Question on Active Bi-Amp

[kahlfin1k]

Well, I didn't like the sound when running wires directly from LF drivers to the amps (bypassing stock passive network). I set the slope at 6db on the Ashly but now also noticed the overall sound was just not as good as before, the bass from the LF drivers is not as tight. Seems the LFs are lacking with this setup.

I am not sure what adjustments are necessary to improve this. I will play with the Ashly settings more when I get a chance but does not seem to be much I could play with on Ashly.

Attached (not sure how to do this) is the speakers spec so not sure how to check "slope of the passive" per your suggestion. Can you provide guidance here from the schematic attached?

Thanks.

Don't see anything attached. If you PM me I'll give you my email and you can attach it. Most likely the reason it doesn't sound good is you have the wrong slope on the active as I suspected earlier you've significantly changed the tuning of the speaker. The 3 critical parts of a speaker design are the driver, cabinet and crossover. Change one and the whole house falls down. You can be sure that the fine engineers at JBL chose those three components very carefully.

 

[edwyun]

need to look at the original XOs schematics. are they only XOs or do they perform other functions.

In the end, it will take time to listen and dial them in (crossover points, slopes, etc.).

 

[McNewbie]

I have soom experience with what you are doing here, have and used the XR 1001 xover, etc.
Here is a switch pannel I made to allow or use the passive networks or the active networks
http://www.audiokarma.org/forums/showthread.php?t=580337

You will have to "dial" in the Asly to get the sound you are looking for. Passive nets works, especially on the low end are real power robbers. Get the right amp(s) and the Ashly set up correctly and you ncan effectively improve you sound. The Ashly if I am correct has a slope of -24/db/oct, probably -18 more than your passives.
I am sure you can get better sound with the right amp and some "me" time with your Ashly. You might get hold of "Tomlinmgt"- I think he has one of these and can help you get set up.
DC

Thanks much dc270 for chiming in to share your experience. I am using mono amp (mc 501) for each LF driver so should be enough. As for the link you provided, it says i don't have the right permission to access it so not sure what that means. I want to spend more time playing with the Ashly (when I get a chance again) as I have heard good feedback on them from many others. I have no doubt that the issue is me lacking the know-how so will take more learning to get it dialed in right.

Thanks again.

 

[dc270]

Should this help, slope is basically how fast or slow the given network decays as it hands off signal to the next driver. Some networks, passive or active, decay or "roll off" slowly, example:

at 500 hz the slope/roll off is set to decay 6db for every octave past 500hz. So the first octave of 500hz is 1000hz, Your power drop, roll off, decay at 1000 hz will be 6db down or less than what it was at 500hz. They call this a "first order" network, a second order network is twice as steep- -12db per octave, 3rd order is -18db/oct, etc.

These xover networks are based around proven design formulas, that given a drivers impedance (at the xover point), what "order" of slope/decay you want, and a few other considerations, will give you what value resisters, inductors and capacitors you will need to built a given network.

You could quite possibly isolate the woofer section of your passive network and measure/read the caps, coils etc to figure the slope and xover freq used by the factory. Duplicate that will your Ashly and there you are! Easier saID THAN DONE PERHAPS.......
DC.

 

[grey]

I set the slope at 6db on the Ashly

Hi,

You're locked in to 24dB slopes with this model. The "response control" adds to the confusion. The manual has a note regarding this.

The passive filters are 12dB slope. I've added an edit in my previous post to note that L3 is also part of the HP filter.

I've not attempted to have active(or line level filters) and passive components do the same job. It's another level of complexity aside from the losses mentioned. Is there any way that can sound favorable in the end? I don't know, but it would be counterintuitive for me to try. If I did somehow need to go down that path I would try to work it out with DSP. At that point there is still power and dynamics left on the table.

I would also check passive crossover mods. (if that is to be done) for severe impedance dips. I wouldn't predict any here, but it's the right thing to do.

 

[dc270]

Sorry the link bombed on ya- it's for those paying members of the AK group.

