Print

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

[Mark Wilkinson]

Working on tuning kf 695z's.  I know the recommendation is to use a UX processor, and I'm looking to at least borrow/rent one.
 
In the meantime, I've loaded these 695 legacy presets into a Danley/Linea SC-48 http://eaw.com/docs/2_Legacy_Products/Processor%20Settings/KFz_PROCS_rev2.pdf

Not bad, not good.....
My question is: are the legacy presets really all that decent ?

It's puzzling because, I'm getting really damn good results using recommended EQ's, but setting x-over frequencies to be the same between low/mid, and mid/high, along with redoing timing between sections.

thx,  mark

[David Sturzenbecher]

Working on tuning kf 695z's.  I know the recommendation is to use a UX processor, and I'm looking to at least borrow/rent one.
 
In the meantime, I've loaded these 695 legacy presets into a Danley/Linea SC-48 http://eaw.com/docs/2_Legacy_Products/Processor%20Settings/KFz_PROCS_rev2.pdf

Not bad, not good.....
My question is: are the legacy presets really all that decent ?

It's puzzling because, I'm getting really damn good results using recommended EQ's, but setting x-over frequencies to be the same between low/mid, and mid/high, along with redoing timing between sections.

thx,  mark

Mark,
What does the phase response look like in Smaart for the passbands in question?

[Steve Eudaly]

In my experience an email to EAW support is very helpful. Check to see if there are any other suggestions they could make or perhaps some unpublished updates to the settings you have.

Do not call. A call gets redirected to the LOUD Technologies umbrella support and you won't get an EAW specific person like you will with an email.

[Peter Morris]

Working on tuning kf 695z's.  I know the recommendation is to use a UX processor, and I'm looking to at least borrow/rent one.
 
In the meantime, I've loaded these 695 legacy presets into a Danley/Linea SC-48 http://eaw.com/docs/2_Legacy_Products/Processor%20Settings/KFz_PROCS_rev2.pdf

Not bad, not good.....
My question is: are the legacy presets really all that decent ?

It's puzzling because, I'm getting really damn good results using recommended EQ's, but setting x-over frequencies to be the same between low/mid, and mid/high, along with redoing timing between sections.

thx,  mark

Hi Mark,

I had some 650Z’s … hated the 3 way factory settings and did my own, but the factory 2 way settings sounded great.

Someone told me that in some of the early boxes there was a wiring error which had the polarity of the mids flipped in 3 way mode … interestingly enough that was one of the things I did with my settings.
 
It would be interesting to see a Smaart or Systune trace of the factory settings (amplitude and phase).

[Mark Wilkinson]

Thanks guys,

Steve, good to know how to get real help.

David, the first screenshot is the EAW legacy settings passbands' phase and mag.

Peter, next is the whole box, again per EAW's settings.

I haven't tried the biamp settings yet.   
Funny, about the polarity flip on the mid section due to a early wiring error....I thought the same thing till I finally found I'd screwed up a speakon breakout cable.
But I did find all three bands are inverted.

Last two screens are my continuing efforts.  First is passbands, second whole box, like above.

I've raised the low pass frequencies on the low (orig 250Hz) and mid sections (orig 1542Hz) , up to the high pass frequencies on the mid (397Hz) and high (2311Hz) sections, still using LR24's on all, per EAW.
Readjusted levels, readjusted timing...phase fell inline much better.
 
I just don't get why EAW would separate x-over freqs between adjacent bands like that (250 to 397) and (1542 to 2311) ...unless maybe it's an on axis compromise for off axis improvement   Lost I am...

Anyway, added some EQ's on top on legacy settings to get to where i am now...getting happy, it's sounding pretty dang good....help always welcomed ! 

[Rich Frembes]

Mark -

I happen to have an MX8750 and a Linea amplifier here in the office. I just spot-checked settings.  If you set the Linea filter preference to BW (instead of Q) and you enter the (Bandwidth) values shown on the EAW processor setting sheet (again instead of Q) you will get an almost exact match to the MX8750 response.  High and low pass filters are all spot on.  It's hard to screw up a Linkwitz-Riley filter definition -- though I have seen it done.

Now as to efficacy of the published settings? That I do not know.  It's been probably 10-15 years since I've heard a KF695z on an MX8750.  But if you follow the instructions above you'll at least know you have a match to the published settings.

I just don't get why EAW would separate x-over freqs between adjacent bands like that (250 to 397) and (1542 to 1542) ...unless maybe it's an on axis compromise for off axis improvement   Lost I am...

As for the separation between high and low pass frequencies, I see there is some EQ going on around both these crossover points. If the parametric definitions are not spot on things will definitely go awry through crossover. It's quite possible this was a compromise for an off axis improvement.  EAW does generally look at 3D polars when developing settings.

Good luck! Looks like you have things in hand.

