passive to active conversion: can someone check my logic?

active conversion : please check my logic

  • theory One

    Votes: 1 100.0%
  • theory Two

    Votes: 0 0.0%
  • theory Three

    Votes: 0 0.0%

  • Total voters
    1
  • Poll closed .
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am in the process of converting my Visaton TQWT from passive xo to active analog xo. I shan't go into much detail here regarding the xo, since ive a thread in line level forum concerning that already.

Essentially i have again used Boxsim to simulate the active xo, as i did with the passive version. Xo point is similar, but steeper slopes. Passive is 2nd order LP+notch, 2nd order HP, electrical. The active version is 5th order LP, the odd order element for BSC, and 4th order HP, plus Fifth 'pole' by way of a DC blocking capacitor. Both versions are very close in frequency response, which is good as that was the intention...

However

I have noticed a resonance in the lower midrange, maybe 200 - 300 Hz region. This was never heard in the passive implementation. It could've been there, but i didn't notice. Everything else is great in the active conversion. Treble is cleaner, bass attack and midrange is great. I don't want to faff around with subjective talk, but when u first tried them in active form i was amazed. I thought id overdone the BSC, but i have run freq curves of the filters and i know they perform as per the simulated electrical transfer functions, within 2%.

I have some ideas to the cause of the resonance. Please check my logic, and speak up if you have ideas I haven't covered, or grasped correctly.
 
Last edited:
theory Two

The BSC is overcooked..

I know the electrical curves are good though.

Either way, active xo mkII has a 4 DIP switch for BSC freq. Could try level adjustment, but as i said in pretty confident the responses accurate for the simulated output, since the passive version were good too.
 
I
The active version is 5th order LP, the odd order element for BSC, and 4th order HP, plus Fifth 'pole' by way of a DC blocking capacitor. Both versions are very close in frequency response, which is good as that was the intention...

However

I have noticed a resonance in the lower midrange, maybe 200 - 300 Hz region. This was never heard in the passive implementation.
Probably is the pipe mode that you started to hear when you listened closer after seeking an upgrade.
The steeper crossover would unmask problems that may have been partially attenuated by the more shallow passive, as well as the acoustical crossover is not going to have the same shape any more, changing from passive to active is not like changing the angle of two straight lines.

Post the measurement of the LF only, passive and active, measured equidistant from the speaker and pipe output, then passive and active measured by the pipe output, and the differences should be apparent, "close" only applies properly to horseshoes and hand grenades.
 
;)

Thanks Weltersys,

I thought it was a pipe mode. It was my initial reaction. Very narrow Q, sounding only on a semi tone, hazard a guess at 220hz. I know i should line them..

Its just I'm often too subjective for some others opinions here....so I'm trying to consider that I could have messed something up with the crossover.

If I had a measurement setup and a weekend free on my horizon, I agree it would be nice to measure them side by side. I'm using a single amp to drive one speaker for the moment...so even real A/B tests aren't possible.

5thOrderXO_ELEC.GIF

Simulated Electrical

5thOrderXO_acousticSPL.GIF

Simmed Acoustic

5thOrderXO_acousticPH.GIF

Simmed Acoustic Phase

Measured_Vs_Simmed.gif

Test results Comparison_mkI

Given that its one speaker, and i'm used to a pair....I guess the room modes could be more prominent than they used to be.

In my measured HP response, the Q wasn't quite there, and the rounding up of frequencies and component values meant the poles were too far apart (i think). A few slight changes for mkII to correct that (Just no measured results from that yet. ...yet)
 
Last edited:
Originally Posted by weltersys
changing from passive to active is not like changing the angle of two straight lines.
whats that supposed to mean?
Lillun,

We like to think of passive crossovers imparting a slope (a straight line of a given slope angle per octave) determined by their "order" :
1st order= one pole (capacitor or coil), 6 dB slope per octave slope
2nd order= two pole, 12 dB per octave slope
3rd order =three pole, 18 dB per octave slope
4th order=4 poles, 24 dB per octave

The reality is the acoustic crossover slope using passive components seldom reflects that "order".

In the charts below are examples, a single capacitor (supposedly first order) is placed in series with a woofer, one value results in a 6 dB per octave slope, another value capacitor (supposedly first order) results in a 24 dB per octave slope. That is a measured, not simulated response.
The slope angle is completely different, as well as the smaller value capacitor causes a 3 dB rise in response before the rapid rolloff.

Using active crossovers, the acoustic crossover slope of a transducer which has flat response in the crossover passband will reflect the "order" of the crossover, it will "look like the line".

Using passive components, the same often is not true, hence my request to see what Mondogenerator's speaker looks like using passive vs. active to make the comparison.

Art
 

Attachments

  • Passive, Active Slopes.png
    Passive, Active Slopes.png
    296.1 KB · Views: 98
In the charts below are examples, a single capacitor (supposedly first order) is placed in series with a woofer, one value results in a 6 dB per octave slope, another value capacitor (supposedly first order) results in a 24 dB per octave slope. That is a measured, not simulated response.

because the drivers do not have flat response correct? and also because with passive crossovers, the impedance of the drivers also affect the slope.
The fact that active crossovers do not depend on the impedance and give a more accurate response is an advantage.
What the OP is trying to do by matching the passive crossover response is pointless. Use the flexibility of the active crossover to tune the speaker to your liking. You can also EQ the response.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.