None the less what I was wanting to show you was a couple of switch terminal plates I designed. With them you could isolate any of a speakers 3 drivers and allow either the passive network OR an active filter to driver them! The grounds are not looped, but isolated so everything plays well together.

My Cornwall used this switch box. I had the woofer using an active network/ seperate amp, and the mid/ twtr using the passive network- all switchable with 3 double throw toggle switches and 3 sets of inputs, It worked very well, just had to power it down before throwing switches.
DC

 

[IBJamon]

If the passive network has only a 6db (first order) slope for that driver, then you will get much less benefit out of switching to the active than other, higher order networks. The less components, the less power/phase/etc. sapped out of the amplification on the way to the driver. You do get the benefit of a higher damping factor, but most of the other benefits are less so.

Unfortunately, it is relatively rare to find a well designed speaker that actually uses first order networks (not to say that they are bad!) because few drivers are that well behaved. If you can figure out the crossover schematics, it would make configuring the active a lot easier.

IBJamon

 

[McNewbie]

If the passive network has only a 6db (first order) slope for that driver, then you will get much less benefit out of switching to the active than other, higher order networks. The less components, the less power/phase/etc. sapped out of the amplification on the way to the driver. You do get the benefit of a higher damping factor, but most of the other benefits are less so.

Unfortunately, it is relatively rare to find a well designed speaker that actually uses first order networks (not to say that they are bad!) because few drivers are that well behaved. If you can figure out the crossover schematics, it would make configuring the active a lot easier.

IBJamon

thanks IBJamon. Here's all the specs for these speakers - not sure what other info is needed. Based on this, I just need to set the xover point for the lows at 350 Hz. Is this not correct? I will have time to play with it more on the weekend.

Any input/suggestion on what I should try next is much appreciated. Thanks.

 

[McNewbie]

Sorry the link bombed on ya- it's for those paying members of the AK group.

None the less what I was wanting to show you was a couple of switch terminal plates I designed. With them you could isolate any of a speakers 3 drivers and allow either the passive network OR an active filter to driver them! The grounds are not looped, but isolated so everything plays well together.

My Cornwall used this switch box. I had the woofer using an active network/ seperate amp, and the mid/ twtr using the passive network- all switchable with 3 double throw toggle switches and 3 sets of inputs, It worked very well, just had to power it down before throwing switches.
DC

Thanks. I want to leave the upper trio speakers at stock and only want to mess with the LF drivers to get tighter bass out of them using active oxer. this is my main goal. So right it seems I just need to play with the Ashly adjustments to get them dialed in. Will play with it some more on the weekend.

Here's the speakers curves I captured this past weekend. Not sure what this tells me yet.

 

[dc270]

your passive xover slope appears to be a -12/oct type, set around 350hz if the graph shows the passive network. As pointed out before, your Ashly is double the slope of your passive networks. Given that the Ashly is so steep, -24db/oct, you might tweak the "responce" & a little around the freq. It's kind of an apples/ oranges mixing here.

Also note the gain on your woofer is pretty hot between 350hz & 100hz you have a 10db gain! That's like triple the power differance. I would settle down that "humping a bit more if possible. Is that graph of the passive or the active? Also the room can cause unusual gains (cabin gain in cars do that) consider that as well.

Did this set of JBL's once come with an optional subwoofer, perhaps for a good reason!

I like what you are trying to do here and think you could even pull it off!
DC

 

[IBJamon]

Do you plan on using the output of the highs from the active x-over for the mid/high section of the speaker? As I noodle that, I'm not sure what to suggest. If you could bypass the high pass on the midrange driver and then use the active for that, that would probably be ideal. Then you wouldn't be doubly applying a high pass on the midrange by using both the active and passive high pass, which would likely double up the slope there (correct me if I am wrong). Otherwise, if you just connect an amp directly to the top post, but not through the active crossover, the high pass slopes won't match, but would probably be additive with the 24db and 12db slopes. Maybe it will be an audible issue, or maybe not; it's hard to say. I may be wrong about this; I desire to be corrected!

