Passive Crossovers (split out of Ti800.4 review thread)

[csperl1]

Ok so about 34# for a pair.

So if I sold my 3 ms 2250TA's for a pair of these I would lose about 17 pounds, gain flexibility, pocket some dough, gain space, and another pair of channels for a 3 way, but lose about half the power per existing channels. Plus I could get rid of my noisy 360.2 and Palm and use the 800.4 crossovers but would need a high quality 4 channel eq with time alignment that I could buy with the $ from the sale of the 360 and 2250's. Could prob get a pretty penny for my pair of consec. serial number 2250TA's too...

Hmmmm....decisions.

Think of it this way, the MS2250TA is 225W x 2 (the first overrated PG amp I have come across). If you bridge this Ti800.4, you end up with 500W x 2 into 4 ohms (unloaded non clipping voltage related), which is far more than the MS2250TA.

I personally would use one Ti800.4 for the right side, and one for the left. Not one for the components, and one for the bass. Use the front of each for a component, and the rear for a sub. Keep the sub at 4 ohm or more to make it easier on the amp. With a pair of these used like that, you will have the best stereo separation money can buy (two separate amps).

I am using separates with no passives so I am going to need a channel for each speaker. I refuse to use passives. So I guess that leaves me buying 3 800.4's. One amp bridged for the left, one for the right, and one bridged to the subs. Would there be an appreciable sonic difference afforded by the doubling of power and true stereo separation? For the record I listen to it loud.

[Eric D]

If you listen loud, then the doubling of power is for you.

You say you refuse passives, but they are the way to go if you want great sound that is still very loud. In most installations passives sound better, as they are designed to work with the speakers at hand. Running things active is hard to get right, and without very advanced test equipment (which you may have for all I know), they never blend together correctly.

I used to also want an all active system, but then when I spent time working at Rockford and seeing what goes into building a good passive, I realized they are the way to go for car audio. You throw away so much power by going active, what little SQ you gain is not worth the cost in space and money.

You still need to run subs on their own channels, as a low frequency passive network is physically large and eats up a lot of power on its own, but from tweeters to midranges, passive networks are well worth it.

All this is assuming good passive setups, not something cheap. I still swear by my old Boston Z6s, which I think were close to $1,000 when new (can't remember). Their network is impressive and reminds me of some nicer home speaker networks.

[Eric D]

Additionally, I notice you mention Dynaudio D260s in your signature. Are those the speakers you will be running?

If so, those speakers are excellent, but the factory crossover really is a joke. If you copied the design of it using larger much better quality components, then ran your speakers with big power (like the Ti800.4 bridged), I am certain you will be very pleased with it.

[csperl1]

Thanks for the info.

I used to use MS amps and cap/coil networks back in the day and an Ax204 for the MB/SW Xover point. I just don't have the time to build and tweak the passives nowadays. I have the system pretty dialed in with the 360.2 so maybe I could use those x-over points as the basis for building my own with high quality inductors and caps. That way I could save $ and space by only buying 2 amps but that may be offset a bit by the cost of HQ passives. I would also LOVE to get this noisy pos 360.2 out of the system.

Yes, I am using the drivers in my sig. I was always under the impression that passives were the power wasters and also that active xovers left the mids and highs susceptible to IM when run with subs. (I mean a girly bass amp not a PG). I realize with the dual power supplies of the 800.4 that will not be an issue.

Um sorry I kinda hijacked the thread...

[smgreen20]

Great review Eric, and yes that pic above was very helpful. For those of us trying to learn the innards of an amp, that was great. I saved the image to my laptop.

Passive xovers are power wasters. General rule of thumb is for every order slope, it's 5% of the power wasted. Example, 100 watts going to a 4th order (24 dB/oct) xover would waste 20 watts. Then there's also the xover phase shift that active doesn't have.

Linky Eric. I know you've based your findings on something. I want to read it too.

[Eric D]

OK, for a bit more on the completely off topic passive network talk.

You have three measurement points, the output of the amp (which is the input to the crossover), the terminals on the tweeter, and the terminals on the midrange.