[Mark Wilkinson]

Mark -

I happen to have an MX8750 and a Linea amplifier here in the office. I just spot-checked settings.  If you set the Linea filter preference to BW (instead of Q) and you enter the (Bandwidth) values shown on the EAW processor setting sheet (again instead of Q) you will get an almost exact match to the MX8750 response.  High and low pass filters are all spot on.  It's hard to screw up a Linkwitz-Riley filter definition -- though I have seen it done.

Now as to efficacy of the published settings? That I do not know.  It's been probably 10-15 years since I've heard a KF695z on an MX8750.  But if you follow the instructions above you'll at least know you have a match to the published settings.

As for the separation between high and low pass frequencies, I see there is some EQ going on around both these crossover points. If the parametric definitions are not spot on things will definitely go awry through crossover. It's quite possible this was a compromise for an off axis improvement.  EAW does generally look at 3D polars when developing settings.

Good luck! Looks like you have things in hand.

Rich,  you just saved me some work.  The guy with the kf695's has the MX8750, and I told him I would check out the website settings vs the unit.  Thanks !!! One less set of transfers to run....

Kinda off-topic, but have you done much with the Linea's LIR filters?  I'm about to try them on the kf695....
I've looked at the electrical summation of the three bands as a start, using a 100Hz highpass on the low section.
I'm impressed, LR24's without much phase rotation, and only 15ms latency. Bands all timed together automagically !

[Rich Frembes]

Kinda off-topic, but have you done much with the Linea's LIR filters?  I'm about to try them on the kf695....
I've looked at the electrical summation of the three bands as a start, using a 100Hz highpass on the low section.
I'm impressed, LR24's without much phase rotation, and only 15ms latency. Bands all timed together automagically !

No, haven't had a particular need to on this end.  We we traditional IIR high pass and low pass filters on our loudspeakers and let our custom FIRs do all the heavy lifting in the phase realm. We typically achieve sub-5 ms latency. What I am eager to see is output FIR filters in the Linea amplifiers.

The LIR filters do look like they would be fun to play with given an unlimited number of hours, monkeys, and typewriters.   

[Peter Morris]

Thanks guys,

Steve, good to know how to get real help.

David, the first screenshot is the EAW legacy settings passbands' phase and mag.

Peter, next is the whole box, again per EAW's settings.

I haven't tried the biamp settings yet.   
Funny, about the polarity flip on the mid section due to a early wiring error....I thought the same thing till I finally found I'd screwed up a speakon breakout cable.
But I did find all three bands are inverted.

Last two screens are my continuing efforts.  First is passbands, second whole box, like above.

I've raised the low pass frequencies on the low (orig 250Hz) and mid sections (orig 1542Hz) , up to the high pass frequencies on the mid (397Hz) and high (2311Hz) sections, still using LR24's on all, per EAW.
Readjusted levels, readjusted timing...phase fell inline much better.
 
I just don't get why EAW would separate x-over freqs between adjacent bands like that (250 to 397) and (1542 to 2311) ...unless maybe it's an on axis compromise for off axis improvement   Lost I am...

Anyway, added some EQ's on top on legacy settings to get to where i am now...getting happy, it's sounding pretty dang good....help always welcomed !

Your plots look fine, mine had a dip at the mid crossover region because of the polarity issue.
 
FWIW what I did was run the 650’s on a Lake Contour using Lakes 24 dB LR crossovers with linear phase. I also did a lot more out of band EQs so that you could use matching crossover frequencies and get the correct summation, by doing that you get better behaviour in the time domain in most cases.

Early DSP's had a limited numbers of PEQ's, often people used asymmetric crossovers as a compromise to get a good flat frequency response and not use up too many EQs ... as you know its the product of the acoustic an electronic response that determine the actual crossover point. 

What I ended up with looked like what you get with the UX8800 and sounded similar.

P.S. I would really like to know how the Linea SC-48 would work with that other box you have :-)

[Mark Wilkinson]

Your plots look fine, mine had a dip at the mid crossover region because of the polarity issue.
 
FWIW what I did was run the 650’s on a Lake Contour using Lakes 24 dB LR crossovers with linear phase. I also did a lot more out of band EQs so that you could use matching crossover frequencies and get the correct summation, by doing that you get better behaviour in the time domain in most cases.

Early DSP's had a limited numbers of PEQ's, often people used asymmetric crossovers as a compromise to get a good flat frequency response and not use up too many EQs ... as you know its the product of the acoustic an electronic response that determine the actual crossover point. 

What I end up with looked like what you get with the UX8800 and sounded similar.

P.S. I would really like to know how the Linea SC-48 would work with that other box you have :-)

Hi Peter, that all makes sense, thx. 
I've been learning about the value of out of band eqs, as I've been whittling down taps while trying to preserve phase.
And for sure, when I feel I have a good grip on the SC-48 with the PM90, I'll post on the other forum to pick it up....

[ProSoundWeb Community]