A bi-wiring question, but NOT what you think (I hope)...

[LeadEars]

I'm setting up a music room in a new much smaller place with the equipment I have on hand. My previous system was fairly high-end, but I've had to down-size a lot these days, due to both retirement-related space & budgetary considerations - so the 450-lb Focal floorstanders have been replaced with KEF LS50 Metas & a pair of KEF C200SW non-powered subs, which I've had since I got them new in '91. I have a question re the electrical effect of bi-wiring non-powered subs with standmount speakers. I'm not re-kindling (igniting?) the question of whether bi-wiring is a good idea or not. I've actually read (or at least skimmed) every post on this forum on the topic - and I'm not doing anything elaborate, nor do I have anything to do anything elaborate with, as far as equalization, crossovers, etc. is concerned.

The C200SW's have an internal crossover & were designed to pair with standmount KEF speakers available back then (or even serve as stands for them using a bracket attachment), with a single pair of speaker wires going to the C200SW input terminals, and a pair of "jumpers" between the C200SW output terminals and the "main" speakers. KEF states in their literature that the C200SWs "relieve" the main speakers of the burden of reproducing the low frequencies. I want to experiment with the 2 pairs of speakers (subwoofers & main speakers) by wiring them both ways - i.e. to compare using and not using the C200SW crossovers, for multiple reasons: I'm not sure how good the crossovers still are, as old as they are, I'm not really sure the LS50's will profit as much from having the frequencies below 120 Hz "stripped out", etc. And, of course, it's an experiment... The C200SW manual says that if for any reason you choose not to connect speakers to the C200SW output terminals, you should put an 8 ohm resistor of at least 20W across them. I'm not sure why it's necessary, or what that will do electrically (to the overall impedance, etc.) or sonically, but they say it's mandatory. My amp is an old Adcom GFA-555, so I'm not worried about the parallel-wired impedance dipping to near 2 ohms occaisionally, but will the arbitrary insertion of the 8 ohm resistor lower it even further, or is it electrically isolated from the amp by the crossover?

My other main electrical question is: If I bi-wire the speakers, since the crossover in the C200SW "feeds" frequencies below 120 Hz to the subwoofer driver, doesn't that effectively "use up" those frequencies, acting as sort of a crossover anyway, without having to pass the signal through the crossover before it gets to the LS50's? In other words, there's no magic ampliflication going on, so any power below 120 HZ that's driving the sub drivers can't be available to the LS50s, correct?

Is there anything else going on in this scenario that I haven't thought about that I need to be worried about? My plan is to try the system out both ways: 1) Bi-wire from the Adcom to the C200SW input terminals with an 8 ohm 50W resistor across the output terminals and a 2nd set of wires from the amp to the LS50s, and 2) A pair of speaker wires to the C200SWs and jumpers from the C200SW output terminals to the LS50s.

As an fyi, the efficiency & sensitivity ratings of both speakers are very comparable, so that shouldn't be an issue. The C200SWs have a nominal impedance of 4 ohms, the LS50s 8 (min 3.5).

 

[quattro98]

In this biwire scenario, the LS50 will play full range. The signal from the amplifier will go to the LS50 completely separately from the subwoofer. The signal has to go through a high pass crossover somewhere to change this.

 

[LeadEars]

I understand that the LS50s will be playing full range, but isn't every watt of energy consumed by the C200SWs in playing low frequencies unavailable to the LS50s? So if, for example, the C200SW crossover is 75% efficient at passing sub-120Hz frequencies off to the sub's driver, aren't the LS50's only getting 25% of it to play with? Otherwise, there's a magic amp somewhere, isn't there?

 

[NTK]

The C200SW manual says that if for any reason you choose not to connect speakers to the C200SW output terminals, you should put an 8 ohm resistor of at least 20W across them. I'm not sure why it's necessary, or what that will do electrically (to the overall impedance, etc.) or sonically, but they say it's mandatory.

My guess is that the subwoofer cross-over filters require the electrical load of the main speaker to work properly. This is very typical of passive cross-over filter designs.

Sometimes, for a passive 2-way cross-over filter, you need 2 "loads", one becoming the dominant load (and sucks up all the power) at high frequencies, while the other becoming dominant at low frequencies. If there is only one load present in the circuit, that load will be dominant all the time and take up all the power, regardless of the signal frequency, making the "filter" useless.

[Edit] In other words, this passive 2-way cross-over works by distributing the signal to 2 separate loads based on the frequency of the signal. The high frequencies go to one load, the low frequencies go to the other load. Some cross-over filters do not work by "blocking" the high frequencies to the low frequency load, or vice versa. They "direct/divert" the high frequencies to the high frequency load (driver), and the low frequencies to the low frequency driver. And for that to work, 2 loads are needed. In the case of the KEF recommendation, the 8 ohm resistor is the dummy high frequency load.

