[Quiet Earth]

If you use an active sub (or subs) to supplement a pair of full range speakers, and you are not concerned with the high pass part of the crossover because you leave the mains uncrossed, do you prefer the sound of the line level or the speaker level input?

Just curious what the consensus is.

[SteveFord]

Line level for me.

[Chromisdesigns]

Line level for me.

Me too. 

[roscoeiii]

Depends on the implementation. IMO, the great thing about REL is its speaker level input circuitry. This is what they recommend and it works beautifully for integrating a sub with your main speakers (see the REL literature multitude of REL subwoofer reviews for the technical reasons why). While the other aspects of their subs are high quality, in my opinion the quality speaker level circuitry is what REL gives you that other sub competitors do not. Not all circuitry is created equal.

You read again and again how well RELs integrate with speakers for music, and this speaker level input seems to be the main reason why.

[Quiet Earth]

You read again and again how well RELs integrate with speakers for music, and this line level input seems to be the main reason why.

You mean speaker level input, right?

[roscoeiii]

You mean speaker level input, right?

Yup. Lemme correct that...

[Davey]

Depends on the implementation. IMO, the great thing about REL is its speaker level input circuitry. This is what they recommend and it works beautifully for integrating a sub with your main speakers (see the REL literature multitude of REL subwoofer reviews for the technical reasons why). While the other aspects of their subs are high quality, in my opinion the quality speaker level circuitry is what REL gives you that other sub competitors do not. Not all circuitry is created equal.

You read again and again how well RELs integrate with speakers for music, and this speaker level input seems to be the main reason why.

Plenty of subwoofer amplifiers have speaker-level inputs.

What is unique about the REL approach?  Is it a transformer that performs the step-down?

Cheers,

Dave.

[roscoeiii]

I am not sure what is unique about their implementation. But from reading reviews when I was shopping for subs, I got the impression that this circuitry was their big IP innovation. Otherwise, they just build solid subs that are well-engineered (espectially their ST line, and now the B and G series).

Not sure about other speaker level inputs, but I do recall reading that the REL's input impedance is very very high. IIRC some speaker level inputs are pass-through and incorporate a high and low pass filter. REL subs don't do that. REL reviews would probably have some of the details you are looking for.

[medium jim]

I think a lot has to do with the quality of the main amp(s) vs. the sub amp(s).   In my case I use speaker level to gain the love of my tube mono blocks, vs. the plate amp doing all the work.   This way, I'm getting the best of both.  If I had so, so amps I would go line level. 

Also, consider the preamp you are using as well, in my case a 50 year old relic that wasn't designed with Subs in mind. 

Jim

[srb]

REL's amplifier input impedance is on the order of ~ 150K ohms and has minimal effect on the main stereo amplifier's speaker connections.  Typical subwoofer high level speaker connections require the main stereo amplifier's speaker outputs to be routed to the subwoofer and the main speakers connect to another set of binding posts on the subwoofer.
 
Although this allows implementation of a high pass filter for the main speakers, the speaker wire length is greatly increased and has to pass through an often questionable filter in the subwoofer as well as two speaker cable connections.
 
Although in theory, it is desireable to limit the main speaker's bass with a high pass filter, the majority of users seem to report they like the sound better running the main speakers full range with the high level connection only supplementing bass.  Because the REL subs also have a simultaneous LFE input with its own level control, those users that can also feed the LFE with a high pass filtered input are able to easily compare the two types of connections.
 
Steve

[medium jim]

REL's amplifier input impedance is on the order of ~ 150K ohms and has minimal effect on the main stereo amplifier's speaker connections.  Typical subwoofer high level speaker connections require the main stereo amplifier's speaker outputs to be routed to the subwoofer and the main speakers connect to another set of binding posts on the subwoofer.
 
Although this allows implementation of a high pass filter for the main speakers, the speaker wire length is greatly increased and has to pass through an often questionable filter in the subwoofer as well as two speaker cable connections.
 
Although in theory, it is desireable to limit the main speaker's bass with a high pass filter, the majority of users seem to report they like the sound better running the main speakers full range with the high level connection only supplementing bass.  Because the REL subs also have a simultaneous LFE input with its own level control, those users that can also feed the LFE with a high pass filtered input are able to easily compare the two types of connections.
 
Steve

Using short runs of speaker cable and decent quality minimizes the capacitance, rendering it a non-issue in most cases.  Being able to use a High-Pass to the Mains to me is the key, especially if you have hard to drive speakers as I do and want to use tubes to drive them.

Jim

[Davey]

REL's amplifier input impedance is on the order of ~ 150K ohms and has minimal effect on the main stereo amplifier's speaker connections.  Typical subwoofer high level speaker connections require the main stereo amplifier's speaker outputs to be routed to the subwoofer and the main speakers connect to another set of binding posts on the subwoofer.
 
