if move the microphone change's of frequency, blabla....but dunno how to solve it.... ( JUST IN CASE SOMEONE STARTS SAYING ITS TOO BIG THE LOSS OF ENERGY BLABLA)

green line XOVER LPF 100HZ // pink line XOVER LPF 125Hz // red line Top FRF HPF 100Hz



thx for all your help.

gracias from Spain.

The narrow notch at the 1200 Hz crossover point indicates the HF polarity probably needs to be reversed, and some time alignment needed, or both.

If using a DriveRackPA+, the LF may actually be electrically reversed from the HF !

You don't show the combined response of the LF and MF, so determining what the acoustic crossover point should be is not really possible from your post.

Ideally, you want to match amplitude and phase at the crossover point. This often requires crossing the LF and HF at different frequencies, using different filter types (LR, BW, 18, 24 dB etc.) and may require a slight delay on the mid and HF.

You also want to match headroom, running the sub into clip is sonically more acceptable than clipping the mid. Raise the crossover if the mids run out of headroom before the sub.

Headroom is program dependent, vocal oriented music would probably require a higher crossover point than dubstep or D&B. For most program, using pink noise to set headroom for a crossover around 100 Hz is a good start.

Using the PEQ to smooth out the top box LF stepped response may help too.

Have fun!

[Arty]

not familiar with those drivers,
would need T/S parameters,
impedance/phase/SPL graph.
then probably the dip could be killed with a proper XO, and probably your other issues could be dealt with.

[endrek]

Originally Posted by weltersys

The narrow notch at the 1200 Hz crossover point indicates the HF polarity probably needs to be reversed, and some time alignment needed, or both.

You think it will be this??? I tryed moving the microphone, and then the notch moves of frequency.... isn't it due to the diference of path??
If I only reverse polarivity of the HF is ok?? I did change the polarity, but of the 12''s. which changed the FRF, (red Top with 12s reversed, green normal)


wich i think is a balance between "losses" and "flat response"...

Originally Posted by weltersys

Ideally, you want to match amplitude and phase at the crossover point. This often requires crossing the LF and HF at different frequencies, using different filter types (LR, BW, 18, 24 dB etc.) and may require a slight delay on the mid and HF.
You also want to match headroom, running the sub into clip is sonically more acceptable than clipping the mid. Raise the crossover if the mids run out of headroom before the sub.

how can i do this?? i don't know how to do phase graph :S is there a kind of "sonometer" that measures phase??
About matching headroom, how do i do that?? :S I have two American Audio Amplifiers, V4001(4 ohms, 1 khz 1% THD 1090W RMS per channel) for the Tops boxes, V5001 for the subs( 8 ohms, 1 khz 1% THD 980W RMS per channel &nbsp, issn't it matched already???

sorry for my newbie questions...but i'm starting with this
about the...

Originally Posted by weltersys

You don't show the combined response of the LF and MF, so determining what the acoustic crossover point should be is not really possible from your post.

the only thing i haven't tryed is changing the type of filter or the slope, it is always BW with 24dB/oct. All graph shows the FRF of the hole system (sub+Top), the first 3 graph with the 12s inverted , the last one with the 12s inverted and the sub inverted.



what the phase is going on?? jeje

i'm a bit lost with all this..... but this is a FRF graph...not a phase graph.... so....is it releavant?? Dunno how to continue....


thx a lot for your help.

[weltersys]

Originally Posted by endrek

You think it will be this??? I tryed moving the microphone, and then the notch moves of frequency.... isn't it due to the diference of path??
If I only reverse polarivity of the HF is ok?? I did change the polarity, but of the 12''s. which changed the FRF, (red Top with 12s reversed, green normal)


wich i think is a balance between "losses" and "flat response"...



how can i do this?? i don't know how to do phase graph :S is there a kind of "sonometer" that measures phase??
About matching headroom, how do i do that?? :S I have two American Audio Amplifiers, V4001(4 ohms, 1 khz 1% THD 1090W RMS per channel) for the Tops boxes, V5001 for the subs( 8 ohms, 1 khz 1% THD 980W RMS per channel &nbsp, issn't it matched already???

sorry for my newbie questions...but i'm starting with this
about the...

the only thing i haven't tryed is changing the type of filter or the slope, it is always BW with 24dB/oct. All graph shows the FRF of the hole system (sub+Top), the first 3 graph with the 12s inverted , the last one with the 12s inverted and the sub inverted.



what the phase is going on?? jeje

i'm a bit lost with all this..... but this is a FRF graph...not a phase graph.... so....is it releavant?? Dunno how to continue....


thx a lot for your help.

The microphone should be on the ground, and the cabinets laid on their side to get your basic polarity settings correct, with the mic any distance off the ground you will get changing “bounce” dips and nulls.

I don’t know why you decided to invert the mid or the sub polarity, but they all show a dip at the crossover, so they should NOT be reversed from each other. Since you don’t show a combined graph with both of the same polarity, can’t say for certain though.

As it turns out, the DriveRack PA inverts the polarity of all the outputs from the input, the DriveRack PA+ inverts the HF. This appears to be a design flaw.

