The Journey Begins - Passive to Active

[Dan Twomey]

Once again, many thanks for all your work involved in helping me get this sorted out.

Regards,
Dan

 

[Dan Twomey]

Just posting some of my experiments here. I did some listening test with the tweeters at 2K and LR2 as recommended in the Fountek data sheet. I noticed a little bit of harshness
that bugged me was gone so I decided to take some measurements. The first set of measurements were both LR4 at 2500Hz with the woofers inverted. The second set of measurements
were LR4 woofers inverted at 2500Hz and the tweeters LR2 at 2000Hz. After reviewing the measurements I noticed that the LR4+LR2 combo has a dip that the LR4+LR4 doesn't. Not I'm inclined
to go back and test the LR4+LR2 with the woofers not inverted.

Regards,
Dan

 

[jtalden]

The new XO filter settings impacts the delay timing needed for good SPL support. It is possible that the new XO settings will work better than the initial settings. It is also possible that this new XO filter setting will not work as well. I can run the needed analysis of this new XO setup if the needed measurements are posted. That way we can see if this is a positive change or not.

 

[Dan Twomey]

The new XO filter settings impacts the delay timing needed for good SPL support. It is possible that the new XO settings will work better than the initial settings. It is also possible that this new XO filter setting will not work as well. I can run the needed analysis of this new XO setup if the needed measurements are posted. That way we can see if this is a positive change or not.

I'm going to do listening tests then measurements of LR4+LR4 inverted and LR4+LR2 non-inverted. After that I'll do separate measurements of both with the usual acoustic timing reference activated.
After getting the 'best' possible performance with strictly XO settings, I will then look at EQ.

Regards,
Dan

 

[jtalden]

The LR4+LR2 option is not a fair comparison if its timing is not set favorably.

 

[Dan Twomey]

With the LR4+LR4 at 2500Hz the evidence is abundantly clear. Inverting the woofers smooths the response at the crossover point.

Regards,
Dan

 

[Dan Twomey]

Included in this set of measurements with timing reference enabled is:
LR4+LR4 at 2.5K non-inverted
LR4+LR4 at 2.5K inverted
LR4+LR2 with LR4 at 2.5K and LR2 at 2K non-inverted
LR4+LR2 with LR4 at 2.5K and LR2 at 2K inverted

Regards,
Dan

 

[Dan Twomey]

Wow! As we can see with the green line that inverting is not helpful with the LR4+LR2 combination.

Regards,
Dan

 

[jtalden]

Measurement 12 is a defective measurement. Look at its impulse response compared to the others. Something went wrong. All the LR2 measurements are meaningless for SPL until the the correct delay timing is determined and implemented. I see no compelling reason to worry about these other XO options though. Why not just use the original 'LR4+LR4 at 2.5K inverted' combination. The small sag in the SPL can be addressed with EQ. If you like though, I can determine the delay needed on one of them if the correct set of 6 measurement are provided as you did before in Post-21.

 

[Dan Twomey]

Interesting. It appears the the non-inverted LR4+LR2 combination is the flattest of all? Now I have narrowed down my listening tests. Inverted LR4+LR4 versus non-inverted LR4-LR2.

Regards,
Dan

Measurement 12 is a defective measurement. Look at its impulse response compared to the others. Something went wrong. All the LR2 measurements are meaningless for SPL until the the correct delay timing is determined and implemented. I see no compelling reason to worry about these other XO options though. Why not just use the original 'LR4+LR4 at 2.5K inverted' combination. The small sag in the SPL can be addressed with EQ. If you like though, I can determine the delay needed on one of them if the correct set of 6 measurement are provided as you did before in Post-21.

I will redo some listening tests. My one and only attraction to the LR4+LR2 is it seemed to eliminate one flaw I had in listen sessions. The LR4+LR4 had a sharp end to S's in the vocals and LR4+LR2 to solve that problem.
I'm going to accept your very generous advice and start using EQ to see if I can get complete 'satisfaction' from the LR4+LR4 inverted crossover.

Regards,
Dan

 

[jtalden]

I didn't bother to look at the LR2 measurements as the delay timing was not established. You found one that provides better SPL response without even having to change the delay timing. That is very unlikely. The timing must have been very close to ideal just by chance. Sometimes it's just better to be lucky.

Why not just use that LR2 setting for the EQ, particularly if it already sounded better to you. There is nothing magic about the LR4 filters. It is just a popular starting place.

 

[Dan Twomey]

Short term the LR4+LR2 plus a quickie house curve has left me very pleased, not perfect, but very pleased.
Some more measurements in the near future and then onto integrating the sub.

Regards,
Dan

 

[Dan Twomey]

Here's my latest measurements with a Woofer LR4 at 2.5Khz and a tweeter LR2 at 2Khz. These measurements are with a quick house curve which I will post soon.
With the exception of no working sub yet I'd say I'm 85% happy with the current sound in my room. There is still just a touch or harshness on some vocals. Rare but still there.

One last thing I found odd with these measurements is the impulse for both speakers. It looks a bit weird.

Regards,
Dan

 

[jtalden]

The impulse looks normal to me. The L+R impulse should be ignored as the mic is a little off center. We can align the L and R and use REW trace arithmetic to sum or average them to see the combined response at equal distances. Below I aligned and averaged the L,R.

 

[Dan Twomey]

Thanks. In the near future I'm going to try and teach myself how to measure the time alignment of my drivers.

Regards,
Dan