With a passive component set hooked up, play a 500Hz tone (at low output so you don't fry your mid). Measure the voltage out of the amp. Measure the voltage at the midrange's terminals. They should be the same, which would mean no power is wasted (if you have the same voltage at the speaker terminals as the amp, it would be the same as if you directly connected them). I don't have a link to this info, this is something I personally have done and you can do it too to prove it to yourself.

If you do this test on a tweeter (and a tone of say 8kHz), you may get the same result, but I doubt it. Most crossovers have built in tweeter attenuation. This will mean less voltage at the tweeter.

At the crossover point you will get all kinds of readings, and that is because the passive is blocking power to the speakers (blocking lows for a tweeter, and highs for a mid).

Now, the problem is a lot of crossovers are cheap junk. With too small of parts and a big amp, the parts of the crossover will "saturate". When this happens everything changes. Less power to everything, the crossover will heat up, things will break, and your system will sound bad. Saturation is why most people argue going active. They probably ran a big amp into their components and it sounded weak or bad, or who knows, so they went active to not limit power.

Generally the passive crossover design is great with the speakers included when ran within the specified power level. If the components sound good at medium volume, but bad at high volume, one way to fix this is by rebuilding the crossover using better parts. I have a set of DIY home speakers in which I have close to $400 in the crossover parts alone. But, each crossover is about as large as a small amplifier.

Now, the specific case of the Dynaudio components. Dynaudio is a great company with some impressive goals. Most of their speakers are designed to be used with only a 1st order (6db/octave) crossover. That means one cap on the tweeter, and one coil on the mid. This is as simple as you can get. They rely on the speakers themselves to blend well. Their tweeter is a wee bit larger than most, letting it play a bit lower, which will happen with only a 1st order filter on it. The best part of this, is you don't need to dump a lot of parts or money into the crossover for Dynaudio components.

Yes I have amplifiers mounted to my walls...

[csperl1]

Ok I was always confused about this issue. Here is how I see it. Please correct me as you know a poop ton more about this than me.

Say you sweep a frequency from 500hz-20k to a mid/tweet with a simple coil to knock the highs out of the mid and cap to knock the lows out of the tweet, at the x-over point say 4k. As the freq increases up to 4k the cap decreases the impedance and the inductor increases its impedance. Nothing theoretically lost there except for loss insertion and if you decrease the number of parts (as is the case with wide pass drivers like dyn) or buy big honkin' parts, you are good except for phase issues. In the case of phase issues they can be remedied by using lower order slopes/stacking slopes/ or swapping terminal polarities. Is this correct?

Should we start a new thread about this?

Call me a Neanderthal...but I would rather look at PG amps on the wall than some moody, homely-looking woman like that Mona Lisa chick any day of the week.

[rscecil007]

Call me a Neanderthal...but I would rather look at PG amps on the wall than some moody, homely-looking woman like that Mona Lisa chick any day of the week.

Oh I like amps on a wall, I was just wondering WHAT amps were on the wall....

[Eric D]

Call me a Neanderthal...but I would rather look at PG amps on the wall than some moody, homely-looking woman like that Mona Lisa chick any day of the week.

Oh I like amps on a wall, I was just wondering WHAT amps were on the wall....

Lots of them!

As for the passive crossover points, I agree with you on them. And, yes we should just start another discussion, as it will be helpful to others.

[rscecil007]

Lots of them!

More pics, I want to see your shop wall art.

[dwnrodeo]

OK, for a bit more on the completely off topic passive network talk.

You have three measurement points, the output of the amp (which is the input to the crossover), the terminals on the tweeter, and the terminals on the midrange.

With a passive component set hooked up, play a 500Hz tone (at low output so you don't fry your mid). Measure the voltage out of the amp. Measure the voltage at the midrange's terminals. They should be the same, which would mean no power is wasted (if you have the same voltage at the speaker terminals as the amp, it would be the same as if you directly connected them). I don't have a link to this info, this is something I personally have done and you can do it too to prove it to yourself.

If you do this test on a tweeter (and a tone of say 8kHz), you may get the same result, but I doubt it. Most crossovers have built in tweeter attenuation. This will mean less voltage at the tweeter.

