Ringing problem with dayton DIY's

I had a friend rebuild my RadioShack 2-way speakers with Dayton parts. He used the following:

DC28F-8 silk dome tweeter 89dB
http://www.partsexpress.com/pe/pshowdetl.cfm?&Partnumber=275-070

DC250-8 Classic woofer 89dB
http://www.partsexpress.com/pe/pshowdetl.cfm?&Partnumber=295-315

Acoustastuff Polyfill
http://www.partsexpress.com/pe/pshowdetl.cfm?&Partnumber=275-070

Dayton XO2W-2K 2-Way Crossover 2,000 Hz
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=260-140

Radio Shack Nova-18 cabinets

These speakers sound awesome on simple clear music, such as Pink Floyd or Bob Dylan or Maggot brain, and it even sounds really good with Cake and other modern pop. But it can't handle metal. When the music gets complicated and the high-pitched four part harmony male vocals start, the midrange spikes in volume rather sharply and I get a ringing, garbled sound. Songs like Savatage's "Not What You See" and Blind Guardian's "And Then There Was Silence" are two big offenders. Lots of Muse songs too. What's more, a lot of my CDs which are range-compressed sound like they're being driven into clipping. This happens at even moderate volumes, but is really noticeable at higher volumes.

If I swap speakers out the problem disappears, so it's not my amp or input source. (amp: Yaqin MC-100B tube amp, 30 watts/channel triode 60/watts channel UL, neither mode makes any difference. Input source: Toshiba HD-A3 HD DVD player. Not exactly HiFi, but will do for now as I'm on a budget).

I'm really at a loss as to what could be wrong. I'm new to Hi Fi, and speaker design is not my thing. The guy that built these never tested them with metal. He says they sounded great when he ran Pink Floyd through them, but admits he didn't do much testing beyond a few songs here and there (and all on vinyl).

He is going to stop by and demo the speakers again on my system with my music, but I'd really like to be armed with some knowledge. He knows a bit about speaker construction - he's got the computer programs and stuff, and he built himself a really nice set of bookshelf speakers that sound great... so I'd like to be armed with some knowledge of what could be wrong. Right now all I can say is that they're "ringy" at certain frequencies. I'm hoping some gurus here can help me determine why.

Thanks all.

Charles.
 

fwater

Member
2007-11-14 6:02 am
The Dayton woofer's response graph shows an almost two octave spike as a result of cone breakup, peaking around +9 dB at 2200Hz. This would explain a garbled or ringing sound. Even when properly attenuated, cone breakup can sometimes still be offensive sounding. The pre-fab crossover can't even begin to handle this spike. An aggressive Zobel might help. I ran a sim without knowing the values of the pre-fab X-O componnents but made an educated guess, and came up with a 50uF cap & 4ohm resitor that pulled the peak down by at least 9 dB. The sim also suggests around a 3ohm restor in series with the tweeter (after X-O) would level out response. The result will be less sensitivety, but a more listenable and balanced sound. Definitely worth a try to save these speakers. BTW, if you don't know the terminology, "a Zobel" refers to a simple X-O component to help control a drivers rising impedance and therefore making the intended slope more effective. It is a capacitor and a resistor, wired in series, and then wired between the woofer's positive and negative terminals.
 

fwater

Member
2007-11-14 6:02 am
Charles, I visited you weblink. Anyone that can turn an inexpensive jap bike into a slick chopper was born to DIY a set of speakers. You'll get five times the results compared to what you have now with your building skills and by asking a few questions, and for probably not much more money. I could help design a more proper crossover for the drivers you already have, if interested.
 
Hi,

As stated the c/o is very wrong for the drivers, and even with a flat
mid/bass unit would only work if it is zobelled. The "design" also has
no baffle step included or any padding for tweeter level.

I'd guess the inductor is tapped and only one of the low pass
capacitors is used for 8 ohm / paralled for 4 ohm.

A 10" bass- mid / 1" tweeter 2 way is a very difficult thing to do.
For these drivers where the tweeter distortion is "average to
poor- Zaphaudio.com" and the 10"ers midrange distortion levels
also not good (see PE PDF) its never going work well IMO.

http://www.speakerbuilder.net/web_files/Projects/Lyra/lyramain.htm

As you already have the bass units and tweeters I think the above
would be a very good to rather excellent idea if you would like what
can be called acccurate clear midrange, which is what mainly counts.

The higher tweeter c/o frequency reduces its distortion and the
dome midrange runs the frequency range of the 10" distortion.

Reducing the 9R midrange resistor slightly may liven it up a touch.