I understand that the LS50s will be playing full range, but isn't every watt of energy consumed by the C200SWs in playing low frequencies unavailable to the LS50s? So if, for example, the C200SW crossover is 75% efficient at passing sub-120Hz frequencies off to the sub's driver, aren't the LS50's only getting 25% of it to play with? Otherwise, there's a magic amp somewhere, isn't there?

Electricity doesn't work that way. The electrons has no idea where you want them to go. The LS50's are "full range" speakers, meaning that they suck up power at all frequencies, even for the low frequencies that they can't reproduce well.

When you connect you multiple speakers (i.e. you main and subwoofer) in parallel, they mostly behave independently of each other, unaware of the existence of the other speaker. The parallel loads will be more challenging for the amplifier, but unless the amplifier's limit is exceeded, this what would happen. (Yes, there is a magic amp, and the magic amp is the existing amp working a lot harder.)

 

[LeadEars]

Thank you both for your replies - they clarify things quite a bit. It sounds like the experiment may be less interesting & useful than I was hoping it would be. With the subs being lower impedance (roughly & generally), perhaps they'll present less resistance to the low frequencies, and that will enable them to accomplish some of what I was hoping for - but that "dummy load" couldn't help the sound, I suspect.

Seems like my best bet is hoping the crossovers haven't suffered as much from age as I'm afraid they have..

Thanks again!

 

[IEales]

I had a look at the brochure on HiFiEngine, but there are no markings on the sub connectors.

The frequency HP is 120Hz.

You could see if the vertical pairs of connectors are a dead short. If not, they are likely rolling the low end out of the mains.

 

[LeadEars]

Thank you for the info. I actually got a reply from KEF, including some info from their archives, and they even passed along some info from a now-retired employee.

You're right about the HP frequency. Another bit of essential info I got was that if you bi-wire primary speakers and don't connect them to the C200SW's output terminals, you have to put an 8-ohm resistor of at least 20 watts across them.

I ended up doing just that, as an experiment, using 50-watt resistors. Apparently, the C200SW crossovers are long past their prime, because the difference in sound quality between running the LS50's through them & bi-wiring them is more significant than any change I've ever made to a system, and that's a lot. It's not a matter of having improved the sound, it's clearly "getting rid of a broken component" better...

Initially, I was concerned about the 8-ohm resistor further reducing the impedance & increasing the load on the amp, but I tried playing orchestral & rock loud enough that I couldn't stay in the room for about 15 minutes, and the amp wasn't clipping or overloading at the beginning or end of the test, and the resistors were still room temp, so they're obviously not passing much current and the amp isn't complaining...

 

[IEales]

You might want to rebuild the sub XO if they are old. Replace electrolytics & mylar with polypropylene

 

[LeadEars]

That's a good point - I should pull them out - I really don't have any way of knowing the quality of the LP side, either, except that the subs are definitely playing. I'm not steady enough in the hands to do anything like that kind of rebuild work, unfortunately, but I should be able to find someone who can.

 

[SIY]

Replace electrolytics & mylar with polypropylene

Whether that's a good idea depends on their position and use in the circuit. I would strongly recommend to @LeadEars not doing that willy-nilly.

 

[LeadEars]

Thanks for the suggestion - "willy-nilly" isn't in my nature, lol. As good as it's sounding right now, "not at all" is more likely.

As i said, doing it myself isn't an option for me anymore - my hands are just too shaky nowadays. My curiosity makes me lean toward replacing/repairing the crossovers, but I also doubt it'll sound better than it does right now, so I suspect it would be a waste of time & money. The only other issue is that the damage they do to the HF sound quality makes me wonder if they're also damaging the LF frequencies going to the sub driver, but I have no way of knowing that without replacing/rebuilding them, I guess...

 

[SIY]

If the speakers are less than about 10 years old, leave stuff alone.

If they're much older than that and there's an electrolytic series cap in the crossover, it won't hurt to replace it for reliability reasons with a good quality bipolar electrolytic. An electrolytic Zobel cap (it's the "C" in an RC connected across the woofer) should also be replaced by an electrolytic. A series film cap can probably be left alone unless there's signs of physical damage. In the photos I've seen of the crossover, it looks like all the caps are bipolar electrolytic (and if you tried replacing the 1000u one with polypropylene, you'd take up most of the cabinet's internal volume!), so use the same for replacements IF any of them show signs of distress like fluid marks, burns, or bulging. There's a 6u cap which could safely be replaced with a film cap if needed.

 

[IEales]

Whether that's a good idea depends on their position and use in the circuit. I would strongly recommend to @LeadEars not doing that willy-nilly.