At the crossover point you will get all kinds of readings, and that is because the passive is blocking power to the speakers (blocking lows for a tweeter, and highs for a mid).

Now, the problem is a lot of crossovers are cheap junk. With too small of parts and a big amp, the parts of the crossover will "saturate". When this happens everything changes. Less power to everything, the crossover will heat up, things will break, and your system will sound bad. Saturation is why most people argue going active. They probably ran a big amp into their components and it sounded weak or bad, or who knows, so they went active to not limit power.

Generally the passive crossover design is great with the speakers included when ran within the specified power level. If the components sound good at medium volume, but bad at high volume, one way to fix this is by rebuilding the crossover using better parts. I have a set of DIY home speakers in which I have close to $400 in the crossover parts alone. But, each crossover is about as large as a small amplifier.

Now, the specific case of the Dynaudio components. Dynaudio is a great company with some impressive goals. Most of their speakers are designed to be used with only a 1st order (6db/octave) crossover. That means one cap on the tweeter, and one coil on the mid. This is as simple as you can get. They rely on the speakers themselves to blend well. Their tweeter is a wee bit larger than most, letting it play a bit lower, which will happen with only a 1st order filter on it. The best part of this, is you don't need to dump a lot of parts or money into the crossover for Dynaudio components.

Yes I have amplifiers mounted to my walls...

I will agree that a lot of crossover design goes into a component set, but for car audio, I don't feel it is needed in every scenario. Home audio, yes it is necessary, as there are not many amplifiers capable of adjusting crossover points, or have high quality equalization options. If you were to choose your own drivers for car audio (that aren't available as a component set) then active would be cheaper, and can sound just as good, as well as offer greater flexability for tuning options. If you know the driver's T/S parameters, set the high pass crossover point greater than 2 octaves above the Fs, have access to an RTA to make sure there aren'g any bumps or dips at a certain frequency, and have equalization with at least 15 bands per channel, I think you would have great success in getting things to sound good. Granted, there will be a lot more tuning needed, it can be done. With the amount of processing available in headunits these days, coupled with amps that have onboard crossovers, it makes going active that much more attainable. Plus no expensive resisters, capacitors, or inductors to purchase.

[Eric D]

I don't think we are on the same page. I am not talking about buying your own drivers, and making them work, I am talking about existing component sets in which the company has already spent a ton of time and effort designing the crossover. Going active in this case puts you right back to square one, and you would be better off using the included crossover, or improving it.

Resistors, capacitors, and inductors can be expensive, but typically 2 more amp channels are going to cost you a lot more. In the case of 3 way, you need 4 more amp channels. Plus you need the space to fit those amp channels, and more current from your electrical system to feed them.

Additionally, getting phase correct is a lot harder when you go active, as you have no way to control it. Sure, time alignment might let you adjust right to left, or front to rear, but very few processors let you do phase adjustment on each driver. A passive crossover already has phase taken into account (I am assuming a good one, not one on a $50 set of components).

[Eric D]

A lot of you guys may think it is pure blasphemy that I am arguing against using active crossovers, but in my experience and from what I have learned about electronics, they just don't make a whole lot of sense for car audio purposes.

They are perfect for sub to mid crossover points where passive crossovers would be physically huge, and of a very low slope. Other than that though, I think the bad outweighs the good.

I for the most part always advocate "less is more" when it comes to car audio. I like one pair of component speakers in the front, a pair of subs in the rear, 4-channels of amplification to drive them, and a 2-way active crossover between the sub and component amp channels. If the system needs it, then add an EQ. But don't stick an EQ in there if it does not need it.

I also advocate power over complexity. If you have 4 channels active, 2 on tweeters, and 2 on midranges, bridging the 4 into just 2 and using a quality passive crossover will make you more happy with the stereo in the long run. IMO, nothing beats playing your stereo so loud you cannot hear your own voice over it, yet it still sounds clean and comfortable at that level. But, this will without a doubt lead to hearing loss...

All of this is just my opinion, but I am posting it for the record so others know where I am coming from on this.