Although this allows implementation of a high pass filter for the main speakers, the speaker wire length is greatly increased and has to pass through an often questionable filter in the subwoofer as well as two speaker cable connections.
 
Although in theory, it is desireable to limit the main speaker's bass with a high pass filter, the majority of users seem to report they like the sound better running the main speakers full range with the high level connection only supplementing bass.  Because the REL subs also have a simultaneous LFE input with its own level control, those users that can also feed the LFE with a high pass filtered input are able to easily compare the two types of connections.
 
Steve

I don't think you're going to find any speaker-level high-pass filters within most subwoofer amplifiers.  These would be VERY large capacitors.

Whether the REL input impedance is 150k or 1.5k it shouldn't make any difference.  It's minimal loading on the amplifier compared to the speaker load that's already there.
I suspect the high-level input circuitry in the REL subwoofers is the same as all others.......just a simple two-resistor divider that reduces level approximately 20-26db.

Whether the subwoofer input is acquired before the main amplifier or after shouldn't make too much difference if the amplifier is functioning as a voltage source.  An attenutor-reduced version of the amplifier high-level output should look identical to the input.
Of course, if amplifiers with high output impedance are used with fairly complex speaker impedance loads, then the response might be skewed considerably.  In that case, maybe it would make sense to monitor the amplifier output vice the input for the subwoofer signal.

Cheers,

Dave.

[Habs Fan]

.

Whether the REL input impedance is 150k or 1.5k it shouldn't make any difference.  It's minimal loading on the amplifier compared to the speaker load that's already there.
I suspect the high-level input circuitry in the REL subwoofers is the same as all others.......just a simple two-resistor divider that reduces level approximately 20-26db.

Agreed and highly suspect equally fine  results  can be had with the line level inputs on REL's,so not sure why they tout it (high level)as the superior method.

[roscoeiii]

Davey and Habs,

Are you suspecting this about the REL speaker level input as a result of research you have done, or any personal experience with REL? OR is this just pure conjecture?

There is a lot written about REL out there, and much of it led me to believe that there is something unique about their speaker level connection in particular. I am no longer using subs so am not willing to put in the time researching this. If you are shopping for a sub, I would certainly investigate what may or may not be unique about the speaker level connection. Don't draw conclusions on what reads to me like conjecture based on neither research into the matter or direct experience. If I am wrong about this, my apologies. But this is how the above two posts read to me.

My direct experience with REL was that their sub integrated FAR better than anything else I had tried. In fact, I went from dual subs (NHT SW2s) to a single REL and the REL worked much better via speaker connections.

[Davey]

Who's drawing a conclusion?
Of course I'm suspecting......that's why I said "I suspect" in my previous posting. 
You "suspect" the REL has some sort of "quality speaker level circuitry," do you not? 

Anyways, my suspicion is based on previously looking at a number of other subwoofer amplifiers (not the REL) in which I found a resistor divider performing this function.  I don't see why you would need anything more complicated than that.  It's a very simple job....electrically speaking.

Cheers,

Dave.

[roscoeiii]

But I "suspect" it based on researching REL specifically rather than looking into speaker level connections and assuming all of them must be the same. I conclude that REL has "quality speaker level circuitry" based on direct comparisons of REL subs to other subs with speaker level inputs as well as subs with RCA inputs and even the well-regarded NHT X2 sub crossover which also uses RCA. Big difference IMO.

All I ultimately meant to say was that someone interested in REL's speaker level inputs would be far better served looking into how REL actually implements speaker level connections rather than basing a decision on what some commenter in a forum "suspects." And even better, listen to a REL (or whatever component is of interest) in your system.

[Davey]

All right.  We're not allowed to speculate on proprietary REL technology in this circle. 

Sorry about that.

Dave.

[roscoeiii]

Just want to make sure readers of this thread are reading closely and don't mistake speculation borne from no direct REL experience but instead based examination of non-REL subwoofers from more informed speculation based on experience with the subwoofer company in question.

Overgeneralizing and speculation based on no direct experience can get one into trouble. As long as we are clear on what the speculation is based on, we can make informed decisions on what opinions to place the most value on.

OK, point made. Hope this thread is helpful to whoever is reading it. My 2¢ has officially been added.

[roscoeiii]

Agreed and highly suspect equally fine  results  can be had with the line level inputs on REL's,so not sure why they tout it (high level)as the superior method.

Oh, and with regard to Habs suspicion, my direct experience is that the speaker level yields much better results than line level inputs on REL. You will see that many reviewers have had the same experience.

[Davey]

Update on my previous speculation regarding sub-amp speaker-level high-pass filters:

I found this subwoofer amplifier which does indeed have 220uF electrolytic capacitors between speaker-level input and output jacks.

http://www.parts-express.com/pe/showdetl.cfm?partnumber=300-803

What a horrible concept......but there it is. 

Cheers,

Dave.