Opinions On This Easy DIY Design

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
What do you experts think of this design using off-the-shelf PE components. Any suggestions for improvement would be greatly appreciated.

OK, here it is:

Peerless 830883 6-1/2" Nomex Cone HDS Woofer
Peerless 830883 6-1/2" Nomex Cone HDS Woofer 264-1094

Vifa XT25TG30-04 1" Dual Ring Radiator Tweeter
Vifa XT25TG30-04 1" Dual Ring Radiator Tweeter 264-1016

Dayton Audio XO2W-2.5K 2-Way Speaker Crossover 2,500 Hz(off the shelf)
Dayton Audio XO2W-2.5K 2-Way Crossover 2,500 Hz 260-142

An DIY 3db L-Pad attenuator for the tweeter using a 1.2ohm & 10ohm resistor

PE 0.5cf ported box.
 

Attachments

  • circuit.jpg
    circuit.jpg
    8.1 KB · Views: 364
Any of the shelf crossover will not work apart from keeping drivers burning up. If you have zobel on the bass driver it will work better, than without that zobel but you really have to have some way of measuring what you are doing. I use omnimic but Parts Express USB mic and the freeware Holmimpulse is one other option and there are many others...
 
Woody is correct. I also would second the notion that you need specially designed crossovers. In speaker building, the order goes as thus:

1. Goal selection
1. Driver selection (whichever can come first depending on the case)
2. Enclosure design
3. Network design

Look at response curves of your drivers if published and note various features. For example, a woofer might be listed as having a frequency response of up to 2kHz. Is this due to acoustic rolloff, or because there is a breakup mode somewhere in the tweeter region which would necessitate extra filtering? You can learn these things by looking at charts, and thankfully, Tymphany (Vifa and Peerless are both part of the company Tymphany) like Scan-Speak and Seas is very good about posting accurate response curves of their drivers.
 
There are certainly experts here that make me look like a concussed Chimp, but I think you can get a reasonable first cut speakers system using off the shelf crossover, but you have to choose them wisely.

In this case, if you look at the Specification sheet of both the Tweeter and the Woofer, it would seem that 2.5Khz is a less than ideal crossover. I think about 1.5khz would be better.

The Zobel Network mentioned is actually a very simple thing, a resistor and capacitor in parallel with the woofer to stabilize the woofer impedance at higher frequencies.

http://www.parts-express.com/pdf/264-1016s.pdf

Best guess the tweeter is functional above 1khz.

http://www.parts-express.com/pdf/264-1094s.pdf

Best guess pushing the woofer to 1.5khz is indeed pushing it.

The woofer above 1.5khz as a break up. An area above the working range where the output peaks. For this woofer that is about 4khz. That peak can also be deal with with a simple notch filter. Easy enough to figure out.

So, in my highly concussed chimp opinion, an off the shelf crossover at 1.5khz with a simple Zobel on the woofer, and a notch filter around 4khz, should yield pretty decent first try results.

As to the L-Pad, get a bunch of values of resistor. What is right in number, is not always right in practice. Treble can be piercing, it tends to cut through and seem louder even when the numbers do not bear it out.

That's why women and babies have high pitched scream, so they can be heard, and that is why men have deeper rumbling voice so the Saber-Tooth Tiger doesn't hear them.

Now, to take it one step farther, for a more precise crossover, rather than using the generic 8 ohm, you find the precise impedance of each speaker at the crossover. That easy enough, signal generator (CD with a 1.5khz tone on it), a small value resistor, and a couple of volt meters. Done.

Based on the specification charts, the Tweeter at 1.5khz is about 3.5 ohms. The woofer is about 8 ohms, if we can trust the spec charts that works out nicely.

However, I never quite trusted the Dayton Crossovers that claim to accommodate 4 ohms and/or 8 ohms, that just didn't seem logical based on the components I see on the crossovers.

Far better to calculate your own using the values of your drivers. At my suggested crossover frequency, they are close enough to 4 ohms and 8 ohms, that you can start with a generic crossover calculator -

Crossover Design Chart and Inductance vs. Frequency Calculator(Low-pass)

You have all types listed at the link above, as well as a Zobel calculator and an L-Pad calculator.

I'm sure someone here can help you with the notch filter.

Making a crossover is not that complicated. You can see photo examples on this forum in great numbers.

Now, once you have this basic crossover with the additional circuits added. Things get really complicated. Beyond this basic point, crossover design gets very complex, and for a first try speakers design, probably beyond your reach.

Though with just the basic design, I think you can do a very good job, and get very good results.

The peak in the woofer response, it a bit more than a octave above my suggested crossover. How much of a problem that peak will be will be for other with more experience to determine. But as I said, a simply notch filter is not that complex. I'm sure someone here could set you up with one.

Just one man's opinion.

Steve/bluewizard
 
diyAudio Moderator
Joined 2008
Paid Member
In speaker building, the order goes as thus:

1. Goal selection
1. Driver selection (whichever can come first depending on the case)
2. Enclosure design
3. Network design
This is often the case but I don't subscribe to this myself.
1. Goal selection
2. Enclosure specification based on goals
3. Driver selection based on enclosure specification
4. Enclosure specifics
5. Network design

This is partly because the desire to use a specific driver will begin to dictate the goals. Using an appropriate driver in an appropriate enclosure can also sound better than using a special driver in a plain, or random enclosure.

Why should there be a zobel on the bass?
This is OK when we understand what we're doing, but ironically few of us would use one because we do ;)

From a generic point of view, the way to get one of these crossovers to work without modification is to compensate the impedances and the levels, and align the drivers and use them where they are most linear. Again, this borders on building the crossover from scratch.
 
This is often the case but I don't subscribe to this myself.
1. Goal selection
2. Enclosure specification based on goals
3. Driver selection based on enclosure specification
4. Enclosure specifics
5. Network design

This is partly because the desire to use a specific driver will begin to dictate the goals. Using an appropriate driver in an appropriate enclosure can also sound better than using a special driver in a plain, or random enclosure.
Indeed! This is a better description than my own. :)
 
Don't waste your time or money with that pre-fab xo. Your chances of it sounding good are about as likely as hitting the bulls-eye while shooting blindfolded.

Driver selection is good and I think they could sound quite nice together if the xo is designed properly. The XT25 needs an impedance compensation circuit and a minimum xo of about 2500Hz. Or a waveguide. (see Zaph|Audio - ZD5 - Scan Speak 15W8530K00 and Vifa XT25 for more info on this) The Peerless ideally wants to be crossed lower than that but could probably manage it the way its brother does in the Veeper TM Monitor.

But unless your goal is to also learn how to design speakers as well as build them, you are much better off building an existing design.

The Veepers would be a good alternative with another Peerless HDS Nomex driver and a different Vifa tweeter. There are lots of other designs to choose from as well.

In this price range tho, I would second the choice of the Piccolo. Those are very high value drivers put together by an expert designer.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.