[dwnrodeo]

All of this is just my opinion, but I am posting it for the record so others know where I am coming from on this.

I believe you hit the nail on the head with this one. Music is subjective for the most part, and everyone isn't going to agree on a similar setup.

[marko]

i love my passive mid to high is exceptionally smooth for focal drivers!

[stipud]

Agreed with everything you said. It bugs me when people take expensively engineered components with a crossover and run them actively. I tried and failed with my Elites a handful of times.

Active is great for using mismatched sets of loose speakers if you have the time to tune everything yourself, but it's a complete waste of money to do it with a full component set (and on a well engineered set, the results are difficult, if even possible, to replicate).

More power on a passive has always sounded better in my experience, as long as the crossover can handle the wattage.

[Eric D]

Another fun aspect to this is how was that passive designed in the first place?

When I worked at Rockford Acoustic Designs, we designed the passive networks to give the best sound possible, on axis, in test boxes, in our listening room. It was about as far from a car as you can get, and was intended to make the speakers sell in a demo room at an audio shop.

When I took some Boston speakers (Pro 6.5s to be exact) and installed them the same way in the same listening room, they measured like crap and sounded like crap. Turn the test boxes in towards each other about 45 degrees and things changed. The speakers measured flatter, and sounded way better.

Some companies design their speakers and the matching networks for them in a lab. Others design them in a car. Huge difference.

Trying to actively duplicate a passive crossover designed specifically for off axis response in a car is for all practical purposes impossible...

[shaheen]

A lot of you guys may think it is pure blasphemy that I am arguing against using active crossovers, but in my experience and from what I have learned about electronics, they just don't make a whole lot of sense for car audio purposes.

They are perfect for sub to mid crossover points where passive crossovers would be physically huge, and of a very low slope. Other than that though, I think the bad outweighs the good.

I for the most part always advocate "less is more" when it comes to car audio. I like one pair of component speakers in the front, a pair of subs in the rear, 4-channels of amplification to drive them, and a 2-way active crossover between the sub and component amp channels. If the system needs it, then add an EQ. But don't stick an EQ in there if it does not need it.

I also advocate power over complexity. If you have 4 channels active, 2 on tweeters, and 2 on midranges, bridging the 4 into just 2 and using a quality passive crossover will make you more happy with the stereo in the long run. IMO, nothing beats playing your stereo so loud you cannot hear your own voice over it, yet it still sounds clean and comfortable at that level. But, this will without a doubt lead to hearing loss...

All of this is just my opinion, but I am posting it for the record so others know where I am coming from on this.

Sir I applaude you ....this is the same principle I have built my cars on for years, I think the new "trend" is active with loads of TA to get everything in the right place so to speak.

[KUB3]

Marko that is passive porno

Eric, I agree with you entirely when it comes to passives. They have soooo many benefits.

I happen to be currently running some three way CDT's passively. But that will be changing shortly. My CDT 3" midrange and tweeter will be going, due to some new point source 3" drivers to handle the entire upper spectrum. That means I will only have a single x-over point in my entire audio system: 250Hz from my mid-subs to my point source.

Initially I will run the new driver with my CDT passives, as that x-over point is spot on. Also both drivers (old and new) have the same sensitivity so it should be a good match. However the passive is adjustable for each driver if required - on both volume and EQ to a degree. The mid range position on the x-over can also be made to run open at the top, so running a point source driver is fine.

But eventually I'd like to have some custom passives made from larger components. So I would need a simple design and a list if suitable high quality components. If you have any input on either it would be appreciated.

[shawn k]

Do I dare get in this conversation..lol???

First let me state that you CAN and WILL find many systems utilizing passive crossovers that sound fantastic. Many companies will invest lot's of R&D on particular crossover designs. This is especially true for high-end components. However, the truth of the matter is that these companies provide these crossovers out of necessity and not necessarily because it's the "best" performing design. Speaker manufacturers "need" to provide a crossover with their components or sales will undoubtably plummit! Can you imagine if you were new to car audio and went to a retailer to buy a set of components and the salesman told you that you have to buy TWO amplifers (or at least a 4 ch amp) with on-board xovers just to get the speaks to perform right??? It just doesn't work.

