passive to active cross over conversion?

[owenhamburg]

Passive cross overs account for complex speaker interactions, and are tuned to design and cost.
Active cross overs can be far easier tuned due to greater component flexibility.

So should one reverse engineer, a cross over such as,

Zaph|Audio - ZMV5 - MCM / Vifa 5" System

Or just put a 4th (or higher?) order at the cross over point and adjust for levels when converting a speaker to an active design?

Thanks for your thoughts?

[JamesTRexx]

I've converted my Tannoy Sixes 611 to active with a Behringer DCX2496 last year and it made a huge improvement.
Originally the crossover points were 2500 and 400, not sure what low order they were, but now it's 4th order LR at 1.71K and 69 Hz. I felt like that got the most out of the drivers.

After adding a DEQ2496 and doing RTA on them I've never heard top high end systems sound as good as these. Ofcourse, being able to place them in the right spot in the room helped with getting a great soundstage.

[Francec]

I started with a two way Vifa 8" and 1" passive which I thought sounded fairly good. I converted one of them to active and was impressed so I took the next logical (?) step and converted both to active 3 way with an 8" Peerless, 5 1/4 Peerless and the original 1" Vifa tweeter. The Vifa woofers went to heaven .
The end result is so much better than passives and easier to tune. I used Rod Elliott's PCBs for the crossovers, amps and balanced driver for the bridged amp for the woofer.
It is well the effort and, as I said, almost devoid of the problems associated with passive cross-over design.

Frank

[navin]

Originally Posted by owenhamburg

Passive cross overs account for complex speaker interactions, and are tuned to design and cost.

Originally Posted by Francec

It is well the effort and, as I said, almost devoid of the problems associated with passive cross-over design.

Please note that guys like Troels and Zaph use complex networks to compensate for time alignment (ladder networks), baffle step, impedance compensation, driver anamolies, phase, etc. If you are replacing a passive network with an active one the "compensations" used in the passive network should be accounted for.

Components used for the crossover slope are often combined with those used for compensation for example an oversized series inductor with the woofer might be used to compensate for baffle step or the impendance compensation circuit might be compbined with the parallel cap if a Low pass 2nd order network.

Time alignment can be done using DSPs.

In short do not expect the system to sound the same if you are replacing passive networks with active.

[owenhamburg]

Thankyou all for your thoughts.

I guess I will leave Baffle step compensation to later stages of the tweaking. I guess this level of compensation requires a measurement microphone rather than by ear.

As for Time alignment, I had always thought it pointless, but so dramatic was the effect of adjusting this on my second hand Behringer cross over, I suspect it can be very significant now.

I guess its time to tweak, until I am mostly happy, and consider building a measurement set up.

[JamesTRexx]

If you have the funds (and go for the DCX), get the DEQ with a mic (mine was €40) at the same time to connect via digital. It allowed me to set up the speakers without toe-in and it created a huge soundstage in my room because the RTA function compensated (among others) the rolloff of high frequencies off-axis.
Ofcourse, YMMV because my sidewalls provide the necessary reflections. (the complete story is on Affordable high end audio! where I dump my rantings about high end :-) )

[navin]

Originally Posted by owenhamburg

As for Time alignment, I had always thought it pointless, but so dramatic was the effect of adjusting this on my second hand Behringer cross over, I suspect it can be very significant now.

Originally Posted by JamesTRexx

If you have the funds (and go for the DCX), get the DEQ with a mic (mine was €40) at the same time to connect via digital. It allowed me to set up the speakers without toe-in and it created a huge soundstage in my room because the RTA function compensated (among others) the rolloff of high frequencies off-axis.

Once you get TA you wont go back.

[owenhamburg]

Originally Posted by JamesTRexx

If you have the funds (and go for the DCX), get the DEQ with a mic (mine was €40) at the same time to connect via digital.

Thats rather interesting. If I could get the same bargain, I would do the same.

This said I have in my kit,

A crossover: Behringer CX3400 (Analogue)
Amplifiers: 3 X Quad 303's and a Quad 306
A computer: Debian - Not relevant the type.
A cheap DAC: ELV ADA24 (costs 34 Euro)
Speakers: Linn Helix LS150 (Cost 50 Pounds with broken tweeter + 47 Euro for newer tweeters from another later Linn)

So to get the same effect, I was thinking I will get a measuring microphone, and put in a fixed analogue cross over / EQ based on measurements, but now I am thinking about time alignment. I cant imagine how to do this well in analogue circuits.

Once again, thank you all. For now I will play some more till I get a measurement microphone.

[Francec]

In short do not expect the system to sound the same if you are replacing passive networks with active.

No, expect it to sound much better. Once you go active, you won't go back.

One problem I found though is badly produced CDs; some that sounded alright (before active) before don't seem as good but well produced CDs do sound better with active.

Frank