Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Kent Thompson]

I thought I would move a question of mine out of an existing thread and put it here like I should have done to begin with...



The measurement I took was at about 8' the microphone was positioned about mid cabinet (dual 15s and horn) on axis at about the top 15 level below the horn.  It would be nice to know what causes this whether I am measuring wrong or something is wrong with the speaker.

I probably should have had the microphone mid way between the top speaker and horn... I am gonna take another measurement in a couple days when its my turn to run the system again.



It was suggested I take measurements at different places. So I did.
These are from various places on axis andslightly off axis(around 10 degrees), close and far.
The black trace is an average.
the dip is pretty consistent from every position.
What I think I see...at the crossover point there is a pretty good phase shift. Is this causing the dip or something else I am not understanding?

[Timo Beckman]

Without seeing the individual respons of all drivers i'm guessing but @ about 200Hz the phase trace goes up meaning the low end of the system is ahead in time compared to the synchronization point of the analyzer . @ about 900Hz the high driver is starting to kick in . @ the point of the frequency dip both drivers have equal level but are 180dg out of phase (might be 540dg but i don't think that's the case here).
So you might try delaying the low's and maybe polarity revers on the high's but as to the polarity thing can't say for sure untill i see the solo respons of individual drivers...

http://sdrv.ms/LW9ck4

Could you sent me a personal mail . I did a edit on the screen shot but for some reason i can't post it here .

[Paul Tucci]

Kent,

Pick a spot, close is good. Can you get within a few feet of the speaker? On axis, halfway between the horn and the closer of the two woofs.
Measure, document it.
Move nothing.
Disconnect/mute the lows.
Measure the highs, document it.
Move nothing.
Reconnect/unmute the lows.  Disconnect/mute the highs.
Measure the lows, document it.

It is imperative that the TF measurements are all done from the same place in time, so don't go redoing an impulse response when the highs have been muted/disconnected. What we want to see is what is the phase response of the lows and the highs are by themselves so that we can see if differing phase responses are destructive to rather than reinforcing each other. Chances are this is the case, just look at those two smiley face phase traces.

One has choices to make in speaker designs. This appears to be an impulse alignment. The bulk of each driver's range is relatively flat phase, indicating alignment with that offset time.
Another choice would be to give higher priority to the crossover region. This comes at the expense of even more phase misalignment elsewhere.
Another choice would be to make it as flat phase as possible at the expense of DSP and latency.
Make youir choice, pay the piper.

Your best advice will come from us examining the phase traces individually. It's worth the effort.

PT

[Kent Thompson]

Well this is a passive crossover so I can't really measure them separately now. There are connections to bi-amp but, will have to make a cable for that. So for now I guess I am just practicing measurements lol.

[Ryan C. Davis]

Well this is a passive crossover so I can't really measure them separately now. There are connections to bi-amp but, will have to make a cable for that. So for now I guess I am just practicing measurements lol.

you're not off the hook cowboy... disconnect the drivers from inside the cab. It will take some time to do all the measuring but if the data is good it will definitely be worth it.

[Paul Tucci]

Well this is a passive crossover so I can't really measure them separately now. There are connections to bi-amp but, will have to make a cable for that. So for now I guess I am just practicing measurements lol.

You most certainly can measure them separately. As many people have said, ( I read the other posts from the other thread) you have to get inside the cabinet and disconnect post crossover.
Secondly, if you measure in a biamp situation, markedly different from what you are now doing, EVERYTHING changes. You will have eliminated the passive crossover from the equation. That would provide different data, not neccessarily valid to your question.

PT

[Ales Dravinec]

Well this is a passive crossover so I can't really measure them separately now. There are connections to bi-amp but, will have to make a cable for that. So for now I guess I am just practicing measurements lol.

Kent,

You should do as Paul suggested ...

I'm offering a simplification :
Instead of disconnecting the driver(s) you are not using in the measurement, just COVER it (them) with thin sheet of wood ot something ....
Follow other steps exactly as mr. Tucci suggested and report back, please.

Respectfully.

P.s.: Btw, phase response (except for anomaly) is painfully flat .... kinda weird for a box of that rank ... I'd also say that HF polarity needs reversal ...

[Paul Tucci]

...mr. Tucci ...

Kinda formal for this playground, don't you think?

PT

[Tim McCulloch]

Kinda formal for this playground, don't you think?

PT

"Oh Great Guru Tucci" has too many syllables.

Paul, what I find interesting is the uniformity of results regardless of measurement coherence.  Also, since the OPs are measuring boxes with passive crossovers I wonder *why* a manufacturer would ship such products; i.e. I think the OPs need to contact the OEM and ask "is this what I should be seeing?"

I'm yet to be convinced we're looking at valid measurements.

[Kent Thompson]

Kent,

You should do as Paul suggested ...

I'm offering a simplification :
Instead of disconnecting the driver(s) you are not using in the measurement, just COVER it (them) with thin sheet of wood ot something ....
Follow other steps exactly as mr. Tucci suggested and report back, please.

Respectfully.

P.s.: Btw, phase response (except for anomaly) is painfully flat .... kinda weird for a box of that rank ... I'd also say that HF polarity needs reversal ...