You wrote :
“I did change the polarity, but of the 12''s. which changed the FRF, (red Top with 12s reversed, green normal)”
that graph clearly shows the 6 dB hole at the 1200 Hz crossover point going away, which makes me think there needs to be a reverse.
The increased dip at around 1800 Hz and peak at 2200 Hz of the red trace are probably due to path length differences which can be addressed with delay.

To measure phase you need a dual FFT type of system like Smaart or TEF or many others. Seeing what the phase response as well as the frequency response makes it easier to set delays and choose what filter type is most appropriate.

That said, you can accurately set delays without a phase reference by first determining polarity (if there is a dip at the crossover area that goes away when one polarity is reversed, the dipped response is out of polarity), then reversing polarity and delaying one output. The dip will either increase or decrease, go for a maximum dip at the crossover point with reversed polarity, then put it back to normal. One caveat, the delays for front loaded speakers should be fairly small. The wavelength at your mid/high is only about 1ms, so you should be experimenting with no more than around 1 ms there.

The wavelength at your low crossover is somewhere in the 7 to 11 ms range (depending on the crossover frequency), if you have added more delay than that, you have added another cycle of delay in, the response can look flat, but the overly delayed portion of the sound will be lagging.

In general for a front loaded system, the LF needs no delay, the mids a short delay 2ms or less, the HF may need a matching delay if the horn used is very short, or less if it's length is long.

Matching headroom can be done by looking at the meters on your amps with music or pink noise after you have set the frequency response the way you want it.
Turn up the sound, if the low cabinet runs out of headroom before the mids, lower the crossover point. If the mids run out first, raise the crossover point.

Edit: Had not realized you are using a passive crossover between mid/high. It still looks like the green trace is polarity reversed, but that may "sound better" than the other peaks when it is "normal".
Passive crossovers are hard to get right...

[Michael Chua]

It's not very clear in your picture but it looks like the TD595 orientation is wrong. The slit should be vertical.

[epa]

Originally Posted by Michael Chua

It's not very clear in your picture but it looks like the TD595 orientation is wrong. The slit should be vertical.

+1 it looks like the "butcrack" is horizontal.
if so ,rotate it.

[endrek]

Originally Posted by Michael Chua

It's not very clear in your picture but it looks like the TD595 orientation is wrong. The slit should be vertical.

+1 jeje...

i'm such a noob!!! (((

[endrek]

Originally Posted by weltersys

The microphone should be on the ground, and the cabinets laid on their side to get your basic polarity settings correct, with the mic any distance off the ground you will get changing “bounce” dips and nulls.

I don’t know why you decided to invert the mid or the sub polarity, but they all show a dip at the crossover, so they should NOT be reversed from each other. Since you don’t show a combined graph with both of the same polarity, can’t say for certain though.

As it turns out, the DriveRack PA inverts the polarity of all the outputs from the input, the DriveRack PA+ inverts the HF. This appears to be a design flaw.

You wrote :
“I did change the polarity, but of the 12''s. which changed the FRF, (red Top with 12s reversed, green normal)”
that graph clearly shows the 6 dB hole at the 1200 Hz crossover point going away, which makes me think there needs to be a reverse.
The increased dip at around 1800 Hz and peak at 2200 Hz of the red trace are probably due to path length differences which can be addressed with delay.

To measure phase you need a dual FFT type of system like Smaart or TEF or many others. Seeing what the phase response as well as the frequency response makes it easier to set delays and choose what filter type is most appropriate.

That said, you can accurately set delays without a phase reference by first determining polarity (if there is a dip at the crossover area that goes away when one polarity is reversed, the dipped response is out of polarity), then reversing polarity and delaying one output. The dip will either increase or decrease, go for a maximum dip at the crossover point with reversed polarity, then put it back to normal. One caveat, the delays for front loaded speakers should be fairly small. The wavelength at your mid/high is only about 1ms, so you should be experimenting with no more than around 1 ms there.

The wavelength at your low crossover is somewhere in the 7 to 11 ms range (depending on the crossover frequency), if you have added more delay than that, you have added another cycle of delay in, the response can look flat, but the overly delayed portion of the sound will be lagging.

In general for a front loaded system, the LF needs no delay, the mids a short delay 2ms or less, the HF may need a matching delay if the horn used is very short, or less if it's length is long.

Matching headroom can be done by looking at the meters on your amps with music or pink noise after you have set the frequency response the way you want it.
Turn up the sound, if the low cabinet runs out of headroom before the mids, lower the crossover point. If the mids run out first, raise the crossover point.

Edit: Had not realized you are using a passive crossover between mid/high. It still looks like the green trace is polarity reversed, but that may "sound better" than the other peaks when it is "normal".
Passive crossovers are hard to get right...

about the microphone at the floor, I am so lucky, that i can go to an anechoic chamber at a friends university to have some fun. And they have just the really old old old old old LabPulse Hardware..... i'll try to find Smaart and enjoy.

really thankfull to your help!! i'll try harder
sorry for so many newbie questions..
i'll try tomorrow again....

muchas gracias amigo!!