On what do you base that?

Age and circuit complexity is also a factor in tackling capacitor changes.

Nelson Pass said: "What is more interesting is that even lowly aluminum electrolytics measure much better than any semiconductor, tube or transformer."

Electrolytics have higher ESR to start, which increases over time, decreasing capacitance over time and far higher variation than film.

Mylar has a dissipation factor an order of magnitude worse than polypropylene.

In either case, non-inductive are preferred especially for high current LP section.

IF replacing caps, it's a good idea to measure the originals out of circuit as some manufacturers select caps that may be inside tolerance but a fair bit away from printed value. OLD electrolytics may measure quite a bit different from printed, depending on the original quality.

 

[SIY]

On what do you base that?

Nearly 60 years of designing, building, and repairing speakers. And seeing some consequences of my less-than-good decisions, which is why I want to help others avoid the same mistakes from following fashion and poor advice.

Mylar has a dissipation factor an order of magnitude worse than polypropylene.

Which is irrelevant in this application. It won't hurt to swap polypropylene, but it won't help anything either.

Electrolytics have higher ESR to start

Which is why just swapping them for film caps can throw off a crossover. I personally experienced this when the craze for cap replacement emerged 40 years ago and I (in my ignorance) ruined the sound of some perfectly good speakers.

non-inductive are preferred

The inductance of modern bipolar electrolytics is negligible for this application.

IF replacing caps, it's a good idea to measure the originals out of circuit

If the caps have aged out, this will give a bad baseline. If they haven't, they should be left alone.

which increases over time, decreasing capacitance over time

Again, with modern components, this is not much of an issue. The electrolytics in my 1995-vintage NHT speakers measure exactly to spec, as an example.

 

[Roland68]

Thanks for the suggestion - "willy-nilly" isn't in my nature, lol. As good as it's sounding right now, "not at all" is more likely.

As i said, doing it myself isn't an option for me anymore - my hands are just too shaky nowadays. My curiosity makes me lean toward replacing/repairing the crossovers, but I also doubt it'll sound better than it does right now, so I suspect it would be a waste of time & money. The only other issue is that the damage they do to the HF sound quality makes me wonder if they're also damaging the LF frequencies going to the sub driver, but I have no way of knowing that without replacing/rebuilding them, I guess...

Maybe you should go a different route, which I think makes more sense, and eliminate the passive subwoofer crossover.

Get a the t.racks DSP 4x4 Mini or Mini Pro from Thomann and an amplifier for your subwoofer as an active crossover and DSP.
This also gives you the option of using the DSP to adapt your speakers to the room.

An SMSL A300 amplifier, for example, has 400 watts of power in BTL mode for €150-200.

 

[SIY]

Maybe you should go a different route, which I think makes more sense, and eliminate the passive subwoofer crossover.

Definitely a better way, but it will take some understanding to get it right. But he's found a great resource for help with that.

 

[LeadEars]

My KEF C200SW subwoofers were made in '89 or '90, so call 'em 35 years old. I don't know if the crossovers have aged badly or were badly designed/constructed from the beginning, but the LS50's sound so much better with the signal NOT passing through them that it's got to be one of those... As I said, my only question about continued use of them is whether they're degrading the base going to the subs or not.

As I said in my original too-long post, I really don't want to go the single-sub route, because I've never heard a set-up I like as well as stereo subs placed near the main speakers. My "LeadEars" moniker is a bit of sarcasm referring to my aging ears, which were pretty good at one time. My Airpods say I have absolutely no hearing loss, but that's not necessarily an assurance that I can still hear the nuances of audiophile sound as well as I once could. Still, I have two passive subs, and I have always liked the sound of that setup better generally, so I'm going to try to make that work. To that end, I'm "all ears" to any suggestions, but I'd most appreciate some guidance on the topic of crossovers to replace the ones inside the C200SW's...

 

[IEales]

+10 on DSP and powered sub amplifiers.
I use a miniDSP Flex Eight to drive two Martin Logan subs set flat in my 4-way system. All EQ, XO and time alignment is done digitally in the Flex.

You would probably need to yank the XO from the C200SW.

Can't comment on the MiniPro other than I'd rather control from a computer than have a separate console. Once set up, the console is only useful as a paperweight.
The reviewed here t.racks DSP Mini looks like it would fit the bill.
You could use the second set of inputs for HT low end.

You could probably find a used suitable 100wpc amplifier on one of the UK marketplaces.
100wpc @ 8Ω is likely plenty given the room size

You may need to roll some low end out of the mains for the best performance. That can be done with tone controls if they are adequate or a simple first order HPF between the pre and main amp if that's available.

With the DSP doing contour and time adjustment, the bottom can be superb.