Active crossover questions

Status
Not open for further replies.
I've looked at MOX-lite properly for the first time on Jan's website. That is what I want, but I want separate boards for each amplifier (each driver).

It looks like 6 dual opamps will do what I want and give me 2 EQs.

Anyone else interested in this approach? It could turn into a group pcb buy, but I have never layed out a pcb before.
 
It seems quite complicated for a system and lots of interconnects. Wouldn't it be easier to have one box per channel with all the chip amps in there? That way you have two boxes only to worry about, less interconnects. short speaker cables?
 
i too am planning a 3-way active system and i will be using a 3886 based GC(Peter's) each for the midrange and tweeter and a 4780 based one for the woofer..
a total of 6 monoblocks....
I can understand the circuits given in siliconchip.com, but, i have no idea as to why u need additional buffers for the amplifiers if they are mounted in seperate boxes....
anyone, care to shed some light on this..

Sachi
 
soongsc said:
It seems quite complicated for a system and lots of interconnects. Wouldn't it be easier to have one box per channel with all the chip amps in there? That way you have two boxes only to worry about, less interconnects. short speaker cables?

Correct, it would be 2 boxes instead of 3, and 4 fewer interconnects.

But my 3 boxes will be flexible they could be reconfigured for other purposes. Also I can use different amps for H/M/L (with different PS requirements) eg LM3875 for High and Mid and maybe dual LM4780 for Low giving 2x120W.

Also I like symmetrical designs and want to use the sidewalls for heatsink. And have as short as possible input and output wiring.

I might start with a Bi-Amp system (passive mid/tweeter crossover) and upgrade to Tri-Amp later.
 
Hmmm. My LM379S based system was been sitting there for a real long time now.🙄

But taken the better resolution active crossovers provide, wouldn't the tweeter be very important? Normally caps effect the high end more than the mids as I recall.

I also had in mind some PWM based chip amp design which is fully digital. JVC seems to have a DVD bedroom system with wooden cones that use this technology. The power effeciency is over 90% so there is little need for heat sinks.
 
sachi said:

I can understand the circuits given in siliconchip.com, but, i have no idea as to why u need additional buffers for the amplifiers if they are mounted in seperate boxes....
anyone, care to shed some light on this..

Sachi


Impedance matching when you have interconnects involved. More complicated. That's why I like the crossover and chip amps in one box.
 
soongsc said:
But taken the better resolution active crossovers provide, wouldn't the tweeter be very important? Normally caps effect the high end more than the mids as I recall.

This was one of my original questions, you can't get a simpler filter than a series cap. And unless you trust the lm3875 datasheet I will be inserting a cap for protection anyway (although it might be different and hence have less effect on signal)?
 
soongsc said:
But taken the better resolution active crossovers provide, wouldn't the tweeter be very important? Normally caps effect the high end more than the mids as I recall.

This was one of my original questions, you can't get a simpler filter than a series cap. And unless you trust the lm3875 datasheet I will be inserting a cap for protection anyway (although it might be different and hence have less effect on signal)?
 
jimbo1968 said:


This was one of my original questions, you can't get a simpler filter than a series cap. And unless you trust the lm3875 datasheet I will be inserting a cap for protection anyway (although it might be different and hence have less effect on signal)?

You're going to use some amp to drive it anyway, even if not the LM3875. But I guess you probably have an exotic cap in mind.

sachi said:

How profound will the effect on the sound be if interconnects are used...

I can hear differences between the various interconnect I have. But in this specific case, maybe you guys can try things and let us know. Comparing a 1M conductor with more than 150pF and one inch of trace with unmeasureable capacitance and inductance and resistance, it seems obvious, but you might have other circuit coupling concerns maybe?
 
buffer / crossover board per amp

A reminder: I would like an active crossover with one board per amp/driver. This would be useful both for people with 2 amps per box or 3 amps per box.

Having studied the mox-lite board details http://www.delta-audio.com/Active_Filter_One/Active_filter_one_ver_1.0_dokumentation.pdf

I've narrowed down my requirement to: IP buffer, HP, LP, OP buffer. Based on the mox-lite and including a square for the PS, this works out to one strip of the mox-lite ie 100mm by30mm.

This assumes the buffers are at opposite ends of the strip, but they could share a dual opamp if it's ok to double back? The board would then be about 80mm by 30mm which matches the chip amp board nicely.

Anyone else interested in something like this? How should I go about proposing it? I could just ask Jens if he'll produce it for me (us?)
 
sachi said:
I can understand the circuits given in siliconchip.com, but, i have no idea as to why u need additional buffers for the amplifiers if they are mounted in seperate boxes....
anyone, care to shed some light on this..

soongsc said:

Impedance matching when you have interconnects involved. More complicated. That's why I like the crossover and chip amps in one box.

Just to confirm > This means I don't need an output buffer on the crossover if it is in the same box as the amplifier?
 
jimbo1968 said:




Just to confirm > This means I don't need an output buffer on the crossover if it is in the same box as the amplifier?


jimbo1968 said:
Under what conditions do you need an input buffer to the active crossover?

Mine will be fed by the pre-out from an AV amp.


If you have that whole circuit you plan to copy in one box (or one box per channel) then you don't need additional buffers. The design alrady has it in there.

You start needing buffers when you jump from one box to another using interconnects.
 
soongsc said:
If you have that whole circuit you plan to copy in one box (or one box per channel) then you don't need additional buffers. The design alrady has it in there.

Let me clarify: My crossover and amp are in 1 box.

preamp --------- || IP buf > HP > LP > OP buff > LM3975 ||------- Midwoofer

Which of the buffers do I need or not need ?
 
jimbo1968 said:


Let me clarify: My crossover and amp are in 1 box.

preamp --------- || IP buf > HP > LP > OP buff > LM3975 ||------- Midwoofer

Which of the buffers do I need or not need ?

Congradulations! If have it all! No need fore more. I assume you are using the OP buff also to adjust gain as in the original design does.
 
jimbo1968 said:
preamp --------- || IP buf > HP > LP > OP buff > LM3975 ||------- Midwoofer

Which of the buffers do I need or not need ?

soongsc said:
Congradulations! If have it all! No need fore more. I assume you are using the OP buff also to adjust gain as in the original design does.

Yes, I realise I don't need more buffers! But can I remove any?

You are right that I want gain adjust. I think that can be done as a part of either buffer. (Which means I can't remove both I suppose)
 
Status
Not open for further replies.