I could hazard a guess on what you would need to bypass (probably the 2mH in-line coil and the C5/C6A/C6B capacitors) but unfortunately it's just that - a guess. We would probably need someone better at crossover design than I to confirm with certainty.

Best of luck on your experiment! I find bi-amping a fun this to mess with.

IBJamon

 

[McNewbie]

your passive xover slope appears to be a -12/oct type, set around 350hz if the graph shows the passive network. As pointed out before, your Ashly is double the slope of your passive networks. Given that the Ashly is so steep, -24db/oct, you might tweak the "responce" & a little around the freq. It's kind of an apples/ oranges mixing here.

Also note the gain on your woofer is pretty hot between 350hz & 100hz you have a 10db gain! That's like triple the power differance. I would settle down that "humping a bit more if possible. Is that graph of the passive or the active? Also the room can cause unusual gains (cabin gain in cars do that) consider that as well.

Did this set of JBL's once come with an optional subwoofer, perhaps for a good reason!

I like what you are trying to do here and think you could even pull it off!
DC

Thanks again. Let me clarify a few things. LF response curves are with Ashly over and Highs are from stock passive network (see setup below). LF measurement is with the mic in front of the driver about 3 feet away to isolate any room response. For the top trio, I set the mic in front and same level with the tweeter.

As for the "optional subwoofer", it's for the earlier version which L250 which has totally different circuit than this one (from what I understand which is not much). Thanks.

 

[McNewbie]

Do you plan on using the output of the highs from the active x-over for the mid/high section of the speaker? As I noodle that, I'm not sure what to suggest. If you could bypass the high pass on the midrange driver and then use the active for that, that would probably be ideal. Then you wouldn't be doubly applying a high pass on the midrange by using both the active and passive high pass, which would likely double up the slope there (correct me if I am wrong). Otherwise, if you just connect an amp directly to the top post, but not through the active crossover, the high pass slopes won't match, but would probably be additive with the 24db and 12db slopes. Maybe it will be an audible issue, or maybe not; it's hard to say. I may be wrong about this; I desire to be corrected!

I could hazard a guess on what you would need to bypass (probably the 2mH in-line coil and the C5/C6A/C6B capacitors) but unfortunately it's just that - a guess. We would probably need someone better at crossover design than I to confirm with certainty.

Best of luck on your experiment! I find bi-amping a fun this to mess with.

IBJamon

At this moment, I have not tried that (using the output of the highs from the active x-over for the mid/high section of the speaker) but I will probably give it a try just to see how it sounds like. I have tried this with the LF drivers though and I really liked it. Right now, it's more about learning then listening to music.

I am not sure what "bypassing in-line coil and the C5/C6A/C6B capacitors" do though? what is the anticipated result?

Thanks.

 

[IBJamon]

If you bypassed the high pass in the lowest non-woofer driver and used the active in its place to handle the high pass duties, then the active would take care of matching those drivers for you. Might have a good result, maybe not.

I was guessing, based on my limited knowledge, which components were part of the high pass circuit for that lower midrange driver. Typically, a 2nd order high pass crossover has a coil in series and a cap in parallel, which makes up the filter, and based on the schematic you provided, those seemed to fit the bill. But I would get another opinion before bypassing anything.

IBJamon

 

[grey]

Typically, a 2nd order high pass crossover has a coil in series and a cap in parallel, which makes up the filter
IBJamon

That's pretty close, but this describes a low pass.
http://ccs.exl.info/calc_cr.html#second

Sorry if my previous posts were not making sense. I was attempting to point out C3/L3 as the main HP components.

 

[grey]

Regarding the subwoofer, it sold for some time. It did start as an accessory with the L250.

1992 B460 Subwoofer $2,150 ea
1992 250TiBQ Ti Series Ltd. Edition $2,499 ea

 

[IBJamon]

That's pretty close, but this describes a low pass.
http://ccs.exl.info/calc_cr.html#second

Sorry if my previous posts were not making sense. I was attempting to point out C3/L3 as the main HP components.

D'oh! Thank you for correcting me. I did indeed have it backwards. See? I told you not to take my word for it.

Do what grey says.

IBJamon