FWIW I doubt in the current state, your speakers are awesome
on any sort of material. They are simply wrong which may sound
interesting with some material but dreadful with other material.

:)/sreten.
 
[IMGDEAD]http://images.craigslist.org/010212010306010406200801120967e6cb45ef4ae3ee007056.jpg[/IMGDEAD]

Hi,

The cabinets appear to be a little small for the driver, 23"x12"x8".
The port should be closed off, left open it will add boom.
Assuming 3/4" stock internal volume pans out at 0.85 cuft.
Well stuffed might get this nearer 1 cuft effective.
This gives a box Q of 0.9, not too bad but an ~ 2 cuft box is ideal.

:)/sreten.
 
interim

Right now I just need a set of speakers that sound okay. Not great, just okay. I eventually want to upgrade the speakers, but I don't have the kind of cash required to buy HiFi speakers right now. If I can get these as "passable" for under a hundred bucks, that will suit me for a few months until I can sell a motorcycle and use the profits for a decent pair of speakers. (my other hobby is fixing up old bikes).

But then I can hopefully use these in the den if I can get them to sound passably good.

Just talked to my friend btw. He didn't use the dayton crossover like I thought. He built the crossovers himself. He said the woofer is crossed at 1kHz to try and reduce that spike. Tweeter's crossover is at 2kHz, and is identical to the recommended crossover in the tweeter's spec sheet.

Charles.
 
Oh, thanks for the compliment on the chopper fwater. That's not the only bike I've done... you should check out the photo gallery for the ninja and the dirt bike.

I've attached a shot of the speaker with the Dayton parts installed.

Charles.
 

Attachments

  • randallshack.jpg
    randallshack.jpg
    57.2 KB · Views: 233
Nevermind...

Since my options were either to put in a fullrange driver (and not take into consideration the box characteristics at all), or actually build new cabinets.... I decided to do neither. I found a like new pair of Cerwin Vega VS-150 speakers for $200 off of Craigslist. They're a little large, but I'm a very happy man right now. In retrospect I think I just outgrew these modified Radio Shacks.... and I'm okay with that.

Charles.
 
Well, this brings up a question I asked a few days ago about how to make a reasonable judgment about the necessary overlap in frequency response between woofer and tweeter in a two way system.

In my view, you need a woofer with a much higher frequency response, so that the crossover point is well within it's workable range. Based on my not-so-expert opinion, this woofer needs to be crossed over at 1khz or less to keep that high end peak out of the circuit, and the tweeter needs to be crossed over at 2.5khz to maybe 3khz. So, obviously there is a big gap in the middle.

So, if you want to salvage or sell these speakers that gap need to be filled. I think there is enough room on the front of the cabinets to add a small midrange. Then add an off-the-shelf 3-way crossover. Just make sure the crossover you choose is within the working range of the midrange.

The alternative, if you hope to salvage them at all, is to replace the woofer with a woofer that has frequency response up to say 4khz or so. Though I doubt you will find many that go higher than that. That way, the 2khz crossover will be well within its working range.

Then use the existing Daytons in a project that allows either a much lower or much faster crossover.

Also, if you decide to use the Tang Band speakers with these woofers, you would need to crossover at about 800hz which is well within the range of the Tang Bands and far from the high end peak found in the Dayton woofers.

Again, I'm not an expert but I'm trying to learn, and the above options are the only ways I can see to fix the existing system.

The only other mods I could think of for the existing system might be a shallow notch filter centered on the woofer's high end peak. That would soften and flatten the woofer frequency response.

So, tell me gang, am I right or am I wrong; I'm trying to learn here.

Steve/bluewizard
 
BlueWizard said:


So, tell me gang, am I right or am I wrong; I'm trying to learn here.

Steve/bluewizard

Hi, OK see below, :)/sreten.

BlueWizard said:
Well, this brings up a question I asked a few days ago about how to make a reasonable judgment about the necessary overlap in frequency response between woofer and tweeter in a two way system.

In my view, you need a woofer with a much higher frequency response, so that the crossover point is well within it's workable range. Based on my not-so-expert opinion, this woofer needs to be crossed over at 1khz or less to keep that high end peak out of the circuit, and the tweeter needs to be crossed over at 2.5khz to maybe 3khz. So, obviously there is a big gap in the middle.

The croosover will have no effect on the distortion levels caused
by the drivers excessive response peak and mundane motor levels
of distortion. It ideally needs crossing low, the lower the better.