Did not think of that! I had thought of opening them up but I did not really want to because they are new and they don't belong to me so this gives me the alternative I need. I will be running sound for the owner of these boxes on occasion so this is also a getting to know the cabinets session for me.

[Kent Thompson]

"Oh Great Guru Tucci" has too many syllables.

Paul, what I find interesting is the uniformity of results regardless of measurement coherence.  Also, since the OPs are measuring boxes with passive crossovers I wonder *why* a manufacturer would ship such products; i.e. I think the OPs need to contact the OEM and ask "is this what I should be seeing?"

I'm yet to be convinced we're looking at valid measurements.

I will contact the OEM and ask. I know I can do that right 
Since measurements are new to me I would not doubt that I could be doing something wrong. Hence my posting here to ask.

[Paul Tucci]

I will contact the OEM and ask. I know I can do that right 
Since measurements are new to me I would not doubt that I could be doing something wrong. Hence my posting here to ask.

My money is on a valid measurement indicating that the acoustic crossover region has issues.
If we could just flip polarity of the top and to see if the current dip raises in amplitude without causing damage elsewhere.
I'm not a doctor, but apparently I play one on the internet. My diagnosos from the pictures presented...I'm seeing a relatively narrow bandwidth of overlap that is equal amplitude and therefore prone to either constructive addition or interference. I'll hazard a guess to say the the combined crossover slopes (the physical passive crossover and the natural rolloffs of the component drivers) are steep. And one is out of polarity in relation to the other IN THAT NARROW BANDWIDTH THAT SHOWS A DIP IN AMPLITUDE RESPONSE.

What is the speaker in question anyways?

PT

[Ales Dravinec]

What is the speaker in question anyways?

PT

http://www.zzounds.com/item--PEVPR12N

... if I've read the original thread correctly .

w/r

[Kent Thompson]

http://www.zzounds.com/item--PEVPR12N

... if I've read the original thread correctly .

w/r

This is a different speaker the model is SP4

http://www.peavey.com/products/download.cfm?file=/literature/specs/00586520_9080.pdf&name=SP_4_Spec_Sheet.pdf

[Ales Dravinec]

This is a different speaker the model is SP4

Ahhh, my bad. Sorry, Kent.

This is now a whole new ball game for me ....

There's no certain way for me to know about LF drivers being paralleled or whether one of them is rolled off lower. The manual doesn't say.

If LF drivers are in fact wired in parallel, you may indeed experience funny behavior, highly dependent on measuring position in vertical plane. Especially in the crossover area where you have now 3 points of origin (sound sources) .... Moving the mic FAR away would lessen the dip if that topology in fact stands.

The 'trick' with covering individual components of the box can be revealing and rewarding enough.

w/r

[Tim McCulloch]

Ahhh, my bad. Sorry, Kent.

This is now a whole new ball game now for me ....

There's no certain way for me to know about LF drivers being paralleled or is one of them rolled off lower. The manual doesn't say.

If LF drivers are in fact wired in parallel, you may indeed experience funny behavior, highly dependent on measuring position in vertical plane. Especially in the crossover area where you have now 3 points of origin .... Moving the mic FAR away would lessen the dip if that topology in fact stands.

The 'trick' with covering individual components of the box can be revealing and rewarding enough.

w/r

This was exactly what I was speculating about in an earlier post.

[Kent Thompson]

Ahhh, my bad. Sorry, Kent.

This is now a whole new ball game for me ....

There's no certain way for me to know about LF drivers being paralleled or whether one of them is rolled off lower. The manual doesn't say.

If LF drivers are in fact wired in parallel, you may indeed experience funny behavior, highly dependent on measuring position in vertical plane. Especially in the crossover area where you have now 3 points of origin (sound sources) .... Moving the mic FAR away would lessen the dip if that topology in fact stands.

The 'trick' with covering individual components of the box can be revealing and rewarding enough.

w/r

Just the sort of info I needed. So I need to make sure to be farther away on these type of cabinets when measuring the whole cabinet.  The only way to tell what is happening is to not only measure the horn and woofers separate but also the woofers individually to see what is happening with them? What distance would  be needed to get a decent measurement on cabinets like these?  Any other hints on the process I should use to measure these?

[Arthur Skudra]

Just the sort of info I needed. So I need to make sure to be farther away on these type of cabinets when measuring the whole cabinet.  The only way to tell what is happening is to not only measure the horn and woofers separate but also the woofers individually to see what is happening with them? What distance would  be needed to get a decent measurement on cabinets like these?  Any other hints on the process I should use to measure these?

As a rule of thumb, I tell my students to place the mic at least 3 to 4 times the diagonal dimension of the front face of the box, and try to center it between the individual components.

[Kent Thompson]

I did get a message back from the manufacturer who echoed what was said about where I measured from. He also gave me a lot of good information on how the cabinet performed and what to expect. It was a very good email I was impressed. I did not expect them to be this forthcoming with information. He was also concerned about what looked like a polarity reversal on the horn. Something I will have to check next time I have access to the speakers.

[ProSoundWeb Community]