Roo

iTrader: (3)

Hmm, perhaps if you can get in touch with Pete Soper (psoper), he can give you a hand.
That's his profession, IIRC.

joltdudeuc

iTrader: (6)

Nice. I guess I don't need anything crazy, just a "Hey, with those impdences on those drivers, you can buy this crossover and x and y and then bamo, better sound."

But I know what's better for me won't be for everyone else. I'm simply looking to just clean up the cut offs, and get the delivery to the drivers where they should be. I'm just not so versed with how to fix the mixed impedances so I can pick a crossover and use it.

The woofers seem to work out well with an impedance of 16ohm since I can run 2 in parallel, and that will make 8. But then these 5.2ohm tweeter are making it slightly more difficult.

joltdudeuc

iTrader: (6)

Well the AVR hasn't turned off in over 3hrs. So it was probably a fluke or I had maybe pushed the wires into the center channel a little too far.

psoper

That sounds like the more likely scenario, the change in wire gauge is going to amount to a net impedence change probably in the <1% range so it shouldn't be a problem, but pushing the leads in too far to where you have a short will shut down an amp pretty immediately.

I'd tend to agree with Concillion- unless you have measurement equipment and enough circuit design expertise to make informed decision, you have about a 50% chance of making things sound worse, but simple upgrades like wire gauge and higher component quality can give you slight improvements with little downside- provided you don't mess anything up.

pignoseWRX

iTrader: (6)

BAIC, a veritable cornucopia of knowledgeable people.

joltdudeuc

iTrader: (6)

Hmmm. So If i'm sticking to these cheap woofers, then I should not try to put in a real crossover like the Dayton's from parts express?

What if I bought this:

http://www.parts-express.com/pe/psho...number=390-804

Would it help me in finding out what the speakers/tweeters are capable of? Perhaps with some data like this I would know where lower and upper fall off is, and then, know where the cross over points should be?

If all I got going for me is better wiring, and my new amp/pre... then that's okay. It saves me money. But leaves me with more to be desired, lol...

I know that watts don't matter as much as transformer or amp quality, and also the DACs. Those heavy monsters definitely sound better, and have better control of the drivers.

psoper

The Dayton thing is basically just an impedance plotting system, it is of some use if you are in the process of building boxes and doing port (bass-reflex) tuning, but probably not terribly useful for doing crossover design with a given woofer- apart from giving you a good picture of what the impedance of the device is through the crossover frequency range- DC resistance is only one small slice of the actual impedance of a driver- the picture in the Dayton ad shows what a driver in free-air usually looks like in impedance vs frequency, so the nominal numbers that get thrown around don't really tell you much about what you need to know when you get into crossover design.

But to get meaningful improvements in tweaking passive crossovers, you'll need to have means to measure frequency response and distortion, so you're looking for gear more like this:
http://www.klippel.de/analyzer/default.asp

or this:
http://www.linearx.com/products/****...LMS/LMS_11.htm

or- even though not intended strictly for this purpose- what I use;
http://www.meyersound.com/products/sim/sim3/

Truth be told, there are lots of reasons that good speakers are expensive- and cheap speakers usually sound marginal or bad- and most of it boils down to selection of components, cheap small voice coils working in lightweight weak magnets just don't work as well as their heavier and more expensive counterparts, and it's a lot easier to throw something together that is "good enough" than to do all of the hard work it takes to design and build an optimized system.

heck even a lot of the so-called "reputable" speaker companies do not do the R&D that they should to get the best out of their systems. R&D requires expensive test equipment and engineers that not only know what all those graphs and charts really mean, but also understand what the implications are to the overall system performance when changing things.

Even with the fancy analyzers, it isn't as simple as the plots might make it seem, you still have power handling, off-axis response, and phase relationships to contend with from the drivers and the RCL elements at crossover and a whole mess of other stuff that- like we've been saying- can wind up doing more damage trying to "improve on" than leaving them alone.

Bottom line- my suggestion would be that if you aren't satisfied with the performance of the speakers you have, save up, and spend some time researching what's out there so you can buy better ones next time, otherwise I expect you can wind up spending a lot more time and money and still be stuck with inferior speakers that do not live up to your expectations.

joltdudeuc

iTrader: (6)

I see. I think I will rewire the internal wires for sure. The internal wires are 20/22ga. I'm putting in 16ga OFC that's fairly well shielded. Definitely better wire.

So today I rewired the right tower. When playing back TV in Stereo I felt the right side, in comparison to the left side, sounded like it had a lower noise floor, was clearer/more transparent. Since it was anchors speaking I could easily hear the deeper mans voice in the 8" woofer, and a blend of both woofer, mid driver, and slightly tweeter in the weather woman's voice. In the left speaker it was not as apparent. The separation wasn't.

Man, I sound like the crazy audiophiles I talk to everyday on the phone!

This was kinda what I was going for in the first place. Just to clean it up. They sound pretty good to me already. They really are a bargain for any consumer looking for solid sound for the money.

I can't wait to move up to the better AVR. Next step will be that, and whatever I buy will have Preouts so that I can eventually buy a sick amp like an Emotiva.

Thanks for your help Matt and Pete. Really needed someone to let me know what I was getting into.

Concillian

iTrader: (2)

1.2uF puts the crossover way the hell up in 10k+ land.

Seriously, just wire the mids and see if you even notice a difference.

It's possible that it has a huge impedence peak somewhere and that's why the value is so small. But 8-ohm impedence with a 1.2uF cap is a 15KHz crossover, which is almost pointless. If it's 6 ohms that's over 20k at crossover. It must have a big impedence spike if you can hear the difference with / without the tweeter.