I hate to say it guys, but there's really only "one" (especially in a vehicle) downside to running active over passive, and that's simply the need for more channels of amplification.

A passive crossover has many downsides that should be considered. For instance. The passive "will" rob power from the amplifier. There's no way around it. The very best of designs will inevitably have a lower effect, but it's still there. Keep in mind that the best xovers can be very expensive.
Also, many people are unaware that passives can actually cause your amp to clip faster than active. This is exactly why you may hear so many people say to up the power (by bridging or adding more amps or whatever)! Why do you need more power if xover's don't steal power (which they do ) right??? I'll explain. We all listen to music and not sinewaves right? With music you will obviously have multiple sinewaves simultaneously. Here the high freq waves "ride" on the low freq waves and this will give you higher output voltage than if you were to amplify "just" the low frequency waves or "just" the high freq waves. Remember with a passive xover the filtering is being done "after" amplification and the amp amplifies full badwidth... duh right?? So.. as what has been said before just add more power to increase headroom right?? Well if you need more power for adequate results, the argument for paying for more channels by going active starts to fade Furthermore, passives are dependent on impedance. Many crossovers are built poorly by using the drivers nominal impedance (4ohm for instance) as a design parameter. A good design should take into account the actuall impedance of the driver/s specifically @ the xover point for best results. Even with these better designs, it cannot be perfect as the driver's impedance will change/ rise as the vc heats up. OK so I'm not done yet..lol.. Due to capacitive and inductive reactance, damping factor is effected. The amplifier cannot drive the speaker as well as if it were simply connected directly to the voice coil. This effects transient response. Imagine your an amp. you have to drive a.... um.. ho(speaker) back and forth. With an active set-up you are holding the ...ho... by the handle and you can stroke that ho back and forth with precision. With a passive set-up, imagine you now have a heavy spring (the passive xover) between you and the ho. As you might imagine, you can still push/pull the ho but it's funky (springy) because of the spring........ I hope that analogy worked.. it's all I could come up with right now

So with an active xover, you have pretty much negated all of the above Short of needing a couple more amplifier channels you have many benefits. Ok so you could argue that an active would/could induce noise, but with all of the internal xovers in amps and HUs these days it would be minimal (inaudible) in anything of decent quality.

On a side note... Marko: Very nice Utopia xover there. This is certainly a slight exception to most passives out there. If I'm not mistaken, that has the ability to adjust phase at multiple points for each driver. This is far in advance of what most xovers provide even if they do provide some type phase shifting. They are no where near what the Focal provides. Super!


So as I said before, Passive crossovers "can" still sound great! Hell, I competed nationally running passive. But I'm confident, with experience, that active is the way to go. If you can at all tune a system and you have enough adjustment, you will benefit from it!

[Eric D]

There are at least "two" downsides, and you forgot one of them...

Only maybe 1 out of 1,000 people in car audio have the required test equipment to setup an active crossover. And out of those people probably only 1 in 10 have the knowledge to do it right.

There is no way to get an active setup sounding better than a quality passive setup without at a minimum an RTA. You should also have an oscilloscope for good measure, as there is no point in fighting clipping related distortion while trying to get the crossover points correct. To really get it right you need to use a swept sine wave analyzer, and those run over $1,000 for a cheap one.

A possible third downside is what do you use for an active crossover? The ones built into the amp? What if they are 4th order and you need 2nd order? Or what if you only need 1st order? So now you have to buy an external crossover. Well, what if you need 1st on the mid, and 3rd on the tweeter? Are you going to run a passive on the mid, and go active on the tweeter with an external crossover because your amp is 4th order? This does not make any sense!

[ttocs]

so how about x-overs that have a bi-amp capabality?



Devils advocate......