So, if you want to salvage or sell these speakers that gap need to be filled. I think there is enough room on the front of the cabinets to add a small midrange. Then add an off-the-shelf 3-way crossover. Just make sure the crossover you choose is within the working range of the midrange.

Adding generic drivers and off the shelf croosovers is a very bad idea.
See the Lyra link for the effect of an off the shelf crossover.


The alternative, if you hope to salvage them at all, is to replace the woofer with a woofer that has frequency response up to say 4khz or so. Though I doubt you will find many that go higher than that. That way, the 2khz crossover will be well within its working range.
Then use the existing Daytons in a project that allows either a much lower or much faster crossover.

The best option IMO is the Lyra.
New front baffle to seal the box and allow the mid to be added.


Also, if you decide to use the Tang Band speakers with these woofers, you would need to crossover at about 800hz which is well within the range of the Tang Bands and far from the high end peak found in the Dayton woofers.

Probably c/o lower to get the baffle step correct and additional
top end filtering probably required. The whole thing would probably
end up bass heavy as the TB is much less sensitive than specified.


The only other mods I could think of for the existing system might be a shallow notch filter centered on the woofer's high end peak. That would soften and flatten the woofer frequency response.

An elliptic crossover (includes a notch) is a possibility but as
stated I think the distortion performance precludes this idea.


Steve/bluewizard
 
Sreten,

Not wanting to doubt your knowledge or ability as they have certainly proven themselves, but...

You say-

The crossover will have no effect on the distortion levels...

then follow that with -

It ideally needs crossing low, the lower the better.

which implies that indeed the crossover does matter and does have an effect. The crossover point has to be well below 1khz, which is what I said.

The crossover will have an effect because if you crossover low enough and/or sharp enough, the distortion/break-up area of the woofer will be substantially attenuated when we reach it.

I then said -

"So, if you want to salvage or sell these speakers that gap need to be filled."

Your suggestion of the Lyra was very close to what I envisioned, a compact midrange dome speaker in a 3-way system. So, on this option we agree.

But I suggested adding off-the-shelf 12db 3-way crossovers, an idea that you shot down. Yet if you look at the Lyra crossovers they are just basic 3-way 12db off the shelf crossovers with a few tweaks added.

Which brings up a general comment that I think needs to be made. Not every question requires a 'state of the art' answer. Every question has a context too it. The level of skill, the available resources (both financial and academic), available tools, and ambition and inclination of the person involved all need to be considered.

You can't offer a $10,000 solution to a $100 question. And again, this isn't directed at Sreten specifically, this is a rant that has been brewing for some time. Sreten has more than proven his skill and knowledge over the years.

So, the context of the question puts limits on the answer. I don't sense that ChopperCharles is trying to build the ultimate speaker system. He is simply trying to take an existing cabinet and re-make it into a decent functional set of speakers. I don't get he sense that he is in a position to invest a big chunk of money or time.

Consequently, I tried to suggest the shortest, easiest, and cheapest route to taking what he has and ending up with decent functional speakers. An off-the-shelf crossover is a quick and easy way to accomplish that.

None the less, there are only three ways that I can see to salvage what he has.

In short, raise the bottom, lower the top, or fill the gap.

The first is to add a midrange, a carefully selected midrange in his price range. Which in turn means a new crossover too.

Your suggestion of the Lyra speaker system is excellent. The woofer and tweeter are virtually the same as what he has, and the midrange is a good match to both. When it comes to crossovers, I say it depends on how much time, money, and inclination ChopperCharles has. Yes, off the shelf crossovers are a second best shortcut, quick and easy solution, and if that is what you are looking for, then it is the perfect solution.

If ChopperCharles has the time, resources, and inclination to make the Lyra crossovers, then more power to him, and it will certainly yield better results. But that doesn't mean ready made crossovers will yield terrible results. Less that ideal yes, but everything we do is a compromise. I think decent results can be achieve with ready made crossovers, though at the same time concede better result can be obtained with custom crossovers. But custom crossovers require a lot of time and effort as well as an available selection of components, and the skill and resources to test and verify them; not everyone has that.

It's all down to context, and only ChopperCharles can determine the best solution within his context.

The second, is to replace the woofer with a woofer that has a much higher frequency response and a much more stable high end roll off. Which means he might be able to keep his existing crossovers.

The third, if you can't bring the woofer up, then you need to bring the tweeter down. The Tang Band is not ideal, but I contend that it is a workable solution. The rated sensitivity of the woofer is 89db; the rated sensitivity of the selected Tang Band is 87db. That seems a reasonable match to me, though I concede it is not as simple as Rated sensitivity.