Concillian

iTrader: (2)

I would second the advice to just save up and buy better speakers next time. Look at that forum I posted and also at designs posted at

zaphaudio.com

These are largely people looking at real value. Zaph especially looks at every aspect of cost when he does his designs. You can get better quality than $1000 speakers from Best Buy for a couple hundred bucks if you're willing to build your own enclosure. And they have the proper measurements, exerience and tools to verify a design is solid.

How big is your TV? The recommendations I see from people who aren't in the business of selling speakers is that anything around 42" or less you are better off without a center channel if you have L-R mains with decent off-axis response. Especially if you're on a strict budget, since you end up compromising sound quality quite a bit to get a 3rd speaker if staying on a budget.

Also, IMO, decent speakers are priority over electronics. Especially if you're sticking with consumer level receivers (which is pretty much anything you'll find in Crutchfield or the like). They really aren't that different in sound quality.

joltdudeuc

iTrader: (6)

Yeah, I found a calculator that said something like 14k/15k... which is crazy, just crazy. This for the tweeter and the two little 4.5" drivers in the center.

in the Tower, which I rewired now, keeping the caps of course I found:

1.2uF on the tweeter again to one of the 8" woofers
2.7uF on the 3" mid driver which was connected to the other 8" woofer.

The towers impedance at the binding posts measured 7.0.
The impedances of the drivers themselves were:

Tweeter:5.0 and 1.2uF cap
Mid Driver: 7.7 and 2.7uF cap
'Mid' Woofer: 13.6
'Low' Woofer: 14.0

Both woofers work the same, produce and move the same. The mid driver definitely does not have any bass and I can hear upper range audio, mostly voices and some instruments, and the tweeter is the very upper ranges of female voices and a good amount of instruments, like cymbals.

To be honest, for me, there is a good deal of separation happening. I'm guessing the caps are like a high pass filter? They only let a certain freq and up pass through to the mid or tweeter?

If that is the case, then the woofers themselves are still getting full range audio right? This is something that I've wanted to correct with a crossover, to keep the frequences clean and to the driver I want them to go to.

It's doesn't sound bad at all, i'm just looking for a little more. The wiring has changed the over all audio in a few ways, and not like a night and day or anything, but noticeable:

-Clearer audio reproduction (less noise, just, more transparent.)
-Putting slightly more emphasis in woofers now, or at least, it output is slightly stronger.

I hope I'm making sense, and this is normal for a rewiring job, otherwise I'm drinking my own snake oil here

joltdudeuc

iTrader: (6)

It's 58", the center is just below it and the towers at to it's sides, slightly toed in.

For me they sound great, and I think if you listened to them, you would find them to be better than what they cost.

International B did when he came over one day. He was impressed with their sound for the money. But of course, I can spend more and get more.

In terms of the AVR, I know the overall sound may not be too different due to processing or perhaps the DACs, but I really do feel that if the amp section is beefier, it should sound better. I used an old 50+ lbs denon AVR and man, the speakers sounded really good on that amp. My amp at home doesn't compare, and the biggest difference is that amp section.

psoper

If you look at the pictures on the Dayton site you can see that at the resonant frequency, the impedance of a free-air woofer goes up to usually at least 4-5X the nominal impedance, which is typically still a couple of ohms higher than the DC resistance.

Tweeters have resonant peaks as well, although usually more like 2-3 times nominal.

This means that simple "calculators" for figuring cut-off based on just DC resistance and capacitor value are all but useless, you'd be better off I think just having a range of caps and trying different ones than relying on equations that do not consider the full "complex impedance" of the drivers.

However, like you are finding- stepping up the wire gauge, especially in lower-cost speakers which came with really thin internal wires can give a noticeable improvement, so yeah, I think you probably are really hearing a slight improvement with that move.

Getting higher quality components in place of the little can caps could be one more degree of improvement but without the ability to analyze the net result of changes, I'd avoid messing with the component values and stick with what the manufacturer had.

joltdudeuc

iTrader: (6)

Got ya.

I think my next move for the speakers in the future will be either to replace the whole thing someday, or maybe replace the drivers with ones that are known to be good and easy to setup with crossovers like the stuff they have at parts express.

psoper

The other thing about low-order passive crossovers (i.e.; a single capacitor in series) is that they often do double duty as an equalization filter as well as acting in that crossover/LF protection role.

Imagine a tweeter that has a response with a big 12dB peak at the lower end of it's operating range, say 2KHz, that naturally slopes down to 0dB out at say 9KHz.

Placing a series cap tuned to 10K would flatten out that whole band so it matches the lower sensitivity and response of the mid or woofer.

Yes, you do lose efficiency on account of it, but it's going to sound a lot better than having all of that high-mid sensitivity blasting away- so just because the numbers seem screwy, doesn't mean things weren't done that way intentionally.

Nearly all of the systems I deal with are active crossovers, where we can separate the crossover function from the equalization, all of which takes place upstream from the amplifier channels, and we usually have lots of DSP or analog filters at our disposal.

Passive crossover design is considerably more of an art since the drivers themselves are not that easily mathematically modeled and passive components are far from ideal, so lots of compromises- and shall we say "creative" design approaches need to go into any passive crossover design.

Going the "parts express" route will require selecting drivers with matched sensitivity on top of all of the other things we've been discussing, not that it can't be done, but I still think it's going to be cheaper and easier to find a better off-the-shelf product.

Then again, if you want to get into speaker building as a hobby, you have nothing to lose but your time and money, which I think pretty much any hobby will relieve you of at some rate... and there are vast resources available out there for the hobbyist to help you get good results and not have to waste too much in the way of time or money