[Eric D]

For some additional information:

When we designed passive crossovers at RAD (Rockford Acoustic Designs), we hooked the mid and tweeter up actively, with a very expensive (and I must stress very expensive) digital crossover. It let you choose slope, and frequency. It would even let you create filters that physically cannot exist in the analog (passive) domain. Once we got the best response we could, we than would build the filter out of coils and caps. At that point we would hook it up and do more testing. We would then tweak the design by adjusting the values of the coils and caps to get the response back to where it was on the digital crossover, or better.

This process generally took about a week from start to finish, with several people from the design team involved. There is no way someone is going to slap a 4-channel amp and some mids and tweets in their car and get their active filter setup right.

[Eric D]

so how about x-overs that have a bi-amp capabality?



Devils advocate......

Have you ever tried it? What results did you get?

From everything I know, the only real advantage to a bi-amp passive crossover is broader adjustment of the gain.

Say in your install your tweeter is 4 db too bright (as verified by an RTA). Your passive crossover has -3db, and -6db padding for the tweeter. Your crossover also has bi-amp ability. Since you can't get the exact -4db you need, you can run a pair of amps, leave the crossover set at 0db, and reduce the gain on the tweeter amp to get the exact -4db you need. But in all honesty, we are splitting hairs here. I don't think anyone will be hearing the 1db difference in this example.

[shawn k]

There are at least "two" downsides, and you forgot one of them...

Only maybe 1 out of 1,000 people in car audio have the required test equipment to setup an active crossover. And out of those people probably only 1 in 10 have the knowledge to do it right.

There is no way to get an active setup sounding better than a quality passive setup without at a minimum an RTA. You should also have an oscilloscope for good measure, as there is no point in fighting clipping related distortion while trying to get the crossover points correct. To really get it right you need to use a swept sine wave analyzer, and those run over $1,000 for a cheap one.

A possible third downside is what do you use for an active crossover? The ones built into the amp? What if they are 4th order and you need 2nd order? Or what if you only need 1st order? So now you have to buy an external crossover. Well, what if you need 1st on the mid, and 3rd on the tweeter? Are you going to run a passive on the mid, and go active on the tweeter with an external crossover because your amp is 4th order? This does not make any sense!

I'm sorry but I still disagree. I don't see where it's necessary to create a good active crossover without said equipment. BTW I have both an O scope and an RTA and I wouldn't use either when setting up an active filter. If you know the response curve for the particular drivers you are using and the fs especially for the tweeter, one can determine at the very least a good starting point for xover points. From there the points can be further adjusted if needed. I feel like you haven't completely read through my post. I did state that you would need enough adjustabliltiy to get it right. With todays processors, and even within some amplifiers it's amazing how much adjustability we have, and we can take advantage of that! Yes there are processors, not even extremely expensive, that can provide a multitude of crossover points AND slopes. Crossovers are by no means rocket science! Any company, including RF, can design a crossover with any goal in mind. As you stated before: RF was designing crossovers to sound good on a demo board. Well this is horrible if we install speakers in cars right??? Ok.. so forget that and let's say a company, like Boston, designed a crossover that has good off axis response. Fanstastic, but it's still not ideal. If the BA sounds great at 45 degrees then I would certainly say this is "better" than the RF philosophy, but what if I need to install my drivers at 30 degrees? You are inevitably going to have phase shifts especially within a vehicle and no design is going to be perfect (well the Utopia is close albeit extremely expensive ) Straight up.. with an active you don't have to worry about this. You can now work with the off axis response of just the driver/s. Sounds nice to me! You have only come back to shunt the active route,and that's fine, but you cannot deny the downsides I have provided about the passive xover. They do outway the negatives of going active. Look. I was never bashing on going passive. They are simple, cheap, and still sound good. I think I made that clear in my first post. I just want to share some knowledge. I think we are on the same page here yes? I'm a guy much like yourself. I like to experiment and run tests of my own. Not too mention just play around and try new things sometimes. I have done comparisons in the past with active vs passive. I can say that there was a particular time where I noticed a much more dynamic soundfeild when utilizing an active set-up compared to a "nice" passive design. I'm not going to sit here and say that everyone should junk their passive's. It's just not logical. At the same time I think it's silly (I'm not pointing fingers at anyone paricular here) when people when people trash an idea without trying and working at it first.