It is the nature of high pitched sounds to sound louder at equal volume when compared to low pitched sounds. That works in CC's favor. This wouldn't be an ideal match, but it would make a functional decent speaker system that would be a little on the warm side due to the slightly lower sensitivity of what would now be the mid/high speaker.

In this case, as I said, the crossover would have to be replaced, but the Tang Band has an amazingly flat response with virtually no aberrations. So, I say to create a decent basic workable speaker, an off the shelf crossover would be adequate in this case. I suggested an 800hz crossover, you suggested lower; either is workable in context. Though I concede finding an 800hz 2-way crossover might be difficult.

Again, ideally, a custom-made custom-tweaked crossover is always the best. But, in context, not everyone is striving for the best. Some people are willing to settle for good.

Still, I would say the most ideal solution, is to work with some variation of the Lyra system.

For what it's worth.

Steve/bluewizard
 
Hi,

I replied to your comments to help. No need to be so defensive.
It is not always obvious what some one is getting at or when
your trying to add to what someone has already said.

But the following points apply :

a) the 10" croosover cannot be only considered in terms of
frequency response, even it this looks ok distortion is a
midrange issue, which was not explained.

b) The Lyra c/o is far from an "off the shelf 3-way", is has midrange
and treble attentuation for a start, and I think you'll find none of
its values match "8 ohm off the shelf" values for the c/o points.

c) We both agree some variant of the Lyra is probably best. We do
not agree about "off the shelf" c/o's. IMO for the Lyra not using the
designed c/o is just completely wasting what is possible.

d) Seems fairly academic now, 15" bass units have won the day ......


[IMGHTTPDEAD]http://www.audioreview.com/channels/audioreview/images/products/product_120900.jpg[/IMGHTTPDEAD]

I wonder what the c/o in these is like .......

:)/sreten.
 
Sorry, Sreten, I was afraid even as I was writing that I might come off a little...well, let's just say 'snarky' and let it go at that.

Keep in mind that I do agree with you, for the most part. In recommending the Lyra design and crossover, you are trying to raise the context and use the available parts to the best possible end, and still stay within reasonably achievable results. Your case for raising the bar within reasonable limits, is a case that definitely needed to be made.

But I'm trying to offer solutions within the existing contexts, though I concede that my interpretation of context is pure speculation. CC has already found a solution, but he still has these parts and still has money invested in these parts. Now the question is, whether today or next year, how can he make use of them in the least complicated fashion, and in a way the produces the best result in the least complicated fashion.

Again, I'm guessing.

As to your point a.) I'm interpreting that to mean that it is just a generally bad idea to try and make a 10" two-way speaker. It is very difficult to get a 10" to effectively cover the bass and mid without complications. 6" and 8" might do nicely, but 10" is a real stretch. Yet, there are 10" woofers that go up to 4000hz to 6000hz. Are they ideal...no. Will they work..yes, within limits.

All I am saying, is that it can be done with acceptably good results, not ideal, not preferred, just acceptable. And the only means to do that are, as I said, to raise the bottom, lower the top, or fill the gap.

The best possible results, and not that hard a process, are to fill the gap,;in other words, add a midrange; hence your reference to the very good Lyra system. But I am trying to offer CC a few 'good enough' options. The more options he has, the more he will be able to resolve the problem in his own working context.

In other words, I'm not saying you are wrong, in fact, I am saying you are right. But, if that option is out of reach for one reason or another, the are more heavily compromised but still acceptable options available.

Between the two of us, I think we have covered all the possibilities from 'get by' to nearly 'ideal'. Somewhere in there is a solution that CC can use when the time comes.

My little rant, as I tried to emphasize, wasn't necessarily directed at you. But frequently, I see responses that are the 'idealized' solutions, when it is clear the person asking the question is not able, for a long list of possible reasons, to implement the idealized solution.

In many cases, the idealized is offered in the hope of getting the person to raise the bar on their skills and equipment, but that raising of the bar takes time, resource, money, and a lot of hard work. Yet, the person has a problem that needs a 'now' solution. Raising the bar, will come with experience and time, but that doesn't fix the immediate problem.

Again, I think you offered the preferred solution, and a solution that would produce excellent results, if CC is in a position to dedicate the time and resources into doing it. If not, then I offered lower, yet I feel acceptable, solutions, but I readily admit they are indeed 'lower' solutions.

Sorry, if my comment produce any uncomfortable feelings, that wasn't my intent. I just want to make sure the full spectrum of potential solutions were covered.

Hoping I haven't made it worse.

Steve/bluewizard